Maintaining the silence: reflections on long-term RNAi

Ghent University Faculty of Pharmaceutical Sciences

Biodegradable dextran nanogels as functional carriers for the intracellular delivery of small interfering RNA

Biodegradeerbare dextraan nanogelen als functionele dragers voor de intracellulaire afgifte van small interfering RNA

Koen Raemdonck Pharmacist

Thesis submitted to obtain the degree of Doctor in Pharmaceutical Sciences

Proefschrift voorgedragen tot het bekomen van de graad van Doctor in de Farmaceutische Wetenschappen

2009

Dean

Prof. dr. apr. Jean Paul Remon

Promoters

Prof. dr. apr. Stefaan C. De Smedt Prof. dr. apr. Jo Demeester

Laboratory of General Biochemistry and

Physical Pharmacy

The author and the promoters give the authorization to consult and to copy parts of this

thesis for personal use only. Any other use is limited by the Laws of Copyright, especially the

obligation to refer to the source whenever results from this thesis are cited.

De auteur en de promotoren geven de toelating dit proefschrift voor consultering

beschikbaar te stellen en delen ervan te kopiëren voor persoonlijk gebruik. Elk ander gebruik

valt onder de beperkingen van het auteursrecht, in het bijzonder met betrekking tot de

verplichting uitdrukkelijk de bron te vermelden bij het aanhalen van resultaten uit dit

proefschrift.

Ghent, October 2nd 2009

Promoters: Author:

Prof. dr. apr. Stefaan C. De Smedt Koen Raemdonck

Prof. dr. apr. Jo Demeester

table of contents

TABLE OF CONTENTS

Table of contents 9

List of abbreviations & symbols 11

General introduction & thesis outline 15

CHAPTER 1 | Maintaining the silence: reflections on long‐term RNAi 21 CHAPTER 2 | In situ analysis of single‐stranded and duplex siRNA integrity in living cells 59

CHAPTER 3 | Dextran microgels for time‐controlled delivery of siRNA

85

CHAPTER 4 | Advanced nanogel engineering for drug delivery 109

CHAPTER 5 | Biodegradable dextran nanogels for RNAi: focusing on endosomal escape

and intracellular siRNA delivery 133 CHAPTER 6 | Evaluation of gene silencing kinetics with siRNA loaded dextran nanogels 163 CHAPTER 7 | Liposomal template‐assisted synthesis of dextran based nanogels for

siRNA delivery 183

CHAPTER 8 | Summary & general conclusions 205

APPENDIX A | Samenvatting & algemene besluiten 215

APPENDIX B | Curriculum vitae 223

DANKWOORD | Acknowledgment 233

9

table of contents

10

list of abbreviations & symbols

LIST OF ABBREVIATIONS

AAV adeno‐associated virus

AF488 Alexa Fluor 488

AFM atomic force microscopy

Ago argonaute

agRNA antigene RNA

AMD age‐related macular degeneration

AON antisense oligonucleotide

ATRP atomic transfer radical polymerization

bp basepair

BSA bovine serum albumin

CaF carboxyfluorescein

CDP cyclodextrin polycation

CLSM confocal laser scanning microscopy

coRNAi combinatorial RNA interference

CPP cell penetrating peptide

CRP controlled radical polymerization

DEAEMA 2‐(diethylamino)ethyl methacrylate

DEPC diethyl pyrocarbonate

Dex‐HEMA dextran hydroxyethyl methacrylate

Dex‐MA dextran methacrylate

DGCR8 DiGeorge syndrome chromosomal region 8

DIC differential interference contrast

diINF‐7 dimer of influenza‐derived fusogenic peptide‐7

DLS dynamic light scattering

DMAEMA 2‐(dimethylamino)ethyl methacrylate

DMEM Dulbecco’s modified Eagle’s medium

DNA deoxyribonucleic acid

DNase deoxyribonuclease

DOPE 1,2‐dioleoyl‐sn‐glycero‐3‐phosphoethanolamine

DOPC 1,2‐dioleoyl‐sn‐glycero‐3‐phosphocholine

DS degree of substitution

DsiRNA dicer substrate RNA

dsRNA double stranded RNA

DTT dithiothreitol

ECM extracellular matrix

EDTA ethylenediaminetetraacetic acid

11


EE encapsulation efficiency

EGFP enhanced green fluorescent protein

eri‐1 enhanced RNAi‐1

FANA 2’‐fluoro‐β‐D‐arabino nucleic acid

FA/FD ratio of acceptor to donor fluorescence upon donor excitation

FBS fetal bovine serum

FFS fluorescence fluctuation spectroscopy

FRET fluorescence resonance energy transfer

GPC gel permeation chromatography

HBV hepatitis B virus

HDL high density lipoprotein

HEPES 4‐(2‐hydroxyethyl)‐1‐piperazine ethanesulfonic acid

HIV‐1 human immunodeficiency virus‐1

HPH high pressure homogenization

Huh‐7 human hepatoma‐7

kDa kilodalton

KPS potassium peroxodisulfate

LCST lower critical solution temperature

LDL low density lipoprotein

LEN lipid enveloped nanogel

LF lipofectamine

LMV large multilamellar vesicle

LNA locked nucleic acid

LPX lipoplex

MEND multifunctional envelope‐type nano device

miRNA microRNA

MOE 2‐methoxyethyl

mRNA messenger RNA

MW molecular weight

MWCO molecular weight cut‐off

nt nucleotide

NPC nuclear pore complex

OD optical density

OTE off‐target effect

PAGE polyacrylamide gel electrophoresis

PbAE poly(β‐amino ester)

PBS phosphate buffered saline

PCI photochemical internalization

12


PDI polydispersity index

pDNA plasmid DNA

PEG poly(ethylene glycol)

PEI poly(ethylene imine)

PNIPAAm poly(N‐isopropylacrylamide)

PRINT particle replication in nonwetting templates

PS photosensitizer

P/S penicillin/streptomycin

PTGS post‐transcriptional gene silencing

RdDM RNA dependent DNA methylation

RdRP RNA dependent RNA polymerase

RhoGr rhodamine green

RISC RNA induced silencing complex

RLU relative light units

RNA ribonucleic acid

RNAi RNA interference

RNase ribonuclease

RT room temperature

SD standard deviation

SDS sodium dodecylsulfate

SEM scanning electron microscopy

shRNA short hairpin RNA

siRNA small interfering RNA

SNALP stable nucleic acid lipid particle

ssRNA single stranded RNA

SUV small unilamellar vesicle

TBE tris, boric acid and EDTA buffer

TEMED N,N,N’,N’‐tetramethylenediamine

TGS transcriptional gene silencing

TLR toll‐like receptor

TMAEMA [2‐(methacryloyloxy)ethyl] trimethylammonium chloride

TPPS2a meso‐tetraphenylporphine disulfonate

TRBP TAR RNA binding protein

TX100 Triton®X‐100

UV ultraviolet

VPG vesicular phospholipid gel

W/O water‐in‐oil emulsion

w/v weight/volume

13


14

LIST OF SYMBOLS

ε extinction coefficient

λex excitation wavelength

λem emission wavelength

ζ zeta potential

general introduction & thesis outline

GENERAL INTRODUCTION &

THESIS OUTLINE

15


16


GENERAL INTRODUCTION

RNA interference and siRNAs

Just slightly over a decade ago, Andrew Fire and Craig Mello described in their Nature publication the

peculiar observation that the exogenous introduction of double‐stranded RNA (dsRNA) into the

nematode C. elegans resulted in the suppression of the complementary gene rather than enhancing

its expression.[1] These authors were the first to launch the term RNA interference (RNAi), a discovery

that was awarded in 2006 with the Nobel Prize for Medicine and Physiology.

At present, RNAi is defined essentially as a naturally occurring and evolutionary conserved

mechanism by which short dsRNA triggers the sequence‐specific inhibition of gene expression. The

knowledge that this gene silencing mechanism is operational in mammalian (and therefore also

human) cells, revolutionized our understanding of endogenous gene regulation by small RNAs and

sparked an explosion of research to harness RNAi as a novel therapeutic modality.[2] RNAi can be

implemented for the treatment of a variety of human diseases with a known genetic origin. Now

with the sequencing of the entire human genome, RNAi may lead to a plethora of drug molecules

against any chosen clinical target.[3] The importance of the discovery of RNAi is further emphasized

by the fast transition from lab to clinic with several clinical trials already in progress and several more

that are scheduled for the coming years.

Intensive investigation on the RNAi mechanism identified small interfering RNAs (siRNAs) as the

genuine effectors of the genetic interference.[4] Synthetic siRNAs can easily be produced through

large‐scale chemical synthesis and the majority of reports on RNAi therapeutics have focused on

siRNA for the activation of gene silencing. However, to trigger RNAi, siRNAs need to be delivered into

the cytoplasm of the target cell and many extra‐ and intracellular barriers are encountered before

the site of action is reached. It is therefore imperative for the widespread development of siRNA

therapeutics to focus on the design of novel nucleic acid delivery systems to overcome these.[5] As

virus‐mediated nucleic acid delivery is still associated with insuperable safety concerns, the use of

non‐viral delivery strategies has gained a lot of interest. The majority of reports on siRNA delivery

describe the application of lipid or polymer based carriers to enhance the fraction of the

administered siRNA dose that effectively reaches the target cell cytoplasm.[6]

17


Hydrogels for siRNA delivery

Hydrogels can be defined as three‐dimensional networks of hydrophilic polymers that are capable of

absorbing large quantities of water or biological fluid. Both synthetic and natural polymers can be

used as core material for their construction. Since their introduction for biological use almost 50

years ago,[7] at present hydrogels are favoured for many biomedical and pharmaceutical purposes,

including controlled drug delivery.[8‐10] Hydrogels may exist in many geometries such as macroscopic

gels (e.g. scaffolds, cylinders,...), microscopic gels and nanogels. Up to now, literature has mainly

focused on macro‐ and microscopic hydrogels for extracellular drug delivery applications but during

the last decade nanogels have gained momentum as drug delivery vehicles. Since siRNAs are needed

in the cell cytoplasm to be effective, as a consequence hydrogels with nanoscopic dimensions are

required to allow intracellular siRNA release.

In essence, due to the relatively recent discovery of siRNA therapeutics, there is a lack of scientific

studies describing nanogels as siRNA carriers for their controlled delivery across the cellular barriers,

which is therefore the major topic of investigation in this thesis.

THESIS OUTLINE

CHAPTER 1 provides insight into the mechanism of RNA interference and discusses the predominant

barriers for siRNA delivery. Moreover, in this chapter an overview can be found of the most

important parameters that govern the silencing efficiency and influence the duration of the gene

silencing effect.

In CHAPTER 2 we aimed to develop a method, based on fluorescence resonance energy transfer (FRET)

to assess the intracellular siRNA stability. Both double stranded (ds) and single stranded (ss) siRNA

are included in this investigation. In addition to chapter 1, we discuss the significance of intracellular

siRNA stability on the RNAi outcome.

In the context of controlled siRNA delivery, CHAPTER 3 investigates if cationic and biodegradable

dextran microgels are suited for siRNA complexation and subsequent degradation controlled release.

This chapter also provides a first indication that these cationic gel particles are able to deliver active

siRNA across the cellular barriers.

Before we investigate intracellular siRNA delivery with our dextran hydrogel particles in more detail,

CHAPTER 4 reviews the state‐of‐the‐art of nanogel design for drug delivery applications.

18


Through mini‐emulsion photopolymerization, in CHAPTER 5 we succeeded in the production of cationic

dextran nanogels, based on the same core material as was used in chapter 3. Their siRNA

complexation potential, intracellular siRNA delivery and eventual gene silencing activity are

evaluated. Additionally, the influence of endosomolytic tools on the RNAi effect is investigated.

Besides the maximal gene suppression, it is equally important to assess the duration of the gene

silencing effect when evaluating a novel siRNA delivery strategy. For this reason, CHAPTER 6 deals with

the kinetics of the RNAi effect obtained following transfection with cationic nanogels of varying

crosslink density.

Finally, in CHAPTER 7, a second approach for the preparation of dextran nanogels is demonstrated.

Here, we apply lipid vesicles as nanoreactors for the selective crosslinking of functionalized dextran

in their aqueous interior. Thus, nanogels are obtained that are surrounded by a lipid shell. This

chapter contains pilot experiments investigating the use of these lipid‐enveloped nanogels (LENs) as

carriers for intracellular siRNA delivery and discusses potential advantages this hybrid nanogel

delivery system may offer.

REFERENCE LIST

1. Fire A., Xu S., Montgomery M.K., Kostas S.A. et al., Potent and specific genetic interference by double‐stranded RNA in Caenorhabditis elegans, Nature 1998, 391: 806‐811.

2. Elbashir S.M., Harborth J., Lendeckel W., Yalcin A. et al., Duplexes of 21‐nucleotide RNAs mediate RNA interference in cultured mammalian cells, Nature 2001, 411: 494‐498.

3. de Fougerolles A.R., Delivery vehicles for small interfering RNA in vivo, Hum. Gene Ther. 2008, 19: 125‐132. 4. Elbashir S.M., Lendeckel W., and Tuschl T., RNA interference is mediated by 21‐ and 22‐nucleotide RNAs,

Genes Dev. 2001, 15: 188‐200. 5. Whitehead K.A., Langer R., and Anderson D.G., Knocking down barriers: advances in siRNA delivery, Nat.

Rev. Drug Discov. 2009, 8: 129‐138. 6. Gao K. and Huang L., Nonviral Methods for siRNA Delivery, Mol. Pharm. 2009, 6: 651‐658. 7. Wichterle O. and Lim D., Hydrophilic gels for biological use, Nature 1960, 185: 117‐118. 8. Gupta P., Vermani K., and Garg S., Hydrogels: from controlled release to pH‐responsive drug delivery, Drug

Discovery Today 2002, 7: 569‐579. 9. Peppas N.A., Bures P., Leobandung W., and Ichikawa H., Hydrogels in pharmaceutical formulations,

European Journal of Pharmaceutics and Biopharmaceutics 2000, 50: 27‐46. 10. Peppas N.A., Hilt J.Z., Khademhosseini A., and Langer R., Hydrogels in biology and medicine: From

molecular principles to bionanotechnology, Advanced Materials 2006, 18: 1345‐1360.

19


20

maintaining the silence: reflections on long‐term RNAi | CHAPTER 1

CHAPTER 1 | Maintaining the silence: reflections on long‐

term RNAi

Parts of this chapter are published:

Koen Raemdonck1, Roosmarijn E. Vandenbroucke1, Niek N. Sanders1,2, Joseph Demeester1 and

Stefaan C. De Smedt1, Drug Discovery Today 2008, 13: 917‐931.

1Laboratory of General Biochemistry and Physical Pharmacy, Department of Pharmaceutics, Faculty of Pharmaceutical

Sciences, Ghent University, Harelbekestraat 72, B‐9000 Ghent, Belgium. 2Laboratory of Gene Therapy, Department of Nutrition, Genetics and Ethology, Faculty of Veterinary Medicine, Ghent

University, Heidestraat 19, B‐9820 Merelbeke, Belgium.

21

CHAPTER 1 | maintaining the silence: reflections on long‐term RNAi

22

ABSTRACT

Since the demonstration of RNA interference (RNAi) in mammalian cells, considerable research and

financial effort has gone towards implementing RNAi as a viable therapeutic platform. RNAi is

without doubt the most promising strategy for the treatment of human genetic disorders. As many

of the targets proposed for RNAi therapy require chronic treatment, researchers agree that the

emphasis now must be placed on the safe and long‐term application of RNAi drugs to finally reap the

benefits.


1 OVERVIEW CHAPTER CONTENTS

Chapter 1 provides:

• a brief introduction to the post‐transcriptional gene silencing mechanism termed RNA

interference (RNAi)

• a summary of the most important parameters that govern the silencing efficiency and

duration of the RNAi effect

• a discussion of the predominant barriers in the context of siRNA delivery

• an introduction to RNAi mediated transcriptional gene silencing, mathematical modelling in

RNAi therapy and combinatorial RNAi

Key terms:

siRNA – miRNA – shRNA – mathematical modelling – transcriptional gene silencing – coRNAi – siRNA

delivery – chemical modification

INTRODUCTION

RNA interference (RNAi) represents a powerful and versatile gene silencing process in which double‐

stranded RNA (dsRNA) triggers the sequence‐specific cleavage of mRNA transcripts. An explosion of

research that followed its discovery in Caenorhabditis elegans in 1998,[1] has recently resulted in both

RNAi based protocols for analysis of gene function and therapeutic applications in humans.[2,3]

Following pioneering research in plants [4,5] and nematodes,[1,6] RNAi was demonstrated in

mammalian cells in 2001 by Tuschl and colleagues, who were the first to apply small interfering RNAs

(siRNAs) to guide the sequence‐specific suppression of gene expression.[7,8]

23


In theory every gene is amenable to RNAi based silencing, and therefore these key papers have led to

a surge of excitement amongst researchers in the medical field. For about a decade now, the quest

for efficient RNAi therapeutics to treat a wide variety of pathologies has been ongoing and has

already made remarkable progress, but we must emphasize that further improvements are still

needed to pave the way for RNAi to develop into a viable therapeutic approach. Obviously, the

translation of RNAi into a broadly applicable therapeutic platform needs appropriate pharmaceutical

considerations at different levels in clinical drug development like e.g. siRNA design [9,10] and siRNA

formulation.[11] In straightforward terms it can be stated that the most efficient delivery of the most

potent siRNA is expected to result in maximal suppression of a target gene.

To extrapolate RNAi from bench to bedside, it is imperative to consider not only the gene knockdown

intensity, but also the dynamics of gene silencing.[12] On one hand, precise control over the duration

of RNAi gene knockdown can provide new information on complex cellular pathways implying

interactions between multiple genes, without having to rely on more elaborate transgene

technologies. On the other hand, from a therapeutic point of view, it may be relevant to silence a

pathogenic gene for a longer period of time, to improve the clinical outcome. Prolongation of the

gene silencing effect of a specific siRNA formulation may be beneficial for chronic patient adherence

if the administration frequency in the dosing schedule could be significantly scaled down.

Although many in vivo studies have already highlighted the enormous therapeutic potential of RNAi

therapeutic strategies,[13] recent studies have shown unintended off target effects (OTEs), activation

of immune and inflammatory pathways and perturbation of endogenous cellular pathways by small

dsRNA drugs. Especially when placing the emphasis on long‐term treatment of chronic diseases,

toxicity issues become increasingly important and should be dealt with accordingly. However, at

present, few data are available on the long‐term toxicity of RNAi triggers but appropriate attention

should be paid to these issues in the context of prolonging RNAi silencing.

In this chapter we will primarily focus on important remarks and strategies in the light of prolonged

RNAi gene silencing for therapeutic purposes. In the different sections we attempt to describe

several topics from the available literature on RNAi that are of particular relevance to the central

theme of this thesis.

INTRINSIC RNAi GENE SILENCING EFFICIENCY AND LONGEVITY

Since the discovery of RNAi, researchers have revealed the basic steps of the RNAi pathway, but

many critical aspects remain to be unveiled. The RNAi effector molecules, termed small interfering

24


RNAs (siRNAs), are RNA duplexes ~21‐23 nucleotides (nt) in length consisting of a duplex stem ~19 nt

and both strands bearing a 5’‐phosphate and 2 nt overhangs at the 3’‐ends (Figure 1.1). These small

RNA fragments are defined as the products of longer dsRNA processing by an RNase III type enzyme,

called Dicer. In primitive organisms, long dsRNA can effectively trigger RNAi. However, in higher

organisms, these long dsRNAs (> 30 nt) also trigger innate immune responses resulting in severe

toxicity. Therefore, by preference synthetic small siRNAs are usually applied to experimentally

activate RNAi.

Once introduced in the cell’s cytoplasm, the heterodimer complex Dicer/TRBP processes dsRNAs into

siRNAs that are subsequently incorporated into a protein complex termed RISC (RNA induced

silencing complex). RISC also contains the Argonaute 2 (Ago‐2) protein.[14] Only one strand of the RNA

duplex is retained inside the RISC, which is the antisense strand or ‘guide’ strand. The sense strand or

‘passenger’ strand is nicked by Ago‐2, unwound and discarded.[15] Subsequently, the activated RISC

(RISC*) uses the guide strand to bind to the complementary region on the target mRNA. Ago‐2

contains an RNase H like domain that is responsible for directed transcript cleavage (also called

‘slicing’) in the RNA:RNA duplex region opposite the phosphate linkage between bases 10 and 11

with respect to the 5’ end of the guide strand.[16] The cleavage fragments are then further processed

by cellular RNases.

Our current knowledge of the RNAi pathway is sufficient to recognize the intrinsic gene silencing

potential of RNAi. The fact that siRNAs employ endogenous cellular machinery is imperative in this

regard. SiRNA activated RISC (RISC*) is defined as a multiple‐turnover enzyme that follows Michaelis‐

Menten kinetics. Once the mRNA target is cleaved, RISC* is recycled in the RNAi pathway to identify

and destroy other mRNAs. This implies that a single enzyme complex can bind and cleave multiple

mRNA transcripts.[17] Moreover, RISC can provide additional intracellular stability to the siRNA guide

strand and protect it against an armada of single‐strand specific RNases. The entrapment of siRNA

into RISC and its catalytic action are key parameters explaining the inherent gene silencing efficiency.

On the other hand, when siRNAs are not incorporated into RISC, their double‐stranded nature makes

them more resistant to intracellular degradation. In contrast to single‐stranded RNA, duplex siRNA

has the structural advantage of being better protected against ubiquitous single‐strand specific

nucleases. This feature makes duplex siRNA significantly more stable intracellularly, which can

already be observed on a short time scale.[18] This inherent stability also has to be taken into account

especially when comparing the gene silencing efficiency and duration of siRNAs and unmodified

antisense oligonucleotides, the latter being rapidly degraded in the cell cytoplasm.[19]

25


FIGURE 1.1 | Main cellular mechanism of RNA interference (RNAi), mediated by small interfering RNAs (siRNAs). The natural RNAi pathway is initiated by the cleavage of long double‐stranded RNA (dsRNA) by the Dicer/TRBP heterodimer complex into siRNAs. These siRNAs are subsequently incorporated into RISC (RNA Induced Silencing Complex). RISC only maintains one strand (antisense strand or guide strand) while Ago‐2, the catalytic unit of RISC, cleaves the passenger strand (sense strand). The siRNA guide strand in active RISC* then recognizes complementary target sites to direct mRNA cleavage, which is again catalyzed by Ago‐2. Further degradation of the unprotected mRNA fragments is carried out by intracellular ribonucleases. TRBP = TAR RNA binding protein, Ago‐2 = Argonaute 2.

Efforts with the aim of improving siRNA potency led e.g. to the development of longer siRNAs (25‐

27mers) that show superior silencing efficiency at certain target sites when compared to the

conventional 21 mers directed against the same region in the mRNA transcript.[10,20,21] With these

types of duplexes, maximal inhibition and longer gene silencing persistence was observed at

concentrations lower than those required for conventional 21mer siRNAs.[10,20‐22] The improved RNAi

silencing is thought to originate from their recognition and cleavage by Dicer (hence the annotation

Dicer substrate RNAs or DsiRNAs), which could facilitate their incorporation into RISC (Figure 1.1) as

Dicer is believed to participate in the early steps of RISC assembly.[23]

Nature holds several examples of convenient RNAi amplification mechanisms, e.g. in nematodes and

plants. One of the known amplification mechanisms is ‘systemic spreading’ of the RNAi silencing

effect. This effect is attributed to a dsRNA‐receptor, denoted as SID‐1 in C. elegans, which is

26


responsible for passive dsRNA translocation over the cell membrane and intercellular dsRNA

transport between neighbouring cells.[24,25] The lack of gene silencing when incubating mammalian

cells with ‘naked’ (i.e. unformulated) siRNA makes it likely that in most mammalian cell types normal

levels of SID‐1 homologue expression can be neglected.[26,27] However, a recent study conducted by

Wolfrum et al. revealed that SID‐1 is at least partially responsible for the in vitro hepatocellular

uptake of lipophilic siRNAs when they are incorporated in lipoprotein complexes such as HDL and LDL

(high and low density lipoprotein respectively) .[28] Specific RNAi silencing of SID‐1 expression in

HepG2 cells and blocking of extracellular SID‐1 epitopes with SID‐1 antibodies decreased the cellular

internalization and subsequent silencing effect of lipophilic siRNAs targeted against apolipoprotein B

(apoB) .[28]

Plants, fungi and worms contain an endogenous RNA dependent RNA polymerase (RdRP) that

produces secondary dsRNAs from an mRNA transcript targeted by a primary siRNA. The resulting

dsRNA can again be recognized and cleaved by Dicer, increasing the intracellular siRNA

concentration. Through this amplification process a small amount of ‘initiator’ dsRNA can lead to

persistent gene silencing.[29‐31] As mammalian cells most likely lack RNA dependent polymerase

activity, one must rely on alternative strategies to achieve prolonged gene silencing.[29,32,33]

STABLE RNAi GENE SILENCING

Twenty‐one mer siRNAs, mimicking Dicer cleavage products, can be chemically synthesized and

introduced into the target cell (Figure 1.1) but they can also be produced intracellularly from short

hairpin RNA (shRNA) precursors that can be continuously expressed from RNA polymerase driven

expression cassettes (Figure 1.2).[34,35] ShRNAs are structurally and functionally related to pre‐

microRNAs, intermediates in the biogenesis of endogenously encoded microRNAs (miRNAs). MiRNAs

constitute a highly conserved class of small RNAs that mediate RNAi mainly through translational

inhibition.[36,37] As Figure 1.2 illustrates, the first step in the production of miRNA occurs in the cell

nucleus, where long primary transcripts (called pri‐miRNA) are expressed from endogenous genetic

regions and processed by Drosha (a RNase III enzyme), with the aid of its cofactor DGCR8, into ~60‐

70 nt hairpins with imperfect complementarities in their stems.[38‐40] These precursor miRNAs (pre‐

miRNAs) are shuttled from the nucleus into the cytoplasm through the Exportin‐5/Ran‐GTP

heterodimer complex.[41] In the cytosol, Dicer processes pre‐miRNA into mature ~22 nt miRNA

duplexes that interact with RISC to modulate transcriptome expression.[42]

27


FIGURE 1.2 | Overview of the microRNA (miRNA) pathway. The presence of miRNAs in the cell cytoplasm leads to modulation of gene expression by translational repression. Pri‐miRNA primary transcripts are processed in the nucleoplasm by the RNase III enzyme Drosha. The resulting pre‐miRNA (precursor miRNA) is shuttled to the cytoplasm by Exportin5/Ran‐GTP, where it is recognized and cleaved by Dicer to form mature miRNAs. On their turn, miRNAs are incorporated into RISC, that only retains one strand (most likely with the help of a helicase), thereby allowing the complex to bind to a (partially) complementary region in the mRNA. It is believed that the targeted transcripts are sequestered to cytoplasmic foci (P‐bodies) that are inaccessible to the translational machinery. TRBP = TAR RNA binding protein, DGCR8 = DiGeorge syndrome chromosomal region 8, also known as Pasha in C. elegans and Drosophila.

Mammalian miRNAs, which tend to contain mismatches with the target mRNA, most often mediate

gene silencing through translational suppression rather than transcript slicing, the latter being the

result of a perfect match between trigger and target. It is believed that the binding of activated

miRISC* to the 3’ untranslated region (3’ UTR) of the target mRNA sequesters the mRNA from the

translational machinery through confinement in cytoplasmic foci, called processing bodies (P‐bodies),

28


thereby preventing protein synthesis.[43] To date, the exact outcome of imperfect complementarity

with target mRNA is still subject to debate as multiple models have already been proposed.[44]

In analogy with miRNA biogenesis, the intracellular processing of shRNAs originates in the cell

nucleus, where Exportin‐5 is responsible for its nuclear export.[45] In the cytoplasm, these shRNAs are

again cleaved by Dicer to yield the active 21mer siRNAs.[34] In correspondence with Dicer substrate

RNAs (DsiRNAs), plasmid vectors can be designed to produce shRNAs (29 nt stem, 4 nt loop) with

higher RNAi potency over smaller hairpin RNAs (19 nt stem, 4 nt loop) due to improved Dicer

recognition and RISC incorporation.[46] However, in a recent report by Li et al., it was shown that in

the context of a 9 nt loop, the shRNA with a 19 nt stem outperformed the longer 29 nt shRNA.[47] The

explanation for this discrepancy again lies in the ability of the shRNA to be recognized by Dicer.

Indeed, whereas shRNAs with a 19 nt stem and a 4 nt loop bypass Dicer cleavage, increasing the loop

length to 9 nt also turns shRNAs with a 19 nt stem into Dicer substrates.[47,48] Building on our

increased understanding of miRNA processing, second‐generation shRNAs were developed (called

shRNAmir).[49] In contrast to first‐generation shRNA that elicit structural analogy with pre‐miRNA,

shRNAmir are transcribed as pri‐miRNA. Their design is based on the human miRNA, miR‐30, of which

the stem is replaced with a RNA sequence of interest. The advantage of shRNAmir over shRNA lies in

their recognition and processing by both Drosha (nucleoplasm) and Dicer (cytoplasm), leading to a

more efficient intracellular production of mature siRNAs and increased knockdown efficiency.[50]

Important in the context of this chapter is to note that, compared with synthetic siRNAs that induce a

transient knockdown, plasmid vector based interference shows better potential for long‐term gene

silencing.[51] To establish stable RNAi gene silencing in cultured cells, researchers mostly appeal to

viral expression vectors.[34] Lentiviral vectors allow genomic insertion of the shRNA expressing

transgene, while adenoviral and adeno‐associated viral vectors (AAVs) show less successful

integration and viral DNA remains largely episomal.[52] Genomic integration of shRNA expression

cassettes per definition results in stable RNAi mediated gene silencing since the target gene remains

silenced as long as transcription of siRNA precursors proceeds. This RNAi gene therapy approach can

be of interest to study loss‐of‐function phenotypes following prolonged knockdown of a target gene

and for long‐term treatment of chronic diseases (e.g. viral infections such as hepatitis B [53] and

hepatitis C), thereby easing the treatment schedule and improving patient comfort. However, in a

clinical setting, the use of synthetic siRNAs for transient RNAi gene silencing may be advantageous

over continuous shRNA/siRNA production as the dosing schedule (and consequently also the

resulting intracellular siRNA concentrations) in a therapeutic regimen can be more easily adapted

related to therapeutic needs.[54] Some applications will call for a fully‐reversible but long‐duration

gene silencing, without loss of activity following repeated administration. Additionally, transgenes

29


that are stably integrated in the host genome can be silenced rapidly by histone modifications and

hypermethylation of CpG islands in the promoter region. Instead of achieving long‐term transgene

expression, this chromatin silencing will result in a gradual extinction of transgene activity.[55]

Recently, Mark Kay and colleagues disclosed that continuous expression of high intracellular levels of

shRNA could result in long‐term toxicity in mice.[56] The observed fatal side effects were primarily

ascribed to saturation of Exportin‐5 leading to interference with nuclear export of miRNA precursors

and miRNA function.[45] Possible solutions to overcome these toxicity issues are (a) lowering the

initial viral load and (b) implementing vector development with inducible or tissue‐specific promoters

that allow more feasible control over intracellular shRNA concentration which may improve the

activity/toxicity ratio.[56‐58] In a mouse model of hyperbilirubinemia, it was shown that the adenoviral

production of shRNAs could silence Abcc2 function (ATP‐binding cassette multidrug resistance

protein 2), involved in liver bilirubin transport, for up to 3 weeks. This effect did not seem to

correlate with changes in the level of endogenous (precursors of) miRNAs.[59]

Nonetheless, it is striking to conclude that the improved RNAi gene silencing of dsRNAs by taking

advantage of the natural miRNA pathway in an earlier stage can have a toxic flip side. In this regard,

synthetic 21mer siRNAs that function further downstream in the RNAi pathway are the better and

safer choice because they bypass Dicer cleavage and do not require nuclear export for their

activity.[45,60] However, when synthetic siRNAs are employed, vigilance is still required, as an

intracellular excess of siRNA may possibly interfere with RISC availability which can also induce

competition with cellular miRNAs.[60‐63] A recent report describing potent and specific knockdown of

hepatocyte‐specific genes in mice and hamsters after systemic delivery of lipid‐formulated siRNAs,

also demonstrated that the treatment schedule did not influence either miRNA biogenesis or miRNA

function.[64] As our knowledge on miRNAs expands, we will be able to better assess any possible

cellular disturbances as a result of siRNA/shRNA treatment. Quantitative data on both the total

number of cellular protein molecules involved in the RNAi pathway and the number of siRNA

molecules that is needed inside a cell for a sufficient silencing effect, could be very helpful towards

optimization of siRNA therapy without the aforementioned toxicity issues.[62]

Another concern that has been raised against the use of (non‐integrating) viral vectors is their

immunogenicity, especially when long‐term use is necessary (e.g. adenoviruses, AAVs) .[3,51]

Furthermore, random genomic insertion of a transgene (e.g. with integrating lentiviral vectors)

coincides with the risk of hazardous in vivo viral recombination and insertional mutagenesis.[65,66]

Although at present viruses are still the most efficient gene delivery vectors, the combination of

these adverse effects could eventually favour a non‐viral approach. Meanwhile, non‐viral strategies

have also been explored for intracellular shRNA production.[51] SiRNA expression plasmids can be

30


introduced in the target cell by complexation with cationic lipids and polymers or by physical

methods like electroporation and microinjection. Unfortunately, the nuclear import of the plasmid

DNA (pDNA) vector remains the predominant bottleneck in the gene delivery process,[67] thereby

giving preference to synthetic siRNAs which function mainly in the cell cytoplasm. Especially non‐

dividing cells are difficult to transfect with non‐viral vectors, since insufficient amounts of the pDNA

can reach the nucleus in the absence of cell division during which the nuclear barrier is temporarily

disassembled.[68]

PERSPECTIVES ON THE DELIVERY OF siRNA: BIOLOGICAL BARRIERS

FIGURE 1.3 | Overview of the main biological barriers for siRNA therapy following intravenous (IV) injection. For in vivo application, siRNAs can be formulated into nanosized carriers that should fulfill some basic requirements. They should be large enough to circumvent renal clearance, but small enough to be able to cross the capillary endothelium and accumulate in the target tissue. Moreover, the siRNA carrier has to be able to evade uptake by the mononuclear phagocyte system and has to protect the siRNA against circulating RNases. Once the carrier has reached the target cell, it needs to deliver the siRNA to the cytoplasm. This usually involves cellular uptake through endocytosis, followed by escape from the endosome and carrier disassembly in the cytosol. ECM: extracellular matrix, NPC: nuclear pore complex, TGS: transcriptional gene silencing.

31


Although it is generally said that chemically synthesized siRNAs are promising therapeutic candidates

for the treatment of various genetic pathologies, their in vivo drug‐like properties are regarded as

very unfavourable. Before siRNAs can reach the desired intracellular location, many extra‐ and

intracellular barriers have to be overcome (Figure 1.3).[26] Obviously, the efficiency with which siRNAs

can by‐pass these obstacles will have major implications on the extent and duration of gene silencing

and, subsequently, on the determination of a clinically relevant dosing interval. Several concepts and

strategies have been proposed to improve the siRNA pharmacokinetics and eventually their

therapeutic performance.

The extracellular compartment: focus on chemical modification and siRNA formulation

It is well known that siRNAs are rapidly degraded in the extracellular environment. When naked (i.e.

uncomplexed, formulated in saline) siRNA is injected intravenously, it is recognized and cleaved by

RNase A type endonucleases [69‐71] or 3’ exonucleases [72‐74] limiting the serum half‐life to < 30 min.

Therefore, much effort has been undertaken to chemically modify siRNA therapeutics, in order to

reduce their susceptibility to serum RNases.[10,75,76] Chemical modifications include RNA backbone

modifications (e.g. phosphorothioates and boranophosphates), 2’‐ribose modifications, and terminal

3’ and 5’ modifications, amongst others.[77] The most important chemical modifications introduced

into an RNA backbone to date are illustrated in Figure 1.4.[72,78] Initially some concerns were raised

with regard to the possible diminished efficacy of chemically modified siRNAs, due to interference

with RISC incorporation, activation and/or recycling. Since then, however, elaborate research in this

field has proved that chemically modified siRNA can also efficiently lower mRNA levels through a

sequence‐specific RNAi dependent pathway and some modifications even show enhanced gene

silencing compared to the unmodified counterpart.[79,80] An important consideration to make is

whether or not chemical modifications, with the aim of improving the extracellular siRNA stability,

are promising in light of prolonging the knockdown of a target gene. One could expect chemically

modified siRNAs to have a higher bioavailability as they stay intact for a longer period of time in the

blood circulation, thereby also improving the intensity and duration of the RNAi gene silencing effect.

Unexpectedly, Layzer and coworkers were not able to demonstrate a stronger silencing or

persistence of silencing for stabilized 2’‐fluoro pyrimidine siRNAs over the unmodified 2’‐OH siRNAs

after hydrodynamic tail vein injection.[81] Most likely, sufficiently high concentrations of both siRNAs

can reach the intracellular target site after hydrodynamic injection and exert an effect before

nuclease digestion becomes prominent.

32


FIGURE 1.4 | Possible chemical modifications that can be incorporated into a siRNA duplex. Modifications include alterations in the phosphodiester backbone (e.g. boranophosphate, methylphosphonate and phosphorothioate) or sugar modifications (LNA, FANA, 4’‐thio and 2’‐OH modifications). FANA = 2’‐fluoro‐β‐D‐arabinonucleic acid, MOE = 2‐methoxyethyl and LNA = locked nucleic acid. Adapted from reference [72].

In contrast to these findings, a modified siRNA duplex, targeted against hepatitis B viral (HBV) RNA

and where all ribose 2’‐OHs were substituted, was shown to be more efficient in decreasing HBV

DNA and HBV surface antigen serum levels when compared to non‐modified siRNA.[76] In the same

report, hydrodynamic injection was used to co‐deliver the siRNA and a replication competent HBV

vector. Surprisingly, the difference in activity between modified and unmodified siRNA was only

significant at the high‐dose level and almost absent at lower doses.[76] When in vivo siRNA

degradation by nucleases is the main limiting factor, one would expect to observe the opposite

trend. This also indicates that factors other than increased nuclease resistance have to be taken into

account when working with chemically modified siRNAs.

Besides degradation by RNases, the short in vivo half life of siRNAs is also influenced by their rapid

renal clearance. Since siRNAs are highly hydrophilic (~40 negatively charged phosphate groups per

siRNA molecule) and have a molecular weight (~14 kDa) far below the cut‐off for glomerular filtration

(~60 kDa), one can expect renal clearance to be prominent. Although RNase digestion can be slowed

33


down by chemically modifying the nucleic acid backbone, renal clearance seems to be the rate

limiting factor governing the in vivo half life of siRNAs.[62,82]

Some chemical modifications (e.g. phosphorothioate internucleotide linkages or 4’‐thio

ribonucleotides) can increase the circulatory half‐life of siRNA by promoting interaction with serum

proteins.[77] A more convenient strategy to modulate and improve the siRNA biodistribution and/or

increase cellular uptake is the more dramatic alteration of siRNA structure by conjugation to small

molecular weight moieties.[77,83‐85] Interesting examples are the coupling of siRNA to heavy‐chain

antibody Fab fragments [86] or lipidic moieties like cholesterol, bile acids and long‐chain fatty

acids.[28,87,88] Lipophilic siRNAs seem to incorporate selectively in lipoprotein particles which are rich

in phospholipids and cholesterol (mainly HDL and LDL). These lipophilic siRNA‐lipoprotein complexes

are able to improve siRNA biodistribution by evading renal clearance and promoting cellular uptake

through HDL and LDL lipoprotein receptors.[28] Chimeric peptide‐siRNA complexes, as an alternative

delivery strategy, have recently been shown to protect mice against infection with Japanese

encephalitis virus by delivering their siRNA payload into the brain after intravenous injection.[89]

Intriguingly, the majority of the siRNA treated mice survived for over 4 weeks, while all animals in the

control group died within 10 days. Rozema et al. recently introduced a novel polymer based siRNA‐

conjugate strategy (termed Dynamic Polyconjugates) for in vivo hepatocytic targeting. These

conjugates (~10 nm in size) consist of a reversibly shielded membrane‐destabilizing polycation,

containing a cleavable targeting ligand, to which siRNA is linked through an intracellularly reducible

disulfide bond. Effective knockdown of two endogenous hepatic genes apoB and peroxisome

proliferators‐activated receptor alpha (ppara) was demonstrated in wild‐type mice following low

pressure i.v. injection.[90] In addition, aptamer‐siRNA conjugates seem promising for cell‐type specific

delivery of siRNA, as demonstrated by two groups using a prostate‐specific membrane antigen

(PSMA) targeted aptamer, either directly conjugated to siRNA or linked through streptavidin.[91,92]

The use of cell penetrating peptides (CPPs) for siRNA delivery to date has also produced some

interesting reports, but the exploration of this area of research is only just commencing.[93,94]

Another convenient way to circumvent renal clearance is by the incorporation of siRNA into gene

silencing nanocomplexes that are large enough to evade glomerular filtration, thereby improving

siRNA pharmacokinetics and biodistribution.[85] Paying proper attention to the “packaging” of siRNAs,

can lead to constructs where the siRNA molecules are shielded from circulating RNases and

destabilizing blood components, eliminating the need for chemical modification.[95‐97] Many inventive

siRNA formulations have already shown to dramatically improve the in vivo potency of

siRNAs.[3,11,62,83,98‐100] A textbook case, underlining the importance of siRNA formulation is given by

Morrissey et al. When administering unformulated backbone stabilized anti‐HBV siRNAs through

34


hydrodynamic tail vein injection, an effective inhibition of HBV replication could be achieved, albeit

at a dramatically high dosing regimen of three daily doses of 30 mg kg‐1.[76] However, formulating the

therapeutic siRNAs into stable nucleic acid lipid particles (SNALPs) of ~140 nm in size, these

investigators succeeded in achieving a significant and long‐term reduction of HBV activity by three

daily injections of 3 mg kg‐1 d‐1 siRNA followed by a weekly administered maintenance dose for up to

6 weeks.[101] A more recent report, using SNALP technology to deliver anti‐ApoB siRNA in cynomolgus

monkeys, describes marked reduction in plasma ApoB protein levels for as long as 11 days after a

single IV injection (2.5 mg kg‐1).[102] Mark Davis’ group formulated siRNAs in cationic cyclodextrin

containing polycations (CDPs) which can be equipped with targeting ligands, e.g. transferrin for

delivery to tumour cells expressing the transferrin receptor. The systemic delivery of these targeted

siRNA nanoparticles, twice‐weekly for > 4 weeks resulted in long‐term inhibition of tumour cell

engraftment and tumour growth in a murine therapeutic model for metastatic Ewing’s sarcoma.[103]

Interestingly, the long‐term inhibition was only observed for the formulation where the surface was

modified with transferrin moieties and no positive outcome was demonstrated for the non‐targeted

formulation. This targeted nanoparticle delivery system also has been evaluated recently in non‐

human primates [104] and is currently in a phase I clinical trial.[105] Some of the cited delivery

approaches are schematically drawn in Figure 1.5.

The intracellular compartment: focus on chemical modification and cellular dilution

Besides improvements at the extracellular level, also the intracellular behaviour of chemically

modified siRNAs may play a pivotal role in improving the RNAi gene silencing potency and duration.

Some reports ascribe a prolonged gene silencing effect in cell cultures to the use of stabilized siRNA

duplexes that show higher resistance towards nuclease attack, suggesting that intracellular

degradation of siRNA can indeed be partially responsible for transient RNAi gene silencing.[106‐110] An

interesting recent report by Takabatake and coworkers demonstrates the dependence of RNAi gene

silencing duration on the tissue expression level of eri‐1 (enhanced RNAi‐1) nuclease.[111] Eri‐1 can be

regarded as a type of siRNAse, which has been discovered in C. elegans mutants.[73] Preliminary

experiments in EGFP transgenic rats revealed that a single administration of siEGFP into the kidney

resulted in the reduction of EGFP expression for up to two weeks, while the transfection of siEGFP

into muscle tissue silenced EGFP for up to 90 days. The authors hypothesize that the higher eri‐1

expression level in the kidney when compared to muscle could be responsible for higher intracellular

siRNA degradation and decreased RNAi longevity. They propose that blocking eri‐1 expression or

applying eri‐1‐resistant siRNAs can result in more sustained gene silencing.[111]

35


FIGURE 1.5 | Delivery systems for siRNA. (a) lipophilic covalent conjugate of siRNA with cholesterol, (b) aptamer‐siRNA conjugate for targeted siRNA delivery, (c) encapsulation of siRNA in stable nucleic acid lipid particles or SNALPs, composed of cationic lipids for siRNA complexation together with a fusogenic and pegylated lipid component, (d) polymer nanoparticle delivery concept consisting of cyclodextrin polycations for siRNA complexation. Targeting moieties can be linked by the formation of adamantane‐cyclodextrin inclusion complexes.

SiRNAs modified with 2’‐fluoro pyrimidines have shown increased persistence in luciferase silencing

over unmodified siRNAs in cells transiently transfected with a luciferase reporter vector.[107]

However, assessment of luciferase silencing with the same type of modified siRNAs in a cell line

stably transfected with the luciferase reporter gene, did not reveal a significant difference in the

duration of gene silencing.[81] As explained by Layzer et al., when trying to investigate gene silencing

in cell culture on a longer time scale, it is preferred to approximate in vivo conditions as much as

possible by using cell culture models that stably express the transgene of interest.[81] It was

questioned recently whether simultaneous transfection of an exogenous gene and the siRNA is

suitable to quantify RNAi.[112] Volkov et al. claimed in outspoken terms that the main purpose of

36


siRNA protection against nuclease attack is the extension of the duration of the siRNA effect.

Interestingly, these authors concluded that only protection of all the nuclease‐sensitive sites in an

siRNA duplex could result in RNAi prolongation, and that partial modification does not induce a more

durable gene silencing when compared to the unmodified duplex.[109]

Fisher et al. investigated the influence of modest altritol (6‐carbon sugar) modifications on siRNA

stability and gene silencing persistence.[113] Although partially modified siRNAs did not show

increased nuclease resistance compared to wild‐type siRNA, the authors did observe enhanced

duration in MDR‐1 gene silencing resulting in more durable suppression of p‐glycoprotein expression.

These data underscore the fact that, at the cellular level, other factors than mere siRNA stability, like

cellular uptake and/or recognition by RNAi machinery, have to be taken into account. Liao and Wang

reported on 2’‐O‐(2,4‐dinitrophenyl) modifications that serve to enhance siRNA lipophilicity.[80] These

chemical adjustments facilitate the intracellular siRNA penetration and improve the siRNA stability

which resulted in a stronger inhibition of insulin‐like growth factor receptor expression, relative to

unmodified siRNA directed against the same target sequence.

Interestingly, it seems that in fast dividing cell lines gene silencing is limited by intracellular dilution

of siRNA (and activated RISC) due to cell division. As a result, in vitro gene silencing usually lasts for

only ~4‐7 days.[54] In slowly dividing and non‐dividing cells the siRNA stability is supposed to be the

main limiting factor for prolonged gene silencing as the intracellular siRNA half‐life is expected to be

shorter than the cell division time.[54] Intracellular persistence of active siRNA has already been

shown to maintain target gene inhibition for several weeks in different types of (non‐dividing)

primary cells.[54,114‐117] In these papers it was assumed that the stability of the RNAi effect is

representative of the physical stability of the intracellular siRNA trigger. Unfortunately, in most

reports this assumption was not experimentally evaluated. Song et al., who did perform a northern

blot analysis for siRNA integrity, suggested that intracellular siRNA survival requires the active

production of the target mRNA transcript.[114] Mantei et al. identified intracellular siRNA stability as

the critical parameter influencing RNAi effect duration in primary T lymphocytes. Whereas target

gene expression was rapidly restored to base levels for unmodified siRNAs upon T cell activation, the

gene inhibition was dramatically lengthened with chemically modified siRNAs. The authors ascribe

this effect to the increased intracellular half‐life of the modified siRNA, rather than dilution resulting

from T cell proliferation.[110]

Also Maliyekkel et al. reported on more stable gene silencing following transient shRNA induction in

growth arrested non‐cycling cells.[115] However, in contrast with other reports on long‐lasting RNAi in

non‐dividing cells,[114] the phenotypic RNAi activity did not coincide with the intracellular siRNA decay

as analyzed by a RNase protection assay. Two possible explanations for this discrepancy were

37


suggested. Firstly, it is conceivable that the RNAi phenotype is mainly governed by the fraction of

siRNA associated with RISC*. As already mentioned previously, it can be expected that incorporation

of the guide strand in the RISC* complex provides extra protection against degradation. In this case

quantifying total intracellular siRNA does not represent the effective intracellular RNAi potential.

Secondly, it was proposed that transcriptional silencing of the target gene could be held responsible

for the increased silencing duration ( more information on transcriptional silencing can be found in in

the following section). Recent in vitro and in vivo data on non‐modified and nuclease stabilized siRNA

duplexes have shown that evading nuclease attack by chemically modifying siRNAs does not

significantly prolong the duration of gene silencing once the siRNA has reached the cytosol of the

target cells.[118]

The contradictory findings on the (supposed) advantages of nuclease stabilized siRNAs described in

this section unfortunately hinder drawing clear conclusions in this regard. Nonetheless, although the

benefits of nuclease resistant siRNAs inside the cell cytoplasm still remain questionable, chemical

modifications have proved to be a very useful approach to abrogate off‐target effects and unwanted

stimulation of the mammalian immune system.[119‐122] Several native siRNA sequences are known to

induce a toll‐like receptor (TLR) mediated immune response through endosomal TLR7/8

recognition.[123,124] Cytoplasmic receptors such as PKR (dsRNA‐binding protein kinase receptor), and

the RNA helicases RIG‐1 (retinoic acid‐inducible gene‐I) and MDA‐5 (melanoma differentiation‐

associated protein‐5) are considered to recognize certain structural RNA characteristics, other than

nucleotide sequence.[122] RIG‐1 for instance is activated by uncapped 5’‐triphosphate RNA and longer

blunt‐ended siRNAs.[122,125] Mitigation of immune response is possible by designing siRNAs devoid of

immunostimulatory sequence motifs and by paying proper attention to dsRNA physical structure.

The most robust approach however is the selective incorporation of modified nucleotides (e.g. 2’‐

OMe) to avoid immune receptor activation.[122,126] Kleinman and coworkers showed recently that 21‐

nt or longer siRNAs inhibited choroidal neovascularization (CNV) in mice in a sequence‐ and target

independent manner. The inhibition resulted from the cell‐surface binding of TLR3 by generic

siRNAs.[127] Although this approach could offer perspectives to harness this immune activation for

antiangiogenic effects,[127,128] excessive cytokine release can cause severe toxicity.[129] Modifying the

siRNA duplex to minimize TLR3 activation could reduce any consequential undesired effect and

enhance target specificity. Therefore, chemical modifications can certainly help to optimise siRNA

design towards minimal in vivo toxicity and maximal siRNA tolerance.

38


RNAi INDUCED SILENCING AT THE TRANSCRIPTIONAL LEVEL

The majority of reports in the literature dealing with mammalian RNAi gene silencing describe siRNA‐

directed cleavage and destruction of cytoplasmic mRNA transcripts, termed post‐transcriptional gene

silencing (PTGS). Although it was first believed that siRNA only functions in the cell cytoplasm,[130] a

series of recent reports suggest that RNAi may also be active in the cell nucleus. As an example, the

RNAi nuclear function was shown by siRNA‐mediated degradation of 7SK snRNA,[131] an abundant,

well‐characterized RNA that has a highly defined structure and specifically localizes in the nucleus.[132‐

134] Furthermore, it was shown that several siRNAs incorporated in functional RISC existed in the

nucleus and cleaved the target RNA with high efficiency.[131] Irrespective of the fact that siRNA

molecules are small enough to passively diffuse through the nuclear pore complex (NPC), recent data

suggest that nuclear RISC (nRISC) originates from a partial translocation of cytoplasmic RISC (cRISC)

followed by shuttling in the nucleus.[135] Although the cited reports describe RNAi action occurring in

the cell nucleus, this still encompasses a PTGS process, for which essentially the same basic rules

apply as discussed for cytosolic RNAi (see section 2 of this chapter).

To date, accumulating evidence emerges that small interfering RNAs have the ability to silence the

expression of eukaryotic genes by targeting their promoter region, a process which is denoted as

transcriptional gene silencing or TGS. This phenomenon was first observed in lower organisms and

plants, e.g. when doubly transformed tobacco plants exhibited a suppressed phenotype of a

transgene caused by a methylation process.[136] This RNA‐dependent DNA methylation (RdDM)

seemed to be induced by RNA sequences identical to genomic promoter regions, leading to TGS.[137‐

141] While the exact molecular mechanisms of RdDM are still to be fully clarified, it is likely to involve

cytosine methylation, histone modification and chromatin remodelling.

Several independent recent reports also demonstrated transcriptional gene silencing by

siRNAs/shRNAs in mammalian cells.[142‐144] SiRNA‐mediated TGS in mammalian cells appears to be the

result of the siRNA‐directed histone H3K9 and H3K27 methylation at the targeted promoter,[145]

although subsequent DNA methylation has also been observed.[144,146] However, the role of DNA

methylation in TGS remains controversial and highly debated since many reports describe that

potent silencing on the transcriptional level does not necessarily require DNA methylation. A paper

from Janowski et al. describes the inhibition of gene expression by so called antigene RNAs

(agRNAs), complementary to transcription start sites within human chromosomal DNA.[147] In

contrast to other reports, no evidence for a role of DNA methylation in promoter silencing could be

obtained, but the silencing was accompanied by dimethylation of Lys9 in histone H3 (H3K9).[148,149]

Kim et al. and Janowski et al. both investigated the involvement of Argonaute proteins in the TGS

39


process. While John Rossi’s group only identified the enrichment of Ago‐1 at the targeted promoter

region,[149] Janowski and co‐workers reported that the siRNA‐mediated silencing of both Ago‐1 and

Ago‐2 seemed to reverse TGS induced by agRNA, suggesting that both Argonautes play a role in the

promoter silencing.[150] Adding to the vast complexity of this field are reports describing TGS by

endogenously encoded miRNAs,[151] or even the transcriptional activation induced by small RNAs

instead of repression.[152,153] It may be clear that the exact molecular mechanisms for TGS or

transcriptional activation still remain rather obscure to date and more data are needed to further

clarify the underlying pathways.

However, evidence on TGS at present is sufficient to envision the use of small RNAs targeting

promoters for therapeutic purposes. An excellent example underlining the therapeutic potential of

TGS can be found in HIV‐1 treatment. To suppress the production of a range of viruses in vitro by

targeting structural and accessory genes, the duration of the PTGS effect is known to be rather

limited, i.e. in HIV‐1 varying from 4 to 7 days.[154] Prolongation of the HIV‐1 silencing up to 14‐25 days

was achieved using adeno‐associated or lentiviral vectors. However, the efficacy of HIV‐1 treatment

based on a PTGS approach is potentially further limited as HIV‐1 is known to adapt to environmental

pressure and rapid selection of siRNA escape mutants has been described in vitro.[155] For this reason,

Suzuki et al. made use of TGS as an alternative approach to prolong the suppressive effect of siRNA

that would be less susceptible to the adaptability of HIV‐1.[156] SiRNA, targeting HIV‐1 LTR, induced a

significant reduction in the reverse transcriptase levels, an effect that was maintained for over one

month. A prolonged effect can be expected as it has been shown that, regardless of the exact

underlying mechanisms, RdDM is long lasting and can be passed on across generations in plant

systems [157‐159] and in C. elegans in the absence of the original RNAi trigger.[160] Therefore,

transcriptional gene silencing through the use of promoter specific siRNAs could be an interesting

method of achieving a more robust gene silencing effect, as recently shown for the first time in

human cells by Kevin Morris’ group, although it has been suggested that this probably needs a

minimal sustained expression of promoter directed siRNA/shRNA in order to increase DNA

methylation events.[161]

MATHEMATICAL MODELLING TO DESCRIBE RNAi GENE SILENCING KINETICS

A broad spectrum of variable parameters can be defined that may influence the outcome of RNAi

gene silencing and the knockdown duration. These parameters are situated on three distinct levels:

(1) the siRNA delivery formulation, (2) the intracellular RNAi pathway and (3) the envisioned target.

40


At the level of siRNA delivery one must keep in mind that in vivo the tissue distribution and

pharmacokinetics of the siRNA formulation (e.g. siRNA containing nanoparticles) will have a major

impact on the RNAi effect in the target cells, as discussed earlier. Because the siRNA targets are

located intracellularly, not only the biodistribution in the extracellular space, but also the different

steps in intracellular trafficking have to be considered. Important processes like the cellular uptake

mechanism, the confinement of siRNA carrier to cytosolic vesicles and siRNA release from its carrier

will influence the RNAi effect (Figure 1.3). For example, several studies have shown that the ability of

siRNA or siRNA carrier complexes to escape from the endosomal compartment can be a limiting step

in their gene silencing efficiency.[162,163]

Obviously, RNAi‐specific parameters such as intracellular siRNA stability, RISC activation and siRNA‐

RISC‐mRNA complex kinetics have to be accounted for. The importance of RISC* recycling in the RNAi

pathway towards siRNA gene silencing potency and the potential influence of siRNA chemical

modification was emphasized earlier.

Last but not least, both the turnover of the intracellular mRNA transcript and target protein stability

should be carefully considered. If the protein half‐life extends over several days, even efficient

knockdown of intracellular mRNA levels after a single siRNA dose may fail in altering the cellular

phenotype owing to residual amounts of the stable protein. The rate of cell division is imperative for

the duration of gene silencing in a transient RNAi approach as it determines how transient the RNAi

effect really is. In this regard, both the targeted cell type and the envisioned therapeutic outcome are

of great importance. If a single administration of siRNA is effective in blocking tumour cell growth,

this inherently could imply a prolonged effect since the intracellular siRNA dilution due to cell

division can be disregarded.[164]

A detailed knowledge of the impact of all the parameters listed above, in relation to RNAi gene

silencing kinetics, could expedite the design of therapeutic siRNA strategies. Predictive mathematical

models, illuminating the key factors that govern the duration of gene silencing, could be helpful

instruments in the setup of siRNA treatment regimens. Several reports that describe mathematical

equations to gain more insight into the RNAi silencing pathway are available in the literature.[165,166]

Raab and Stephanopoulos studied the impact of the siRNA concentration and the time of siRNA

transfection relative to reporter plasmid co‐transfection in mouse hepatoma cells.[12] Their model

could be used to select the appropriate time of siRNA transfection as a function of gene induction.

Moment analysis was used by Takahashi and coworkers to evaluate the time‐course of endogenous

protein expression as a function of siRNA concentration.[167] Likewise, an interesting report by Arciero

et al. describes mathematical modelling as a tool to predict tumour‐immune evasion and to study the

41


impact of siRNA administration directed against the immune suppressive cytokine TGF‐β on tumour

growth.[164,168]

Bartlett and Davis were the first to construct a mathematical model incorporating parameters that

govern the in vivo siRNA delivery process, such as biodistribution of the siRNA carrier and

intracellular trafficking (vector unpackaging and endosomal escape), to study the kinetics of siRNA‐

mediated gene silencing.[54] Model calculations were also applied to define a dosing schedule,

consisting of repeated siRNA injections, which results in persistent gene silencing, dependent on the

half‐life of the target protein and the target cell division rate. An illustrative example for non‐dividing

fibroblasts is given in Figure 1.6. The same authors further employed their mathematical model in

several in vitro and in vivo gene silencing studies performed with targeted siRNA‐nanoparticles based

on cationized cyclodextrin, to provide information that could aid in designing more effective siRNA

delivery strategies.[169]

FIGURE 1.6 | Effect of siRNA dose frequency on the duration of luciferase knockdown by siRNA in non‐dividing fibroblasts that stably express the luciferase gene. (A) Experimental results obtained with Oligofectamine formulated siRNA, directed against pGL3 luciferase. Squares, 100 nM (day 0); diamonds, 100 nM (day 0) + 10 nM (day 4); triangles, 100 nM (day 0) + 100 nM (day 4). (B) Luciferase knockdown following siRNA transfection, as predicted by mathematical modelling. Reprinted with permission from Bartlett and Davis, 2006 [54]. © Oxford University Press.

When considering different siRNA delivery strategies, also note that it could be of interest to

incorporate the delivery kinetics of siRNA from its formulation into predictive mathematical models,

as the amount of siRNA delivered into the cytoplasm as a function of time could well be a

determining factor towards the eventual therapeutic outcome and the intracellular toxicity.

42


TIME‐CONTROLLED INTRACELLULAR RELEASE OF siRNA

‘More may not always be better’, it was sharply put by Marsden in a recent edition of the New

England Journal of Medicine,[170] and can be regarded as a basic toxicological paradigm. In theory

every substance is potentially harmful when exceeding a certain concentration level or exposure

time. This is also true for siRNA therapeutics, judging by the concentration dependency of adverse

effects such as off‐target silencing, induction of immune response and saturation of the endogenous

RNAi pathway.[63,122] This obviously stresses that it is imperative to work with the most potent siRNA

designs at the lowest concentration possible. Keeping this in mind, one cannot overemphasize the

need for rigorous control over the intracellular siRNA/shRNA concentrations.

To regulate the number of active siRNA molecules in the cell cytosol, one could consider the use of

conditional shRNA expression from plasmid expression cassettes delivered in the target cells by viral

or non‐viral carriers. However, the regulation of intracellular siRNA concentrations becomes more

complex in the case of (non‐viral) synthetic siRNA delivery where mostly (electrostatic) complexes

between siRNA and (cationic) polymers (i.e. polyplexes) or (cationic) lipids (i.e. lipoplexes) are applied

to the cells. With these poly‐ and lipoplexes a transient RNAi effect lasting for less than one week is

generally observed. For many of these formulations, one can expect that upon escape from the

endosome, they provide a direct release of siRNA in the cytoplasm. The released siRNA is

subsequently prone to dilution through cell division and (possibly) intracellular degradation, leading

to a transient gene silencing. In theory, time‐controlled release of siRNA in the cell cytoplasm could

be interesting to maintain intracellular siRNA concentrations for a longer period of time above the

minimal threshold required for efficient gene silencing, without flooding the cell cytoplasm with

uncomplexed (naked) siRNAs. Controlling the amount of siRNA that is released in the cytosol could

therefore result in a prolonged RNAi effect.

To achieve this goal, there is a need for delivery vehicles that exhibit tailored time‐controlled siRNA

delivery. Our group recently published on cationic biodegradable poly‐β‐amino esters (PbAEs) which

are able to form nanosized electrostatic complexes with the negatively charged siRNA (Figure 1.7).

Following endocytosis of the polyplexes, the hydrolytic degradation of the cationic polymer is

believed to increase the osmotic pressure inside the endosomal vesicles due to an accumulation of

its degradation products.[171] When exceeding a critical pressure threshold, this eventually leads to

endosomal rupture. It is hypothesized that the polymer degradation rate and the number of

polyplexes residing in a single endosome will govern the kinetics of the rise in osmotic pressure over

the endosomal membrane. Therefore, not all endosomes are expected to rupture at the same time.

43


In this way a more gradual siRNA release into the cell cytoplasm is expected in function of polymer

degradation, thereby altering gene silencing kinetics (Figure 1.7).

FIGURE 1.7 | Long‐term gene silencing with biodegradable poly‐β‐amino esters (PbAEs) in HuH‐7 hepatoma cells that stably express the luciferase gene. The cells are transfected with (a) PbAE1:siRNA and (b) PbAE2:siRNA complexes containing siRNA targeting the pGL3 luciferase gene in different molar ratios. The chemical structure of both cationic polymers is depicted. It is hypothesized that the slower degradation rate of PbAE2 results in a more sustained gene silencing effect when compared to the more labile PbAE1 polymer. Adapted with permission from reference [170].

44


In a comparable approach, which is explained in more detail in chapter 3, our group is also involved

in the design of biodegradable siRNA impregnated gel beads (microgels) which elicit time‐controlled

release of the incorporated siRNA governed by the hydrolysis of the hydrogel crosslinks.[172]

Depending on the hydrogel characteristics, the siRNA release time can be tailored from hours over

days to several weeks. As such gels can be taken up by target cells they could serve as intracellular

siRNA depots slowly disintegrating and releasing the entrapped siRNA.

Due to the high gene silencing potential of siRNAs, only a small number of active siRNAs per cell

(~several hundred) are needed for activity. Veldhoen et al. provided a detailed quantitative analysis

of cellularly internalised siRNA through a liquid hybridization protocol.[173] Combining these data with

the observed RNAi effect revealed that for a half maximal silencing, ~104 siRNA molecules were

necessary in case of a cell penetrating peptide (CPP) assisted transfection and only ~300 in the case

of transfection with lipofectamine™2000.[173,174] This indicates that the number of siRNA molecules

that becomes available for biological activity may depend strongly on the delivery agent. Data on the

minimal number of siRNA molecules needed to trigger a sufficient RNAi effect and knowledge on the

kinetics of the RNAi effect in relation to the intracellular siRNA concentration are very important

when attempting to achieve a long‐term maximal inhibitory effect.

We emphasised that a number of crucial parameters need to be considered for an understanding of

the RNAi gene silencing kinetics. Similarly, an optimal controlled siRNA release depends on both the

cell type and target gene. For instance, it should be clear that the optimal siRNA release profile to

achieve a more sustained gene silencing will differ significantly for slowly dividing, fast dividing cells

and non‐dividing cells. On the other hand, also the expression level of the target gene will greatly

influence the amount of siRNA needed inside the cell cytoplasm for maximal gene silencing.

COMBINATORIAL APPROACH IN RNAi THERAPY

Because siRNAs/shRNAs can be developed to target viral transcripts in a sequence‐specific manner,

RNAi could prove to be the next best thing to block viral replication.[175] Unfortunately, many reports

have mentioned drug resistance through induced viral mutagenesis under pressure of RNAi mono‐

therapy.[155,175‐177] The perplexing viral genetic flexibility could also entail the emergence of

genetically encoded viral RNAi suppressors.[178] Despite the great potential of RNAi for antiviral

therapy, the issues raised above complicate the achievement of a persistent viral suppression using

RNAi mono‐therapies. Even highly active anti‐retroviral therapy (HAART), where a cocktail of three or

more antiretroviral drugs is used, cannot eradicate viral replication and only delays disease

45


progression. Moreover, HAART is notorious for the induction of severe toxicity, putting up an extra

barrier for efficient therapy.[179]

In correlation with HAART, a novel RNAi approach was put forward, called combinatorial RNAi or

coRNAi.[177] This multiplex approach involves the concerted action of different siRNA/shRNA

effectors, directed against multiple viral sequences. Although the majority of antiviral RNAi studies

focus on the targeting of viral transcripts, some studies have also been devoted to suppressing the

expression of cellular genes that are of key importance in the viral life cycle.

The primary goal of coRNAi is therefore to minimize the emergence of viral escape mutants in order

to maintain a long‐term antiviral effect. To give an example, it was shown for HIV‐1 that in response

to a single shRNA inhibitor, mutations inside and even outside the RNAi target sequence provided an

escape route for the virus eventually nullifying the antiviral effect.[180] A combinatorial strategy

applying multiple shRNAs expressed from a single lentiviral vector and targeting a different region in

the HIV‐1 genome, increased the viral inhibitory activity. Moreover, it seemed that a double

expression vector also provided a more durable viral inhibition. Comparable observations are

presented in the literature for Coxsackievirus B3 [181] and SARS‐associated coronavirus, where a clear

synergistic antiviral effect was obtained through combination of siRNAs targeting different functional

genes in the coronaviral genome.[182] Currently, the concept of coRNAi also includes combining RNAi

triggers with non‐RNAi based suppressors of gene expression [183] or even anti‐viral proteins.[184] For

instance, a multiplex strategy resulting in long‐term HIV‐1 inhibition in primary cells was reported

using a triple combination of an anti‐HIV shRNA, a hammerhead ribozyme targeted against the HIV

co‐receptor chemokine receptor 5 (CCR5) and a RNA decoy of HIV‐TAR.[183]

It would be beyond the scope of this review to discuss the plethora of reports available in the

literature with regard to combinatorial RNAi. Recently, an excellent review solely devoted to this

topic has been published by Mark Kay’s group.[177] We would like to refer the reader to this review

(and references herein) for a comprehensive overview of the different strategies already explored for

coRNAi therapy.

In the latter paper, the authors focus mainly on the antiviral potential of coRNAi. Nevertheless, RNAi

can be regarded as a broad therapeutic platform that can additionally be deployed in other

challenging‐to‐treat human pathologies, such as metabolic disorders and cancer, where a

combinatorial approach could be a therapeutic asset in the context of achieving long‐term silencing.

Recent reports established a synergistic effect of an RNAi trigger especially when combined with

conventional low molecular weight chemotherapeutics. Takei et al. employed an effective siRNA

against midkine (MK) together with low (and essentially non‐toxic) doses of paclitaxel in human

46


prostate cancer xenografts and found that paclitaxel significantly augmented the antitumour effect

of MK‐siRNA.[185] In this way, tumour growth inhibition could be maintained for several weeks

without the need to switch to higher (and most likely toxic) doses of paclitaxel. Unfortunately, the in

vivo administration of the siRNA‐atelocollagen formulation used in that report was restricted to

intratumoural injection. The same synergistic phenomenon, however, was found for an intravenously

administrated targeted nanoparticle formulation of anti‐EGFR siRNA, in combination with

intraperitoneal cisplatin, in a human lung cancer xenograft model. In contrast to mono‐therapy with

targeted nanoparticles containing anti‐EGFR siRNA, which only partially reduced tumour growth, the

combination of the targeted siRNA‐nanoparticles with cisplatin completely inhibited tumour

proliferation for ~1 week.[186] A third and earlier report describes the packaging of siRNA into neutral

liposomal vesicles for the targeting of the tyrosine kinase receptor EphA2 oncogene that is

overexpressed in ovarian cancer. When liposomal anti‐EphA2 siRNA was administered in addition to

paclitaxel, a significant reduction in tumor growth was observed when compared to paclitaxel in

combination with non‐silencing siRNA.[187] The same neutral liposomal formulation was applied to

target focal adhesion kinase (FAK) and again a synergistic effect could be obtained when combining

therapeutic siRNA with conventional chemotherapeutics such as docetaxel and cisplatin.[188]

Altogether, the studies cited in this section exemplify the tremendous potential of coRNAi in the

battle against various challenging therapeutic targets in human disease. A particular advantage when

combining different RNAi effectors (with or without alternative therapeutics) is the maximization of

the silencing effect, both acute and long‐term. Needless to say that the safety concerns which have

been raised for RNAi monotherapy also hold true for coRNAi. Multiplexing several siRNA/shRNA

triggers could even augment the risks inherently linked to RNAi, such as off‐targeting,

immunostimulation and competition with endogenous miRNAs. Following coadministration of

different types of RNAi drugs, they can even compete with each other for the (limited) amount of

cellular RNAi proteins, thereby diminishing each other’s efficacy.[60,189] On the other hand, combining

RNAi effectors with other non‐RNAi based therapeutics may allow lowering the dosing of both drugs.

In this way, coRNAi can still profit from the synergistic effect along with the benefit of reduced

toxicity.

To date, the concept of coRNAi even has expanded (paradoxically) to single‐molecule approaches,

where distinct functions are synergetically combined in the same molecule. An example of a single‐

molecule combinatorial approach was recently published by Poeck et al. in Nature Medicine, where

they used 5’‐triphosphate siRNAs as a potential therapeutic to combat melanoma.[190] The siRNAs

were targeted against Bcl‐2, a known key tumor survival factor involved in the regulation of

apoptosis. The triphosphate moiety is an activator of RIG‐1, which then leads to the stimulation of

47


the innate immune system and cytokine production. The synergy of sequence‐specific Bcl‐2 silencing

and RIG‐1 activation induces tumour apoptosis and tumour growth inhibition, and is a perfect

example of how the immune‐stimulating potential of small interfering RNA may have its therapeutic

benefit.[191] Another original approach is presented by Hossbach et al., who demonstrated the

simultaneous knockdown of two distinct genes with a single asymmetric ‘double‐guide’ siRNA, where

both strands can trigger efficient RNAi on separate targets.[192]

CONCLUSIONS

Advances in biotechnology have led to the emergence of many macromolecular drugs, such as

peptides, proteins and nucleic acids. Researchers are trying ever since to improve the therapeutic

outcome of this new spectrum of drugs using different strategies, related to the drug itself or the

formulation needed to deliver the drug in vivo. RNA interference has the advantage of being able to

profit from lessons learned during preclinical development of antisense oligonucleotides (AONs) and

ribozymes.[193] After several decades of research on AONs, leading experts of the industrial and

academic laboratories still have to acknowledge that inventive engineering strategies to improve the

design of siRNAs/shRNAs and to optimize their in vivo delivery are required to turn the therapeutic

promise of RNAi into clinical reality.

The clinical application of siRNA will call for repeated (often intravenous) administration, so it is

desirable to aim for a durable effect to lengthen treatment intervals. Direct application of synthetic

siRNAs has the disadvantage that the RNAi effect is transient, mainly due to intracellular dilution of

active siRNA dependent on the cell division rate or intracellular siRNA degradation. To maintain a

sufficient silencing of the target gene expression over a prolonged period of time, different strategies

have been proposed. Paying proper attention to their practical implication should eventually lead to

a maximal RNAi effect, both in terms of magnitude and duration. SiRNA duplexes can be designed

and modified to optimally exploit the endogenous RNAi pathway in order to increase their gene

silencing potency. Pursuing optimal siRNA design will lead to siRNA drug candidates with lower IC50

values which should enable effective medical treatment at lower doses. Incorporation of chemically

modified nucleotides into the siRNA sequence has proven to enhance their half‐life in the

bloodstream by protecting them against nuclease activity. However, it is conceivable that the

application of naked (i.e. unformulated or non‐conjugated) siRNAs will be mainly limited to confined

target sites after local administration, such as the eye or the respiratory tract.[189,194,195]

Advances in materials science increased chances of designing new nucleic acid delivery concepts. The

nano‐revolution led to the development of intelligent nanodevices which should enable

48


extrapolating drug delivery from the laboratory to a real in vivo situation. Multifunctional delivery

vehicles and siRNA conjugation strategies offer many advantages for the systemic application of

siRNA as they are usually equipped with targeting ligands and carry stabilizing hydrophilic polymers

to avoid aggregation in the bloodstream and to prevent non‐specific uptake by the reticulo‐

endothelial cells.[196] Moreover, delivery vehicles are often modified to improve the intracellular

trafficking (e.g. endosomal escape) which can lead to increased intracellular bioavailability. Including

all these factors in carrier design should enhance the percentage of the administered siRNA dose that

reaches the target site after systemic delivery. The ultimate goal is to define the appropriate dosing

schedule for a given siRNA formulation to maintain the desired therapeutic outcome. Model

predictions can aid in recognizing the main factors that govern the duration of the siRNA effect in

order to adjust the siRNA dose and frequency of administration accordingly. This can provide

researchers with general meaningful insights on systemic siRNA delivery that may apply to a variety

of siRNA carriers.

Depending on the pathological target, a more sustained gene silencing than readily achievable with

synthetic siRNA may be advisable. For this purpose researchers seek solace in the intracellular shRNA

production from plasmid or viral vectors. Although much progress has been made in the

development of viral gene therapy vectors, there are still important safety concerns that remain

troublesome.[197] Moreover, little information is available on the adverse effects that are linked to

sustained shRNA expression in vivo.

Most likely, a safe and long‐term application of RNAi drugs will require rigorous spatiotemporal

control over the intracellular siRNA/shRNA concentrations. Smart nanodevices for controlled

intracellular siRNA delivery or smart plasmids for controllable shRNA expression may fulfil some of

the necessary requirements.

49


REFERENCE LIST

1. Fire A., Xu S.Q., Montgomery M.K., Kostas S.A. et al., Potent and specific genetic interference by double‐stranded RNA in Caenorhabditis elegans, Nature 1998, 391: 806‐811.

2. Dorsett Y., siRNAs: applications in functional genomics and potential as therapeutics, Nature Reviews Drug Discovery 2004, 3: 318‐329.

3. Kim D.H. and Rossi J.J., Strategies for silencing human disease using RNA interference, Nat. Rev. Genet. 2007, 8: 173‐184.

4. Hamilton A.J. and Baulcombe D.C., A species of small antisense RNA in posttranscriptional gene silencing in plants, Science 1999, 286: 950‐952.

5. Napoli C., Lemieux C., and Jorgensen R., Introduction of A Chimeric Chalcone Synthase Gene Into Petunia Results in Reversible Co‐Suppression of Homologous Genes in Trans, Plant Cell 1990, 2: 279‐289.

6. Lee R.C., Feinbaum R.L., and Ambros V., The C‐Elegans Heterochronic Gene Lin‐4 Encodes Small Rnas with Antisense Complementarity to Lin‐14, Cell 1993, 75: 843‐854.


8. Elbashir S.M., Lendeckel W., and Tuschl T., RNA interference is mediated by 21‐and 22‐nucleotide RNAs, Genes & Development 2001, 15: 188‐200.

9. Birmingham A., Anderson E., Sullivan K., Reynolds A. et al., A protocol for designing siRNAs with high functionality and specificity, Nat. Protoc. 2007, 2: 2068‐2078.

10. Peek A.S. and Behlke M.A., Design of active small interfering RNAs, Current Opinion in Molecular Therapeutics 2007, 9: 110‐118.

11. de Fougerolles A., Vornlocher H.P., Maraganore J., and Lieberman J., Interfering with disease: a progress report on siRNA‐based therapeutics, Nature Reviews Drug Discovery 2007, 6: 443‐453.

12. Raab R.M. and Stephanopoulos G., Dynamics of gene silencing by RNA interference, Biotechnology and Bioengineering 2004, 88: 121‐132.

13. Behlke M.A., Progress towards in vivo use of siRNAs, Molecular Therapy 2006, 13: 644‐670. 14. Tolia N.H. and Joshua‐Tor L., Slicer and the Argonautes, Nature Chemical Biology 2007, 3: 36‐43. 15. Leuschner P.J.F., Ameres S.L., Kueng S., and Martinez J., Cleavage of the siRNA passenger strand during

RISC assembly in human cells, EMBO Reports 2006, 7: 314‐320. 16. Liu J.D., Carmell M.A., Rivas F.V., Marsden C.G. et al., Argonaute2 is the catalytic engine of mammalian

RNAi, Science 2004, 305: 1437‐1441. 17. Haley B. and Zamore P.D., Kinetic analysis of the RNAi enzyme complex, Nature Structural & Molecular

Biology 2004, 11: 599‐606. 18. Raemdonck K., Remaut K., Lucas B., Sanders N.N. et al., In Situ Analysis of Single‐Stranded and Duplex

siRNA Integrity in Living Cells, Biochemistry 2006, 45: 10614‐10623. 19. Bertrand J.R., Pottier M., Vekris A., Opolon P. et al., Comparison of antisense oligonucleotides and

siRNAs in cell culture and in vivo, Biochemical and Biophysical Research Communications 2002, 296: 1000‐1004.

20. Kim D.H., Behlke M.A., Rose S.D., Chang M.S. et al., Synthetic dsRNA Dicer substrates enhance RNAi potency and efficacy, Nature Biotechnology 2005, 23: 222‐226.

21. Amarzguioui M., Rossi J.J., and Kim D., Approaches for chemically synthesized siRNA and vector‐mediated RNAi, FEBS Letters 2005, 579: 5974‐5981.

22. Hefner E., Clark K., Whitman C., Behlke M.A. et al., Increased potency and longevity of gene silencing using validated Dicer substrates, J. Biomol. Tech. 2008, 19: 231‐237.

23. Lee Y.S., Nakahara K., Pham J.W., Kim K. et al., Distinct roles for Drosophila Dicer‐1 and Dicer‐2 in the siRNA/miRNA silencing pathways, Cell 2004, 117: 69‐81.

24. Winston W.M., Molodowitch C., and Hunter C.P., Systemic RNAi in C‐elegans requires the putative transmembrane protein SID‐1, Science 2002, 295: 2456‐2459.

25. Feinberg E.H. and Hunter C.P., Transport of dsRNA into cells by the transmembrane protein SID‐1, Science 2003, 301: 1545‐1547.

26. Oliveira S., Storm G., and Schiffelers R.M., Targeted delivery of siRNA, Journal of Biomedicine and Biotechnology 2006, 63675.

27. Duxbury M.S., Ashley S.W., and Whang E.E., RNA interference: A mammalian SID‐1 homologue enhances siRNA uptake and gene silencing efficacy in human cells, Biochemical and Biophysical Research Communications 2005, 331: 459‐463.

50


28. Wolfrum C., Shi S., Jayaprakash K.N., Jayaraman M. et al., Mechanisms and optimization of in vivo delivery of lipophilic siRNAs, Nat. Biotechnol. 2007, 25: 1149‐1157.

29. Baulcombe D.C., Amplified silencing, Science 2007, 315: 199‐200. 30. Sijen T., Steiner F.A., Thijssen K.L., and Plasterk R.H.A., Secondary siRNAs result from unprimed RNA

synthesis and form a distinct class, Science 2007, 315: 244‐247. 31. Pak J. and Fire A., Distinct populations of primary and secondary effectors during RNAi in C. elegans,

Science 2007, 315: 241‐244. 32. Chiu Y.L. and Rana T.M., RNAi in human cells: Basic structural and functional features of small interfering

RNA, Molecular Cell 2002, 10: 549‐561. 33. Stein P., Svoboda P., Anger M., and Schultz R.M., RNAi: Mammalian oocytes do it without RNA‐

dependent RNA polymerase, RNA 2003, 9: 187‐192. 34. Cullen B.R., Induction of stable RNA interference in mammalian cells, Gene Therapy 2006, 13: 503‐508. 35. Brummelkamp T.R., Bernards R., and Agami R., A system for stable expression of short interfering RNAs

in mammalian cells, Science 2002, 296: 550‐553. 36. Barnes M.R., Deharo S., Grocock R.J., Brown J.R. et al., The micro RNA target paradigm: a fundamental

and polymorphic control layer of cellular expression, Expert Opin. Biol. Ther. 2007, 7: 1387‐1399. 37. Bartel D.P., MicroRNAs: Genomics, biogenesis, mechanism, and function, Cell 2004, 116: 281‐297. 38. Zeng Y., Yi R., and Cullen B.R., Recognition and cleavage of primary microRNA precursors by the nuclear

processing enzyme Drosha, EMBO Journal 2005, 24: 138‐148. 39. Han J., Lee Y., Yeom K.H., Nam J.W. et al., Molecular basis for the recognition of primary microRNAs by

the Drosha‐DGCR8 complex, Cell 2006, 125: 887‐901. 40. Yeom K.H., Lee Y., Han J., Suh M.R. et al., Characterization of DGCR8/Pasha, the essential cofactor for

Drosha in primary miRNA processing, Nucleic Acids Res. 2006, 34: 4622‐4629. 41. Lund E., Guttinger S., Calado A., Dahlberg J.E. et al., Nuclear export of microRNA precursors, Science

2004, 303: 95‐98. 42. Kim V.N., MicroRNA biogenesis: coordinated cropping and dicing, Nat. Rev. Mol. Cell Biol. 2005, 6: 376‐

385. 43. Liu J.D., Valencia‐Sanchez M.A., Hannon G.J., and Parker R., MicroRNA‐dependent localization of

targeted mRNAs to mammalian P‐bodies, Nature Cell Biology 2005, 7: 719‐723. 44. Grimm D., Small silencing RNAs: state‐of‐the‐art, Adv. Drug Deliv. Rev. 2009, 61: 672‐703. 45. Yi R., Qin Y., Macara I.G., and Cullen B.R., Exportin‐5 mediates the nuclear export of pre‐microRNAs and

short hairpin RNAs, Genes & Development 2003, 17: 3011‐3016. 46. Siolas D., Lerner C., Burchard J., Ge W. et al., Synthetic shRNAs as potent RNAi triggers, Nature

Biotechnology 2005, 23: 227‐231. 47. Li L.M., Lin X.Y., Khvorova A., Fesik S.W. et al., Defining the optimal parameters for hairpin‐based

knockdown constructs, RNA 2007, 13: 1765‐1774. 48. Vlassov A.V., Korba B., Farrar K., Mukerjee S. et al., shRNAs targeting hepatitis C: Effects of sequence and

structural features, and comparison with siRNA, Oligonucleotides 2007, 17: 223‐236. 49. Fewell G.D. and Schmitt K., Vector‐based RNAi approaches for stable, inducible and genome‐wide

screens, Drug discovery today 2006, 11: 975‐982. 50. Silva J.M., Li M.Z., Chang K., Ge W. et al., Second‐generation shRNA libraries covering the mouse and

human genomes, Nature Genetics 2005, 37: 1281‐1288. 51. Vorhies J.S. and Nemunaitis J., Nonviral delivery vehicles for use in short hairpin RNA‐based cancer

therapies, Expert Review of Anticancer Therapy 2007, 7: 373‐382. 52. Song S.H., Lu Y.Q., Choi Y.K., Han Y.N. et al., DNA‐dependent PK inhibits adeno‐associated virus DNA

integration, Proceedings of the National Academy of Sciences of the United States of America 2004, 101: 2112‐2116.

53. Chen C.C., Ko T.M., Ma H.I., Wu H.L. et al., Long‐term inhibition of hepatitis B virus in transgenic mice by double‐stranded adeno‐associated virus 8‐delivered short hairpin RNA, Gene Therapy 2007, 14: 11‐19.

54. Bartlett D.W. and Davis M.E., Insights into the kinetics of siRNA‐mediated gene silencing from live‐cell and live‐animal bioluminescent imaging, Nucleic Acids Research 2006, 34: 322‐333.

55. Mutskov V. and Felsenfeld G., Silencing of transgene transcription precedes methylation of promoter DNA and histone H3 lysine 9, EMBO Journal 2004, 23: 138‐149.

56. Grimm D., Streetz K.L., Jopling C.L., Storm T.A. et al., Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways, Nature 2006, 441: 537‐541.

57. Wiznerowicz M., Szulc J., and Trono D., Tuning silence: conditional systems for RNA interference, Nat. Methods 2006, 3: 682‐688.

51


58. An D.S., Qin F.X.F., Auyeung V.C., Mao S.H. et al., Optimization and functional effects of stable short hairpin RNA expression in primary human lymphocytes via lentiviral vectors, Molecular Therapy 2006, 14: 494‐504.

59. Narvaiza I., Aparicio O., Vera M., Razquin N. et al., Effect of adenovirus‐mediated RNA interference on endogenous microRNAs in a mouse model of multidrug resistance protein 2 gene silencing, J. Virol. 2006, 80: 12236‐12247.

60. Castanotto D., Sakurai K., Lingeman R., Li H. et al., Combinatorial delivery of small interfering RNAs reduces RNAi efficacy by selective incorporation into RISC, Nucleic Acids Res. 2007, 35: 5154‐5164.

61. Koller E., Propp S., Murray H., Lima W. et al., Competition for RISC binding predicts in vitro potency of siRNA, Nucleic Acids Research 2006, 34: 4467‐4476.

62. Dykxhoorn D.M. and Lieberman J., Knocking down disease with siRNAs, Cell 2006, 126: 231‐235. 63. Snove O., Jr. and Rossi J.J., Toxicity in mice expressing short hairpin RNAs gives new insight into RNAi,

Genome Biol. 2006, 7: 231. 64. John M., Constien R., Akinc A., Goldberg M. et al., Effective RNAi‐mediated gene silencing without

interruption of the endogenous microRNA pathway, Nature 2007, 449: 745‐747. 65. Kohn D.B., Sadelain M., and Glorioso J.C., Occurrence of leukaemia following gene therapy of X‐linked

SCID, Nature Reviews Cancer 2003, 3: 477‐488. 66. Cavazzana‐Calvo M. and Fischer A., Gene therapy for severe combined immunodeficiency: are we there

yet?, Journal of Clinical Investigation 2007, 117: 1456‐1465. 67. Vandenbroucke R.E., Lucas B., Demeester J., De Smedt S.C. et al., Nuclear accumulation of plasmid DNA

can be enhanced by non‐selective gating of the nuclear pore, Nucleic Acids Res. 2007, 35: e86. 68. Munkonge F.M., Dean D.A., Hillery E., Griesenbach U. et al., Emerging significance of plasmid DNA

nuclear import in gene therapy, Advanced Drug Delivery Reviews 2003, 55: 749‐760. 69. Haupenthal J., Baehr C., Kiermayer S., Zeuzem S. et al., Inhibition of RNAse A family enzymes prevents

degradation and loss of silencing activity of siRNAs in serum, Biochemical Pharmacology 2006, 71: 702‐710.

70. Turner J.J., Jones S.W., Moschos S.A., Lindsay M.A. et al., MALDI‐TOF mass spectral analysis of siRNA degradation in serum confirms an RNAse A‐like activity, Mol. Biosyst. 2007, 3: 43‐50.

71. Haupenthal J., Baehr C., Zeuzem S., and Piiper A., RNAse A‐like enzymes in serum inhibit the anti‐neoplastic activity of siRNA targeting polo‐like kinase 1, Int. J. Cancer 2007, 121: 206‐210.

72. Behlke M.A., Chemical modification of siRNAs for in vivo use, Oligonucleotides. 2008, 18: 305‐319. 73. Kennedy S., Wang D., and Ruvkun G., A conserved siRNA‐degrading RNase negatively regulates RNA

interference in C. elegans, Nature 2004, 427: 645‐649. 74. Zou Y., Tiller P., Chen I.W., Beverly M. et al., Metabolite identification of small interfering RNA duplex by

high‐resolution accurate mass spectrometry, Rapid Commun. Mass Spectrom. 2008, 22: 1871‐1881. 75. Choung S., Chemical modification of siRNAs to improve serum stability without loss of efficacy,

Biochemical and Biophysical Research Communications 2006, 342: 919‐927. 76. Morrissey D.V., Blanchard K., Shaw L., Jensen K. et al., Activity of stabilized short interfering RNA in a

mouse model of hepatitis B virus replication, Hepatology 2005, 41: 1349‐1356. 77. Corey D.R., Chemical modification: the key to clinical application of RNA interference?, Journal of Clinical

Investigation 2007, 117: 3615‐3622. 78. Watts J.K., Deleavey G.F., and Damha M.J., Chemically modified siRNA: tools and applications, Drug

Discov. Today 2008, 13: 842‐855. 79. Allerson C.R., Sioufi N., Jarres R., Prakash T.P. et al., Fully 2 '‐modified oligonucleotide duplexes with

improved in vitro potency and stability compared to unmodified small interfering RNA, Journal of Medicinal Chemistry 2005, 48: 901‐904.

80. Liao H.T. and Wang J.H., Biomembrane‐permeable and ribonuclease‐resistant siRNA with enhanced activity, Oligonucleotides 2005, 15: 196‐205.

81. Layzer J.M., McCaffrey A.P., Tanner A.K., Huang Z. et al., In vivo activity of nuclease‐resistant siRNAs, RNA 2004, 10: 766‐771.

82. Dykxhoorn D.M., Palliser D., and Lieberman J., The silent treatment: siRNAs as small molecule drugs, Gene Therapy 2006, 13: 541‐552.

83. Novobrantseva T.I., Akinc A., Borodovsky A., and de Fougerolles A., Delivering silence: Advancements in developing siRNA therapeutics, Current Opinion in Drug Discovery & Development 2008, 11: 217‐224.

84. De Paula D., Bentley M.V., and Mahato R.I., Hydrophobization and bioconjugation for enhanced siRNA delivery and targeting, RNA. 2007, 13: 431‐456.

52


85. de Fougerolles A.R., Delivery vehicles for small interfering RNA in vivo, Hum. Gene Ther. 2008, 19: 125‐132.

86. Song E.W., Zhu P.C., Lee S.K., Chowdhury D. et al., Antibody mediated in vivo delivery of small interfering RNAs via cell‐surface receptors, Nature Biotechnology 2005, 23: 709‐717.

87. Soutschek J., Akinc A., Bramlage B., Charisse K. et al., Therapeutic silencing of an endogenous gene by systemic administration of modified siRNAs, Nature 2004, 432: 173‐178.

88. Nishina K., Unno T., Uno Y., Kubodera T. et al., Efficient in vivo delivery of siRNA to the liver by conjugation of alpha‐tocopherol, Molecular Therapy 2008, 16: 734‐740.

89. Kumar P., Wu H.Q., McBride J.L., Jung K.E. et al., Transvascular delivery of small interfering RNA to the central nervous system, Nature 2007, 448: 39‐43.

90. Rozema D.B., Lewis D.L., Wakefield D.H., Wong S.C. et al., Dynamic PolyConjugates for targeted in vivo delivery of siRNA to hepatocytes, Proceedings of the National Academy of Sciences of the United States of America 2007, 104: 12982‐12987.

91. Chu T.C., Twu K.Y., Ellington A.D., and Levy M., Aptamer mediated siRNA delivery, Nucleic Acids Res. 2006, 34: e73.

92. McNamara J.O., Andrechek E.R., Wang Y., Viles K.D. et al., Cell type‐specific delivery of siRNAs with aptamer‐siRNA chimeras, Nat. Biotechnol. 2006, 24: 1005‐1015.

93. Eguchi A., Meade B.R., Chang Y.C., Fredrickson C.T. et al., Efficient siRNA delivery into primary cells by a peptide transduction domain‐dsRNA binding domain fusion protein, Nat. Biotechnol. 2009, doi: 10.1038/nbt.1541.

94. Juliano R., Alam M.R., Dixit V., and Kang H., Mechanisms and strategies for effective delivery of antisense and siRNA oligonucleotides, Nucleic Acids Res. 2008, 36: 4158‐4171.

95. Sato A., Choi S.W., Hirai M., Yamayoshi A. et al., Polymer brush‐stabilized polyplex for a siRNA carrier with long circulatory half‐life, Journal of controlled release 2007, 122: 209‐216.

96. Urban‐Klein B., RNAi‐mediated gene‐targeting through systemic application of polyethyleneimine (PEI)‐complexed siRNA in vivo, Gene Therapy 2005, 12: 461‐466.

97. Buyens K., Lucas B., Raemdonck K., Braeckmans K. et al., A fast and sensitive method for measuring the integrity of siRNA‐carrier complexes in full human serum, J. Control Release 2007, 126: 67‐76.

98. Howard K.A. and Kjems J., Polycation‐based nanoparticle delivery for improved RNA interference therapeutics, Expert Opinion on Biological Therapy 2007, 7: 1811‐1822.

99. Howard K.A., Rahbek U.L., Liu X.D., Damgaard C.K. et al., RNA interference in vitro and in vivo using a chitosan/siRNA nanoparticle system, Molecular Therapy 2006, 14: 476‐484.

100. Takeshita F., Minakuchi Y., Nagahara S., Honma K. et al., Efficient delivery of small interfering RNA to bone‐metastatic tumors by using atelocollagen in vivo, Proceedings of the National Academy of Sciences of the United States of America 2005, 102: 12177‐12182.

101. Morrissey D.V., Lockridge J.A., Shaw L., Blanchard K. et al., Potent and persistent in vivo anti‐HBV activity of chemically modified siRNAs, Nature Biotechnology 2005, 23: 1002‐1007.

102. Zimmermann T.S., Lee A.C., Akinc A., Bramlage B. et al., RNAi‐mediated gene silencing in non‐human primates, Nature 2006, 441: 111‐114.

103. Hu‐Lieskovan S., Heidel J.D., Bartlett D.W., Davis M.E. et al., Sequence‐specific knockdown of EWS‐FLI1 by targeted, nonviral delivery of small interfering RNA inhibits tumor growth in a murine model of metastatic Ewing's sarcoma, Cancer Research 2005, 65: 8984‐8992.

104. Heidel J.D., Yu Z.P., Liu J.Y.C., Rele S.M. et al., Administration in non‐human primates of escalating intravenous doses of targeted nanoparticles containing ribonucleotide reductase subunit M2 siRNA, Proceedings of the National Academy of Sciences of the United States of America 2007, 104: 5715‐5721.

105. Davis M.E., The First Targeted Delivery of siRNA in Humans via a Self‐Assembling, Cyclodextrin Polymer‐Based Nanoparticle: From Concept to Clinic, Mol. Pharm. 2009, 6: 659‐668.

106. Czauderna F., Structural variations and stabilising modifications of synthetic siRNAs in mammalian cells, Nucleic Acids Research 2003, 31: 2705‐2716.

107. Chiu Y.L. and Rana T.M., siRNA function in RNAi: a chemical modification analysis, RNA. 2003, 9: 1034‐1048.

108. Dowler T., Bergeron D., Tedeschi A.L., Paquet L. et al., Improvements in siRNA properties mediated by 2'‐deoxy‐2'‐fluoro‐beta‐D‐arabinonucleic acid (FANA), Nucleic Acids Res. 2006, 34: 1669‐1675.

109. Volkov A.A., Kruglova N.S., Meschaninova M.I., Venyaminova A.G. et al., Selective protection of nuclease‐sensitive sites in siRNA prolongs silencing effect, Oligonucleotides. 2009, 19: 191‐202.

110. Mantei A., Rutz S., Janke M., Kirchhoff D. et al., siRNA stabilization prolongs gene knockdown in primary T lymphocytes, Eur. J. Immunol. 2008, 38: 2616‐2625.

53


111. Takabatake Y., Isaka Y., Mizui M., Kawachi H. et al., Chemically modified siRNA prolonged RNA interference in renal disease, Biochem. Biophys. Res. Commun. 2007, 363: 432‐437.

112. Tagami T., Barichello J.M., Kikuchi H., Ishida T. et al., The gene‐silencing effect of siRNA in cationic lipoplexes is enhanced by incorporating pDNA in the complex, Int. J. Pharm. 2007, 333: 62‐69.

113. Fisher M., Abramov M., Van A.A., Xu D. et al., Inhibition of MDR1 expression with altritol‐modified siRNAs, Nucleic Acids Res. 2007, 35: 1064‐1074.

114. Song E.W., Lee S.K., Dykxhoorn D.M., Novina C. et al., Sustained small interfering RNA‐mediated human immunodeficiency virus type 1 inhibition in primary macrophages, Journal of Virology 2003, 77: 7174‐7181.

115. Maliyekkel A., Davis B.M., and Roninson I.B., Cell cycle arrest drastically extends the duration of gene silencing after transient expression of short hairpin RNA, Cell Cycle 2006, 5: 2390‐2395.

116. Omi K., Tokunaga K., and Hohjoh H., Long‐lasting RNAi activity in mammalian neurons, FEBS Letters 2004, 558: 89‐95.

117. Kakutani K., Nishida K., Uno K., Takada T. et al., Prolonged down regulation of specific gene expression in nucleus pulposus cell mediated by RNA interference in vitro, J. Orthop. Res. 2006, 24: 1271‐1278.

118. Bartlett D.W. and Davis M.E., Effect of siRNA nuclease stability on the in vitro and in vivo kinetics of siRNA‐mediated gene silencing, Biotechnol. Bioeng. 2007, 97: 909‐921.

119. Marques J.T. and Williams B.R., Activation of the mammalian immune system by siRNAs, Nat. Biotechnol. 2005, 23: 1399‐1405.

120. Snove O., Jr. and Rossi J.J., Chemical modifications rescue off‐target effects of RNAi, ACS Chem. Biol. 2006, 1: 274‐276.

121. Jackson A.L., Burchard J., Leake D., Reynolds A. et al., Position‐specific chemical modification of siRNAs reduces "off‐target" transcript silencing, RNA. 2006, 12: 1197‐1205.

122. Judge A. and MacLachlan I., Overcoming the innate immune response to small interfering RNA, Human Gene Therapy 2008, 19: 111‐124.

123. Judge A.D., Sood V., Shaw J.R., Fang D. et al., Sequence‐dependent stimulation of the mammalian innate immune response by synthetic siRNA, Nature Biotechnology 2005, 23: 457‐462.

124. Hornung V., Guenthner‐Biller M., Bourquin C., Ablasser A. et al., Sequence‐specific potent induction of IFN‐alpha by short interfering RNA in plasmacytoid dendritic cells through TLR7, Nature Medicine 2005, 11: 263‐270.

125. Marques J.T., Devosse T., Wang D., Zamanian‐Daryoush M. et al., A structural basis for discriminating between self and nonself double‐stranded RNAs in mammalian cells, Nature Biotechnology 2006, 24: 559‐565.

126. Judge A.D., Bola G., Lee A.C.H., and MacLachlan I., Design of noninflammatory synthetic siRNA mediating potent gene silencing in vivo, Molecular Therapy 2006, 13: 494‐505.

127. Kleinman M.E., Yamada K., Takeda A., Chandrasekaran V. et al., Sequence‐ and target‐independent angiogenesis suppression by siRNA via TLR3, Nature 2008, 452: 591‐597.

128. Cho W.G., Albuquerque R.J., Kleinman M.E., Tarallo V. et al., Small interfering RNA‐induced TLR3 activation inhibits blood and lymphatic vessel growth, Proc. Natl. Acad. Sci. U. S. A 2009, 106: 7137‐7142.

129. Robbins M., Judge A., and MacLachlan I., siRNA and innate immunity, Oligonucleotides. 2009, 19: 89‐102. 130. Zeng Y., RNA interference in human cells is restricted to the cytoplasm, RNA 2002, 8: 855‐860. 131. Robb G.B., Brown K.M., Khurana J., and Rana T.M., Specific and potent RNAi in the nucleus of human

cells, Nat. Struct. Mol. Biol. 2005, 12: 133‐137. 132. Gurney T., Jr. and Eliceiri G.L., Intracellular distribution of low molecular weight RNA species in HeLa

cells, J. Cell Biol. 1980, 87: 398‐403. 133. Matera A.G. and Ward D.C., Nucleoplasmic organization of small nuclear ribonucleoproteins in cultured

human cells, J. Cell Biol. 1993, 121: 715‐727. 134. Wassarman D.A. and Steitz J.A., Structural analyses of the 7SK ribonucleoprotein (RNP), the most

abundant human small RNP of unknown function, Mol. Cell Biol. 1991, 11: 3432‐3445. 135. Ohrt T., Mutze J., Staroske W., Weinmann L. et al., Fluorescence correlation spectroscopy and

fluorescence cross‐correlation spectroscopy reveal the cytoplasmic origination of loaded nuclear RISC in vivo in human cells, Nucleic Acids Res. 2008, 36: 6439‐6449.

136. Matzke M.A., Primig M., Trnovsky J., and Matzke A.J., Reversible methylation and inactivation of marker genes in sequentially transformed tobacco plants, EMBO J. 1989, 8: 643‐649.

137. Hamilton A., Voinnet O., Chappell L., and Baulcombe D., Two classes of short interfering RNA in RNA silencing, EMBO J. 2002, 21: 4671‐4679.

54


138. Mette M.F., Aufsatz W., van der W.J., Matzke M.A. et al., Transcriptional silencing and promoter methylation triggered by double‐stranded RNA, EMBO J. 2000, 19: 5194‐5201.

139. Wassenegger M., Heimes S., Riedel L., and Sanger H.L., RNA‐directed de novo methylation of genomic sequences in plants, Cell 1994, 76: 567‐576.

140. Zilberman D., Cao X., and Jacobsen S.E., ARGONAUTE4 control of locus‐specific siRNA accumulation and DNA and histone methylation, Science 2003, 299: 716‐719.

141. Zilberman D., Cao X., Johansen L.K., Xie Z. et al., Role of Arabidopsis ARGONAUTE4 in RNA‐directed DNA methylation triggered by inverted repeats, Curr. Biol. 2004, 14: 1214‐1220.

142. Castanotto D., Tommasi S., Li M., Li H. et al., Short hairpin RNA‐directed cytosine (CpG) methylation of the RASSF1A gene promoter in HeLa cells, Mol. Ther. 2005, 12: 179‐183.

143. Kawasaki H. and Taira K., Induction of DNA methylation and gene silencing by short interfering RNAs in human cells, Nature 2004, 431: 211‐217.

144. Morris K.V., Chan S.W., Jacobsen S.E., and Looney D.J., Small interfering RNA‐induced transcriptional gene silencing in human cells, Science 2004, 305: 1289‐1292.

145. Weinberg M.S., Villeneuve L.M., Ehsani A., Amarzguioui M. et al., The antisense strand of small interfering RNAs directs histone methylation and transcriptional gene silencing in human cells, RNA. 2006, 12: 256‐262.

146. Strunnikova M., Schagdarsurengin U., Kehlen A., Garbe J.C. et al., Chromatin inactivation precedes de novo DNA methylation during the progressive epigenetic silencing of the RASSF1A promoter, Mol. Cell Biol. 2005, 25: 3923‐3933.

147. Janowski B.A., Huffman K.E., Schwartz J.C., Ram R. et al., Inhibiting gene expression at transcription start sites in chromosomal DNA with antigene RNAs, Nat. Chem. Biol. 2005, 1: 216‐222.

148. Ting A.H., Schuebel K.E., Herman J.G., and Baylin S.B., Short double‐stranded RNA induces transcriptional gene silencing in human cancer cells in the absence of DNA methylation, Nat. Genet. 2005, 37: 906‐910.

149. Kim D.H., Villeneuve L.M., Morris K.V., and Rossi J.J., Argonaute‐1 directs siRNA‐mediated transcriptional gene silencing in human cells, Nat. Struct. Mol. Biol. 2006, 13: 793‐797.

150. Janowski B.A., Huffman K.E., Schwartz J.C., Ram R. et al., Involvement of AGO1 and AGO‐2 in mammalian transcriptional silencing, Nature Structural & Molecular Biology 2006, 13: 787‐792.

151. Kim D.H., Saetrom P., Snove O., Jr., and Rossi J.J., MicroRNA‐directed transcriptional gene silencing in mammalian cells, Proc. Natl. Acad. Sci. U. S. A 2008, 105: 16230‐16235.

152. Janowski B.A., Younger S.T., Hardy D.B., Ram R. et al., Activating gene expression in mammalian cells with promoter‐targeted duplex RNAs, Nat. Chem. Biol. 2007, 3: 166‐173.

153. Li L.C., Okino S.T., Zhao H., Pookot D. et al., Small dsRNAs induce transcriptional activation in human cells, Proc. Natl. Acad. Sci. U. S. A 2006, 103: 17337‐17342.

154. Capodici J., Kariko K., and Weissman D., Inhibition of HIV‐1 infection by small interfering RNA‐mediated RNA interference, J. Immunol. 2002, 169: 5196‐5201.

155. Boden D., Pusch O., Lee F., Tucker L. et al., Human immunodeficiency virus type 1 escape from RNA interference, J. Virol. 2003, 77: 11531‐11535.

156. Suzuki K., Shijuuku T., Fukamachi T., Zaunders J. et al., Prolonged transcriptional silencing and CpG methylation induced by siRNAs targeted to the HIV‐1 promoter region, J. RNAi Gene Silencing 2005, 1: 66‐78.

157. Robertson K.D., DNA methylation and chromatin ‐ unraveling the tangled web, Oncogene 2002, 21: 5361‐5379.

158. Sijen T., Vijn I., Rebocho A., van B.R. et al., Transcriptional and posttranscriptional gene silencing are mechanistically related, Curr. Biol. 2001, 11: 436‐440.

159. Burgers P.M. and Eckstein F., Absolute configuration of the diastereomers of adenosine 5'‐O‐(1‐thiotriphosphate): consequences for the stereochemistry of polymerization by DNA‐dependent RNA polymerase from Escherichia coli, Proc. Natl. Acad. Sci. U. S. A 1978, 75: 4798‐4800.

160. Vastenhouw N.L., Brunschwig K., Okihara K.L., Muller F. et al., Long‐term gene silencing by RNAi, Nature 2006, 442: 882‐882.

161. Hawkins P.G., Santoso S., Adams C., Anest V. et al., Promoter targeted small RNAs induce long‐term transcriptional gene silencing in human cells, Nucleic Acids Res. 2009, 37: 2984‐2995.

162. Oliveira S., van Rooy I., Kranenburg O., Storm G. et al., Fusogenic peptides enhance endosomal escape improving siRNA‐induced silencing of oncogenes, International Journal of Pharmaceutics 2007, 331: 211‐214.

55


163. Oliveira S., Fretz M.M., Hogset A., Storm G. et al., Photochemical internalization enhances silencing of epidermal growth factor receptor through improved endosomal escape of siRNA, Biochimica et Biophysica Acta‐Biomembranes 2007, 1768: 1211‐1217.

164. Bartlett D.W. and Davis M.E., Impact of tumor‐specific targeting and dosing schedule on tumor growth inhibition after intravenous administration of siRNA‐containing nanoparticles, Biotechnol. Bioeng. 2007,

165. Groenenboom M.A., Maree A.F., and Hogeweg P., The RNA silencing pathway: the bits and pieces that matter, PLoS. Comput. Biol. 2005, 1: 155‐165.

166. Bergstrom C.T., McKittrick E., and Antia R., Mathematical models of RNA silencing: unidirectional amplification limits accidental self‐directed reactions, Proc. Natl. Acad. Sci. U. S. A 2003, 100: 11511‐11516.

167. Takahashi Y., Yamaoka K., Nishikawa M., and Takakura Y., Moment analysis for kinetics of gene silencing by RNA interference, Biotechnol. Bioeng. 2006, 93: 816‐819.

168. Arciero J.C., Jackson T.L., and Kirschner D.E., A mathematical model of tumor‐immune evasion and siRNA treatment, Discrete and Continuous Dynamical Systems‐Series B 2004, 4: 39‐58.

169. Bartlett D.W., Su H., Hildebrandt I.J., Weber W.A. et al., Impact of tumor‐specific targeting on the biodistribution and efficacy of siRNA nanoparticles measured by multimodality in vivo imaging, Proc. Natl. Acad. Sci. U. S. A 2007, 104: 15549‐15554.

170. Marsden P.A., RNA interference as potential therapy‐‐not so fast, N. Engl. J. Med. 2006, 355: 953‐954. 171. Vandenbroucke R., De Geest B.G., Bonné S., and Van Haecke T., Prolonged gene silencing in hepatoma

cells and primary hepatocytes after siRNA delivery with biodegradable poly(β‐amino esters), Journal of Gene Medicine 2008, 10: 783‐794.

172. Raemdonck K., Van Thienen T.G., Vandenbroucke R.E., Sanders N.N. et al., Dextran Microgels for Time‐Controlled Delivery of siRNA, Advanced Functional Materials 2008, 18: 993‐1001.

173. Veldhoen S., Laufer S.D., Trampe A., and Restle T., Cellular delivery of small interfering RNA by a non‐covalently attached cell‐penetrating peptide: quantitative analysis of uptake and biological effect, Nucleic Acids Res. 2006, 34: 6561‐6573.

174. Mescalchin A., Detzer A., Wecke M., Overhoff M. et al., Cellular uptake and intracellular release are major obstacles to the therapeutic application of siRNA: novel options by phosphorothioate‐stimulated delivery, Expert. Opin. Biol. Ther. 2007, 7: 1531‐1538.

175. Dykxhoorn D.M. and Lieberman J., Silencing viral infection, PLoS. Med. 2006, 3: e242. 176. Boden D., Pusch O., and Ramratnam B., Overcoming HIV‐1 resistance to RNA interference, Front Biosci.

2007, 12: 3104‐3116. 177. Grimm D. and Kay M.A., Combinatorial RNAi: A winning strategy for the race against evolving targets?,

Molecular Therapy 2007, 15: 878‐888. 178. Zheng Z.M., Tang S.A., and Tao M.F., Development of resistance to RNAi in mammalian cells, Therapeutic

Oligonucleotides: Transcriptional and Translational Strategies for Silencing Gene Expression 2005, 1058: 105‐118.

179. Meadows D.C. and Gervay‐Hague J., Targeting HIV, Chemmedchem 2006, 1: 16‐29. 180. ter Brake O., Konstantinova P., Ceylan M., and Berkhout B., Silencing of HIV‐1 with RNA interference: A

multiple shRNA approach, Molecular Therapy 2006, 14: 883‐892. 181. Schubert S., Grunert H.P., Zeichhardt H., Werk D. et al., Maintaining inhibition: siRNA double expression

vectors against coxsackieviral RNAs, Journal of Molecular Biology 2005, 346: 457‐465. 182. He M.L., Zheng B.J., Chen Y., Wong K.L. et al., Kinetics and synergistic effects of siRNAs targeting

structural and replicase genes of SARS‐associated coronavirus, FEBS Lett. 2006, 580: 2414‐2420. 183. Li M.J., Kim J., Li S., Zaia J. et al., Long‐term inhibition of HIV‐1 infection in primary hematopoietic cells by

lentiviral vector delivery of a triple combination of anti‐HIV shRNA, anti‐CCR5 ribozyme, and a nucleolar‐localizing TAR decoy, Molecular Therapy 2005, 12: 900‐909.

184. Unwalla H.J., Li H.T., Bahner I., Li M.J. et al., Novel Pol II fusion promoter directs human immunodeficiency virus type 1‐inducible coexpression of a short hairpin RNA and protein, J. Virol. 2006, 80: 1863‐1873.

185. Takei Y., Kadomatsu K., Goto T., and Muramatsu T., Combinational antitumor effect of siRNA against midkine and paclitaxel on growth of human prostate cancer xenografts, Cancer 2006, 107: 864‐873.

186. Li S.D., Chen Y.C., Hackett M.J., and Huang L., Tumor‐targeted delivery of siRNA by self‐assembled nanoparticles, Molecular Therapy 2008, 16: 163‐169.

187. Landen C.N., Chavez‐Reyes A., Bucana C., Schmandt R. et al., Therapeutic EphA2 gene targeting in vivo using neutral liposomal small interfering RNA delivery, Cancer Research 2005, 65: 6910‐6918.

56


188. Halder J., Kamat A.A., Landen C.N., Han L.Y. et al., Focal adhesion kinase targeting using in vivo short interfering RNA delivery in neutral liposomes for ovarian carcinoma therapy, Clinical Cancer Research 2006, 12: 4916‐4924.

189. Bitko V., Musiyenko A., Shulyayeva O., and Barik S., Inhibition of respiratory viruses by nasally administered siRNA, Nature Medicine 2005, 11: 50‐55.

190. Poeck H., Besch R., Maihoefer C., Renn M. et al., 5'‐Triphosphate‐siRNA: turning gene silencing and Rig‐I activation against melanoma, Nat. Med. 2008, 14: 1256‐1263.

191. Schlee M., Hornug V., and Hartmann G., siRNA and isRNA: Two edges of one sword, Molecular Therapy 2006, 14: 463‐470.

192. Hossbach M., Gruber J., Osborn M., Weber K. et al., Gene silencing with siRNA duplexes composed of target‐mRNA‐complementary and partially palindromic or partially complementary single‐stranded siRNAs, RNA Biol. 2006, 3: 82‐89.

193. Corey D.R., RNA learns from antisense, Nature Chemical Biology 2007, 3: 8‐11. 194. Campochiaro P.A., Potential applications for RNAi to probe pathogenesis and develop new treatments

for ocular disorders, Gene Therapy 2006, 13: 559‐562. 195. Shen J., Samul R., Silva R.L., Akiyama H. et al., Suppression of ocular neovascularization with siRNA

targeting VEGF receptor 1, Gene Therapy 2006, 13: 225‐234. 196. Remaut K., Sanders N.N., De Geest B.G., Braeckmans K. et al., Nucleic acid delivery: Where material

sciences and bio‐sciences meet, Materials Science & Engineering R‐Reports 2007, 58: 117‐161. 197. Thomas C.E., Ehrhardt A., and Kay M.A., Progress and problems with the use of viral vectors for gene

therapy, Nature Reviews Genetics 2003, 4: 346‐358.

57


58

in situ analysis of single‐stranded and duplex siRNA integrity in living cells | CHAPTER 2

59

CHAPTER 2 | In situ analysis of single‐stranded and duplex siRNA integrity in

living cells


Koen Raemdonck1, Katrien Remaut1, Bart Lucas1, Niek N. Sanders1,2, Joseph Demeester1 and Stefaan

C. De Smedt1, Biochemistry 2006, 45: 10614‐10623.


Sciences, Ghent University, Harelbekestraat 72, B‐9000 Ghent, Belgium.

2Laboratory of Gene Therapy, Department of Nutrition, Genetics and Ethology, Faculty of Veterinary Medicine, Ghent


CHAPTER 2 | in situ analysis of single‐stranded and duplex siRNA integrity in living cells

60

ABSTRACT

To attain the full therapeutic promise of small interfering RNA (siRNA) it is believed that

improvements, such as increased biostability, are critical. Regrettably, thus far insufficient in situ

data are on hand regarding the intracellular stability of siRNAs. We report on the use of an advanced

fluorescence‐based method to probe nucleolytic decay of double labelled siRNAs which are subject

to fluorescence resonance energy transfer (FRET). In vitro measurements with RNase A and cellular

extracts demonstrate that the ratio of acceptor (5’‐Cy5) to donor (3’‐rhodamine green) fluorescence

can be used to study the degradation of the labelled siRNA substrates upon donor excitation.

Intracellular FRET analysis showed substantial degradation of single‐stranded siRNA, whereas duplex

siRNA stayed intact during the measured time period. These data underline the high intrinsic

nuclease resistance of unmodified duplex siRNA and prove that cellular persistence is much more

critical for the single‐stranded structure. It is for the first time that the stability of siRNA is

investigated in real‐time inside living cells. The presented fluorescence‐based method is a

straightforward technique to gain direct information on siRNA integrity inside living cells and

provides a bright outlook to learn more about the intracellular fate of siRNA therapeutics.



Chapter 2 provides:

• details on the validation and application of a fluorescence based method (FRET‐FFS) to

evaluate siRNA integrity

• a comparison of the stability between ssRNA/dsRNA upon incubation with RNase A and

cellular extracts

• an evaluation of siRNA integrity in living cells with FRET‐FFS following microinjection

Key terms:

FRET – ssRNA – dsRNA – FFS – FA/FD ratio – chemical modification – microinjection

INTRODUCTION

Knockdown of gene expression by RNA interference (RNAi) has evolved as a powerful tool for reverse

genetic analysis and holds a great therapeutic potential.[1,2] RNAi represents an evolutionary

conserved gene silencing mechanism by which ~21 nucleotide double‐stranded RNAs, termed small

interfering RNAs (siRNAs),[3] guide sequence specific cleavage and subsequent degradation of the

targeted mRNA and thus knockdown of the corresponding gene.[4] Intracellularly, duplex siRNA is

recognized by key components of the RNAi pathway which eventually leads to the incorporation of

the antisense (guide) strand in a ribonucleoprotein complex termed RISC (RNA‐induced silencing

complex). This activated RISC (RISC*) subsequently degrades the target mRNA employing the guide

strand as ‘docking site’ to the homologous mRNA sequence.[1‐5] The discovery that siRNAs can also

target mammalian genes without invoking the interferon response lightened the path to the

development of siRNA therapeutics.[5]

61


To effectively use siRNAs as drug molecules, many barriers still need to be overcome.[6] For successful

in vivo application of siRNAs, their biostability and cellular delivery should be improved.[1,2,6‐8] Many

efforts have already been undertaken to chemically modify the siRNA molecules in order to reduce

their susceptibility to RNases.[7,9‐13] Some of the investigated modifications indeed provide a higher

resistance against serum‐derived nucleases.[7,12,13] Moreover, evidence exists that chemical

modifications of siRNAs can lead to a prolonged gene silencing effect in cell culture, suggesting that

the transient and short lived RNAi induced gene knockdown, as usually observed in mammalian cells,

may be partially due to intracellular siRNA degradation.[9,11,12,14,15] However, in an attempt to confirm

this hypothesis, Layzer and coworkers had to conclude that the use of nuclease‐resistant siRNA does

not by definition result in a more potent and/or prolonged RNAi effect in vitro and in vivo.[13] Besides

chemical modification, the packaging of siRNAs into suitable delivery carriers may also protect them

from degradation,[16,17] also in vivo.[8] More detailed information can be found in chapter 1.

The interest of numerous research groups with regard to improvement of siRNA stability emphasizes

the need for useful and straightforward methods to probe siRNA integrity. To the best of our

knowledge, experiments which are able to follow the degradation of siRNA in real‐time in living cells

are not yet reported, mainly due to a lack of suitable non‐invasive methods. It may be clear however

that the development of siRNA as a drug needs an understanding of its intracellular fate. The major

focus of this manuscript is therefore to investigate the intracellular turnover of siRNAs. For this

purpose, we assessed the enzymatic degradation of double fluorescently labelled single‐stranded (ss)

and double‐stranded (ds) siRNA inside living cells by fluorescence resonance energy transfer (FRET;

Figure 2.1a). The intracellular tracking of siRNA duplexes carrying a single fluorophore makes it

impossible to distinguish between intact and degraded siRNA duplexes, strand separation or even

the emergence of free dye. Opposed to this approach, FRET has the advantage that it is based on the

direct and distance‐dependent interaction between two spectrally different fluorescent labels.[17] The

fluorescence intensities of both dyes ‐ i.e. rhodamine green and Cy5, attached to the 3’ and 5’ end of

the sense strand respectively ‐ were monitored using a dual‐colour fluorescence fluctuation

spectroscopy setup (FFS) as depicted in Figure 2.1b. Essentially, FFS measures fluorescence intensity

fluctuations in the excitation volume of a (confocal) microscope (Figure 2.1b ‐ 2.1d). The fluorescence

fluctuations are due to the movement of the fluorophores in and out the confocal volume (Figure

2.1c ‐ 2.1d). In dual‐colour FFS, a sample containing two spectrally different fluorophores (e.g.

rhodamine green and Cy5) is analyzed and their emission light is registered separately by two

detectors monitoring the same excitation volume (Figure 2.1b).[18]

Our results demonstrate that duplex siRNA is more resistant against intracellular nucleases when

compared with single‐stranded siRNA which is degraded within minutes. Additionally, this study

62


shows the potential of the delineated fluorescence‐based technique for future research on the cell

biological behaviour of siRNA.

FIGURE 2.1 | (a) Schematical representation of fluorescence resonance energy transfer (FRET) and its applicability to probe the degradation of double labelled siRNA (further information: see experimental section). (b) Setup of the dual‐colour fluorescence fluctuation spectroscopy (dual‐colour FFS) instrument. The excitation light, emerging from a Kr‐Ar ion laser, is reflected by a triple chroic mirror and focused into the objective of a confocal microscope. The fluorescence light, emitted by the fluorophores when diffusing through the excitation volume, passes the same objective and triple chroic mirror to be split by a dichroic mirror into a red and a green component. (c) Animated view of fluorescent molecules diffusing through the excitation volume (typically some femtoliters) which causes fluorescence intensity fluctuations as depicted in (d).

MATERIALS AND METHODS

siRNA oligonucleotides

21 nt single‐stranded (ss) siRNAs were synthesized and purified by Eurogentec (sense strand: 5’‐

GUGCGCUGCUGGUGCCAACdTdT‐3’, antisense strand: 5’‐GUUGGCACCAGCAGCGCACdTdT‐3’). Double

fluorescent labelling occurred at the 3’ and 5’ end of the sense strand with rhodamine green (RhoGr)

and Cy5 respectively. The concentration of the siRNA stock solutions (in DEPC‐treated water) was

calculated from absorption measurements at 260 nm (1 OD260 = 40 µg mL‐1) with a Nanodrop ND‐

1000 spectrophotometer. The absorption of the rhodamine label at 260 nm was taken into account.

The presence of the fluorescent labels was verified by absorption measurements at 500 nm for

RhoGr (εRhoGr, 500 nm = 54 000 L.mol‐1.cm‐1) and 647 nm for Cy5 (εCy5, 647nm = 250 000 L.mol‐1.cm‐1).

Fluorescent siRNA duplexes were prepared through hybridization of non‐labelled antisense and

fluorescently labelled sense RNA strands (either single or double labelled). Equimolar amounts of

63


both RNAs were mixed in annealing buffer (final buffer concentration 50 mM Tris, pH 7.5‐8.0, 100

mM NaCl) and annealing was performed in a thermal cycler (2 min incubation at 94°C followed by

slow cooling to 25°C over a time period of 45 min). Duplex formation was confirmed by 20% non‐

denaturing polyacrylamide gel electrophoresis.

Cell culture and microinjection experiments

VERO cells (African green monkey kidney cells) were maintained in Dulbecco’s modified eagle’s

medium (DMEM) without phenol red (Gibco™ Invitrogen) supplemented with 2 mM glutamine, 10%

heat deactivated fetal bovine serum (FBS) and 1% penicillin‐streptomycin at 37°C in humidified

atmosphere (5% CO2). Cells were prophylactically treated against mycoplasma with Plasmocin

(Invivogen). For intracellular fluorescence measurements, cells were seeded (2.5 x 104 cells per cm2)

on culture dishes with central 150 µm thick glass‐bottomed surface (MatTek Corporation). Cells were

incubated 24 h or 48 h prior to microinjection.

Microinjection experiments were conducted with a Femtojet® microinjector and an Injectman NI 2

micromanipulator (Eppendorf). All injections were performed in the cytoplasm. Directly following

microinjection, intracellular fluorescence measurements were carried out as described below.

Preparation of cytosolic and nuclear cell extracts

Cytosolic and nuclear cell fractions were derived from VERO cells according to a slightly modified BD

Bioscience Transfactor Extraction protocol. Cells were harvested at high confluency by trypsinization

and centrifuged at 450 g for 5 min (4°C). Subsequently, the cell pellets were rinsed twice by

resuspension in cold phosphate buffered saline (PBS), again collected and redispersed in a hypotonic

lysis buffer [10 mM HEPES pH 7.9, 1.5 mM MgCl2 and 10 mM KCl, supplemented with Complete Mini

EDTA‐free Protease Inhibitor Cocktail (Roche)]. Following incubation on ice (15 min) and subsequent

centrifugation (5 min at 420 g, 4°C), the cells were again dispersed in lysis buffer equal to twice the

cell pellet volume. The resulting cell suspension was slowly drawn into a narrow‐gauge syringe and

rapidly ejected to disrupt the outer membrane (disruption step was repeated 10 times). Nuclei were

pelleted by centrifugation of the lysed cells at 11000 g for 20 min. The supernatant (cytosolic extract)

was collected and stored in small aliquots at ‐80°C.

For preparation of the nuclear extract, the remaining nuclear pellet was resuspended in nuclear

extraction buffer (20 mM HEPES pH 7.9, 1.5 mM MgCl2, 0.42 M NaCl, 0.2 mM EDTA, 25% (v/v)

glycerol) again supplemented with Complete EDTA‐free Protease Inhibitor Cocktail in appropriate

concentration. Following disruption of the nuclei (using a new syringe), the nuclear suspension was

64


shaken gently for 30 min on ice and then centrifuged at 18000 g for 5 min. The supernatant (nuclear

extract) was stored in small aliquots at ‐80°C.

Dual‐colour fluorescence fluctuation spectroscopy setup

Fluorescence measurements were performed on a dual‐colour FFS setup installed on a MRC1024 Bio‐

Rad confocal laser scanning microscope (depicted in Figure 2.1b). An inverted microscope (Eclipse

TE300D, Nikon), equipped with a water‐immersion objective lens (Plan Apo 60x, NA 1.2, collar‐rim

correction, Nikon) was employed. The 488 nm and 647 nm laser lines of a krypton‐argon laser (Bio‐

Rad) were used to excite the RhoGr and Cy5 dye respectively and their emission signal was registered

by ultrasensitive Avalanche photodiode detectors. The fluorescence intensity fluctuations were

recorded on a digital ALV 5000/E correlator. The FFS setup was calibrated as described by Schwille et

al. [19] and summarized by Remaut et al. [20] to optimize the overlap of the excitation and detection

volumes and to determine the size of the latter.

A detailed outlining of the theoretical concept and applications of FFS can be found elsewhere.[21,22]

In this study the dual‐colour FFS setup is mainly employed to measure the fluorescence intensities of

labelled siRNA present at nanomolar concentrations. Additional correlation analysis of the

fluorescence fluctuations to obtain information on the mobility of the siRNA molecules is not

performed. Therefore, the mathematical basics and diffusion models to be applied on FFS data will

not be covered in this chapter.

siRNA stability analysis

Fluorescence resonance energy transfer (FRET) is an important research tool to study inter‐ and

intramolecular processes and is defined as the non‐radiative excitation energy transfer from a donor

fluorophore to an appropriate acceptor fluorophore when both dyes are fixed in close proximity

(typically ~10‐100 Å).[23] The energy transfer efficiency is inversely proportional to the sixth power of

the inter‐dye distance.[23] Hence, virtually any process that induces distance variations between

donor and acceptor on the nanometer scale can be monitored with this technique.[24] In this report,

FRET is used as a tool to study the degradation of double labelled siRNAs (see results section). Intact

nucleic acid molecules, bearing both donor and acceptor, show a significant acceptor emission upon

donor excitation due to energy transfer, as schematically illustrated in Figure 2.1a. When the double

labelled substrates are cleaved into single labelled products, FRET can no longer occur resulting in

negligible acceptor fluorescence. Basically, the ratio of the 5’‐Cy5 fluorescence (acceptor

fluorophore) to the 3’‐RhoGr fluorescence (donor fluorophore) upon donor excitation (termed FA/FD)

65


correlates with siRNA integrity (see results section). For the determination of the FA/FD ratio on

degrading siRNAs, the fluorescence intensities of RhoGr and Cy5 were recorded by the “green” and

“red” detector respectively (Figure 2.1b) during 3 seconds, with laser excitation set at 488 nm. All

samples were analyzed in glass‐bottomed 96‐well plates (Greiner Bio‐One, certified DNase/RNase

free) and the focal volume was placed 50 µm above the bottom of the wells.

Degradation experiments with RNase A (RPA grade, manufactured from bovine pancreas, Ambion)

were performed at ambient temperature by first diluting the siRNA stock solutions with RNase A

buffer (100 mM Tris‐acetate pH 6.5, 1 mM EDTA) to a final concentration of 13 nM. Subsequently, 1

µL of RNase A solution was added and FA/FD was measured as a function of time.

To monitor the degradation of siRNAs in cytosolic and nuclear extracts of VERO cells, a siRNA solution

(400 nM) in “degradation buffer” (20 mM Tris‐HCl, 50 mM Na‐acetate, 2 mM Mg‐acetate, adjusted

with sodium hydroxide to pH 7.4) was mixed with 50% (v/v) of cellular extract and incubated at 37°C

in non‐stick RNase free microfuge tubes (Ambion). Aliquots were taken at different time points and

diluted with sodium citrate buffer to stop degradation (for dsRNA degradation: 10 mM sodium

citrate pH 4, 0.025 % sodium dodecylsulphate (SDS); for ssRNA degradation: 10 mM sodium citrate

pH 6, 20 mM DTT, 0.1 % SDS). Samples (end concentration 13 nM) were analyzed immediately or

incubated on ice prior to FA/FD measurements.

To monitor the stability of siRNA in VERO cells, first confocal images were taken of the cells and the

FFS observation volume was positioned inside the nucleus. Solutions of intact ssRNA (10 µM), intact

dsRNA (5 µM), (RNase A) degraded ssRNA (10 µM) and (RNase A) degraded dsRNA (6 µM) were

prepared in RNase‐free water and loaded in a microinjection needle (Eppendorf II). Subsequently,

cells were microinjected in the cytoplasm. Approximately 30 to 60 sec after injection, the cells were

exposed to 488 nm laser light and the FA/FD ratio was determined in the nucleus. Low laser excitation

power and short measurement time ensured minimal photobleaching and cellular damage. For time‐

dependent stability analysis, the FA/FD ratio was registered inside the nucleus at different time‐points

after injection. At the end of each data series we verified that the nucleus was still in focus.

Polyacrylamide gel electrophoresis combined with dual‐colour FFS

27.5 µL siRNA duplex (8 µM) was added to 0.48 µg RNase A in 12.5 µL RNase A buffer and incubated

at room temperature. At specified times, 5 µL aliquots (for gel electrophoresis, ~ 0.4 µg siRNA) and

0.5 µL aliquots (for dual‐colour FFS) were removed. Electrophoresis aliquots were mixed with 10 µL

formamide (with 0.05% bromophenol blue), snap frozen at ‐80°C and stored afterwards at ‐20°C. A

20% polyacrylamide gel, supplemented with 7 M urea in TBE buffer (0.089 M Tris base, 0.089 M boric

66


acid (pH 8.35), 2 mM Na2EDTA), was applied to analyze siRNA degradation (40 min run at 100V).

Following electrophoresis, the gel was stained (40 min) in a 1:104 dilution of SYBR Green II RNA stain

(Molecular Probes) in DEPC‐treated water and samples were visualized on a UV Trans‐illuminator.

The 0.5 µL aliquots were added to 7 µL of RNase A buffer and also snap frozen at ‐80°C. The FA/FD

ratio (upon 488 nm excitation) of every aliquot was calculated.

RESULTS

FRET as a tool to distinguish intact and degraded single‐stranded and duplex siRNA

Figure 2.2 shows representative fluorescence fluctuation profiles as obtained for intact ssRNA (Figure

2.2a) and intact dsRNA (Figure 2.2b) in DEPC‐treated water, upon 488 nm excitation. Although with

these settings only rhodamine green (RhoGr) can be efficiently excited, the Cy5 fluorescence (red

signal) markedly exceeds the RhoGr fluorescence (green signal). This clearly indicates that energy

from the RhoGr excited state (donor) is transferred to and excites the Cy5 acceptor dye (FRET). The

ratio of acceptor fluorescence to donor fluorescence upon 488 nm excitation (FA/FD ratio) is used

here as a measure for energy transfer (see Materials and Methods). Table 2.1 summarizes the FA/FD

values as measured for ssRNA and dsRNA. Both intact ssRNA and dsRNA display a FA/FD ratio >> 1 in

DEPC‐treated water which is mainly due to FRET. Upon annealing with non‐labelled complementary

siRNA, the FA/FD ratio of ssRNA clearly lowers, indicating that the energy transfer is partially inhibited

in the duplex form. This is probably explained by a higher rigidity of the duplex structure which

makes the relative position of both dyes less optimal for FRET. Also, the FA/FD value seems to depend

on the medium in which the siRNA is diluted (Table 2.1). When the siRNA is placed in RNase A buffer

or sodium citrate buffer, a lower FA/FD ratio was measured in comparison with FA/FD in DEPC‐treated

water (Table 2.1), indicating that in the buffer media energy transfer is significantly diminished.

Fluorescence fluctuations (upon 488 nm excitation) of degraded siRNA samples are shown in Figure

2.2c and 2.2d. In contrast with the fluorescence fluctuation profiles of intact siRNA samples, now the

RhoGr emission becomes dominant. This red‐to‐green shift is due to the degradation of the siRNAs

which separates donor and acceptor dye, thus precluding energy transfer (Figure 2.1a). SiRNA

degradation is also clearly reflected in the FA/FD values (Table 2.1). Incubation of siRNA in the

presence of RNase A and cellular extracts (derived from VERO cells) does no longer allow FRET and

significantly lowers the FA/FD ratio (FA/FD << 1).

67


TABLE 2.1 | Overview of the measured FA/FD ratios of single‐stranded (ss) and double‐stranded (ds) siRNA. Intact siRNA oligos show a high ratio primarily due to FRET. Degradation of siRNA in the presence of cellular extracts or RNase A corresponds with FA/FD << 1. Ratios were calculated based on average red and green fluorescence intensities, measured over a 30 s time frame (excitation wavelength equalled 488 nm). Different samples were analyzed and FA/FD ratios are presented as mean values ± standard deviation of at least three independent experiments. Buffer A: RNase A buffer; buffer B: sodium citrate buffer, supplemented with SDS.

INTACT DEPC treated water Buffer A Buffer B

ssRNA 29.3 ± 2.0 15.5 ± 2.1 19.6 ± 1.9

dsRNA 9.5 ± 0.8 1.5 ± 0.1 1.9 ± 0.1

DEGRADED RNAse A Nuclear extract Cytoplasmic extract

ssRNA 0.11 ± 0.01 0.13 ± 0.01 0.12 ± 0.01

dsRNA 0.11 ± 0.00 0.09 ± 0.01 0.14 ± 0.01

FIGURE 2.2 | Fluorescence fluctuation profiles of intact/degraded ssRNA (a,c) and dsRNA (b,d) as obtained in DEPC treated water upon 488 nm excitation. A graphical representation of the predominant siRNA species in the measured samples is added. Intact siRNA shows a higher Cy5 fluorescence signal as compared to the RhoGr emission due to fluorescence resonance energy transfer (FRET). When the oligos are degraded, FRET can no longer occur eventually leading to the inversion of the emission profile.

68


FIGURE 2.3 | (a,b) Real‐time emission behavior of double fluorescently labelled siRNAs (488 nm excitation). Analysis was performed through continuous excitation over a 400 s time‐interval of a 13 nM siRNA solution (in RNase A buffer) incubated at ambient temperature together with an excess of RNase A (2.5 ng and 1.5 µg RNase A for ssRNA and dsRNA respectively). Arrows indicate the time‐point of RNase A addition. Degradation of ssRNA results in an increase in RhoGr emission (grey data) and a concomitant decrease in Cy5 fluorescence (black data) due to the disappearance of FRET. Regarding the siRNA duplex, the FA/FD ratio is primarily influenced by the increase in donor fluorescence. (c,d) Time‐dependent profiles of the resulting FA/FD ratio. To normalize the results, the FA/FD ratio of intact siRNA was set to 1.

To further unravel the FA/FD turnover during degradation, the red and green fluorescence were

monitored in time after addition of RNase A. In this experiment an excess of RNase A was added to

allow monitoring a significant decrease in FA/FD on a short time‐scale (400 s) with continuous laser

excitation. As can be seen in Figure 2.3a, the degradation of ssRNA corresponds with an increase in

green fluorescence and a concomitant decrease in red fluorescence, due to the disappearance of

FRET. During degradation of dsRNA, the green fluorescence also gradually increases, while the red

fluorescence, however, remains almost unchanged (Figure 2.3b). We observed that degradation of

single RhoGr labelled siRNAs also increases the green fluorescence which we attributed to RhoGr

dequenching (data not shown). It cannot be excluded that this phenomenon also occurs in double

labelled siRNAs. In other words, one can say that the change in FA/FD upon degradation of ssRNA and

69


dsRNA is governed by both a loss of FRET and intramolecular RhoGr fluorescence dequenching. It is

important to emphasize that a direct comparison of ssRNA and dsRNA degradation kinetics cannot be

made from FA/FD measurements due to the fact that different processes influence the decrease in

FA/FD ratio of respectively ssRNA and dsRNA. More importantly however, Figure 2.3c and 2.3d clearly

illustrate that the FA/FD ratio allows following the conversion between intact and degraded siRNA.

Monitoring the degradation of ssRNA and dsRNA in RNase A solutions

First, we tried to characterize the correlation between the FA/FD ratio and the percentage of

degraded siRNA in a sample. Hereto, different amounts of intact and degraded siRNA (resulting from

long‐term incubation at 37°C with nuclear or cytosolic extract) were mixed and the FA/FD ratio was

determined. As can be seen in Figure 2.4, a higher amount of degraded siRNA corresponds with a

lower FA/FD ratio, although a linear relation was not found. These calibration curves clearly show that

the FA/FD ratio is mainly sensitive towards the onset of siRNA degradation (from 0% to 20% degraded

siRNA). When the sample contains only degraded siRNA, the FA/FD value drops to circa 0.6% and 5%

of its initial value for respectively ssRNA (Figure 2.4a) and dsRNA (Figure 2.4b).

FIGURE 2.4 | Correlation between the FA/FD ratio and the percentage of degraded siRNA in a mixture of intact and degraded siRNA (in sodium citrate buffer, supplemented with SDS). The FA/FD ratio decreases with increasing concentrations of degraded siRNA as only intact RNA can contribute to the FRET signal. The FA/FD ratio of intact RNA was accounted value 1.

Figure 2.5 shows the degradation profiles of ssRNA and dsRNA after addition of different amounts of

RNase A. Breakdown of the RNA initially strongly lowers the FA/FD ratio followed by a more gradual

decrease. As expected, adding higher amounts of enzyme results in a faster decrease in the FA/FD

ratio, which reflects a higher degradation rate. RNase A is an endonuclease with a single‐strand

substrate specificity that cleaves 3’ from C and U residues. Only low amounts of RNase A are needed

to degrade a nanomolar ssRNA sample (Figure 2.5a). However, at higher enzyme concentrations also

70


dsRNA is subject to degradation (Figure 2.5b). Indeed, it is known that the specificity of RNase A is

dependent on the concentration used.

FIGURE 2.5 | Time‐dependent degradation profiles (normalized FA/FD in function of degradation time) of ssRNA (a) and dsRNA (b) with different amounts of RNase A (depicted in both figures) in RNase buffer. The siRNA concentration was equal in all samples (13 nM). FFS measurements were performed at ambient temperature with laser excitation set at 488 nm. Control measurements were in RNase buffer alone.

To visualize dsRNA degradation we performed denaturing PAGE gel electrophoresis (Figure 2.6). As

expected, a clear degradation pattern could be seen. After 24 h of incubation almost no detectable

nucleic acid fragments were remaining on the gel. FFS based stability analysis on the degrading

samples was performed in parallel (Figure 2.6) and elucidated that merely mixing the dsRNA with

RNase (corresponding with the 2’ time point) already reduced the FA/FD ratio by 37 %. Following 2 h

of incubation the FA/FD ratio does not decrease any further. This indicates that the degradation

primarily takes place during the first two hours of incubation. As almost no full length fragments

remained visible on the gel at the 2 h time point, the information obtained from respectively the gel

electrophoresis measurements and the FA/FD measurements are in good agreement, indicating that

the FA/FD ratio effectively reflects the amount of intact and/or degraded double labelled siRNA in a

sample.

Monitoring the degradation of ssRNA and dsRNA in cellular extracts

In a next step we wondered whether the FRET‐FFS technique also allows studying the stability of

siRNA in more complex media. Hereto, cytoplasmic and nuclear fractions were isolated from VERO

cells and incubated with single‐ and double‐stranded RNA.

71


FIGURE 2.6 | Gel electrophoresis on degrading dsRNA samples. 0.48 µg RNase A was added to a 5.5 µM dsRNA solution in RNase A buffer. At different incubation times (ambient temperature), FFS aliquots were taken and snap frozen at ‐85°C. Prior to measurement, the samples were diluted to a concentration suitable for FFS analysis. The normalized FA/FD ratio in function of degradation time is shown. Electrophoresis aliquots (~0.4 µg dsRNA) were taken at the same time points and also stored immediately at ‐85°C. Graph inset shows the corresponding dsRNA degradation profile as visualized with SYBR Green II RNA stain on a denaturing 20% polyacrylamide gel.

Figure 2.7 shows the corresponding degradation profiles. A first interesting observation is that single‐

stranded RNA is almost completely degraded after 60 min of incubation (Figure 2.7a), while complete

degradation of duplex siRNA, even after several hours of incubation, was not obtained (Figure 2.7b).

A significantly slower degradation rate was seen when experiments were repeated with 1:10 diluted

extract (data not shown). Surprisingly, ssRNA seems to degrade equally fast in cytoplasmic extracts as

in nuclear extract (Figure 2.7a) indicating that both extracts possess a comparable ssRNase activity.

However, regarding the siRNA duplex the FA/FD ratio shows a considerably faster decay in nuclear

extract compared to cytosolic extract. This observation suggests that, under the given assay

conditions, higher nuclease activity was present in the nuclear fraction of the VERO cells.

Monitoring the intracellular degradation of ssRNA and dsRNA

Having established that the FA/FD ratio is a reliable parameter to follow the siRNA integrity in cellular

extracts, we next aimed to analyze the intracellular stability of siRNA in living cells using

microinjection to deliver the siRNA to the cytosol.

72


FIGURE 2.7 | FFS measurements performed on degrading siRNA oligos in cellular extracts. Crude extracts were mixed with twice the volume of a 400 nM RNA‐solution in degradation buffer and incubated at 37°C. At different time‐points an aliquot was mixed with SDS‐sodium citrate buffer to a final concentration of 13 nM. Excitation wavelength equaled 488 nm and fluorescence intensities were recorded during 3 s. FA/FD ratios were again normalized to value 1. (a) No difference in degradation rate could be observed for ssRNA in pure cytoplasmic and nuclear extract. (b) Concerning dsRNA however, the FA/FD ratio shows a faster decay in nuclear extract compared to cytosolic extract. Control measurements are performed in degradation buffer alone (dsRNA) or degradation buffer supplemented with 0.1% SDS (ssRNA). Influence of siRNA‐protein binding on the measured FA/FD signal was minimized through dilution of the samples in the denaturing buffer prior to measurement.

Figure 2.8a shows confocal images (488 nm excitation) of VERO cells taken at different time points

after cytoplasmic injection of a high concentration of intact dsRNA. Duplex siRNA seems to rapidly

accumulate in the nucleus subsequent to injection and clearly does not show a homogenous

intranuclear distribution. A comparable behavior has already been observed for antisense

oligonucleotides.[25] Also single‐stranded siRNA showed a fast nuclear accumulation following

microinjection (data not shown). Hence, further exploration of the intracellular integrity of both

siRNAs was performed in the nucleus of the injected VERO cells. Figure 2.8b depicts the FA/FD ratio as

measured in the nuclei directly after microinjecting respectively intact or previously degraded (ss and

ds) siRNA in the cytosol of the cells. As the FFS setup makes use of more sensitive detectors when

compared with a conventional fluorescence microscope, a lower amount of siRNA was injected here.

For injected intact siRNA a FA/FD ratio > 1 was calculated, while FA/FD remained far below 1 when

degraded siRNA was introduced in the cytosol. This clearly evidences that the dual‐colour FFS setup

can distinguish between intact and degraded siRNA inside living cells.

Next, we aimed to characterize the intracellular degradation rate of duplex siRNA in VERO cells. For

this purpose, we microinjected intact double‐stranded siRNA in the cytosol of VERO cells and

positioned the FFS observation volume in the nucleus of the cell.

73


FIGURE 2.8 | (a) Transmission image and confocal images (488 nm excitation) of VERO cells taken at different time points after cytoplasmic injection of dsRNA. (b) FA/FD ratio as measured in cells after microinjecting intact or (previously) degraded ssRNA and dsRNA. The FFS observation volume was positioned in the nucleus of the injected cell with laser excitation set at 488 nm. The FA/FD ratio is presented as mean value ± standard deviation obtained from at least three independent experiments under identical calibration.

During the first half hour following injection of dsRNA an increase in both the red and green nuclear

fluorescence was observed (Figure 2.9a) in agreement with the nuclear accumulation of duplex

siRNA as seen in Figure 2.8a. Subsequently, a slow decay in the fluorescence intensities was

observed, which can be attributed to a redistribution of intact nuclear dsRNA to the cytosolic

compartment. A similar behaviour was reported for antisense oligonucleotides which seem to shuttle

between nucleus and cytoplasm.[26] Interestingly, the FA/FD ratio remains constant over the measured

time interval (Figure 2.9c) demonstrating that the duplex siRNA stays intact. This is in contrast with

the degradation of dsRNA in cellular extracts as concluded from Figure 2.7b. These observations

prove that siRNA stability in cellular extracts cannot just be translated to the actual cellular

environment and highlight that in situ measurements are needed. Intracellular measurements with

ssRNA revealed a vast lowering of the FA/FD ratio (Figure 2.9d), indicating the disappearance of FRET

and RhoGr dequenching and thus ssRNA degradation.

74


FIGURE 2.9 | Intranuclear fluorescence intensity profile of (a) dsRNA and (b) ssRNA. The emission signal in both the green (■ gray) and red (● black) detector is shown in function of degradation time. (c,d) Corresponding time‐dependent degradation profiles of dsRNA and ssRNA obtained in the nucleus of VERO cells. Laser excitation was set at 488 nm and microinjection was conducted in the cytoplasm of the focus cell. Normalized FA/FD ratios of dsRNA are presented as mean values ± standard deviation of three independent experiments under different calibration. ssRNA data are from one representative of three independent experiments.

The intracellular data underline the high stability of duplex siRNA while single‐stranded siRNA is

highly unstable, in line with other reports in literature.[10,11] Interestingly, in Figure 2.9b again a

concomitant increase in red and green fluorescence was registered in the nucleus during the first 5

min, attributed to the nuclear import of the ssRNA. Clearly, the single‐stranded siRNA shows a faster

accumulation when compared with duplex siRNA. However, the relative increase in red fluorescence

is less pronounced than the increase in green fluorescence explaining the FA/FD decline within this

time frame. It can be postulated that upon injecting ssRNA in the cytoplasm, a fraction of the RNA

will be readily degraded. Hence, during the first few minutes after microinjection, a mix of degraded

and intact ssRNA is prone to nuclear import. It goes without saying that the intact single‐stranded

siRNA molecules that reach the nucleus will also degrade within a reasonable time span. As outlined

previously, ssRNA degradation results in an increase in green fluorescence and a parallel lowering of

75


the red fluorescence. As the red emission already drops at 5 min after injection (Figure 2.9b) we

conclude that little intact siRNAs still reach the nucleus, but that the siRNA molecules at this moment

are subject to intranuclear degradation. One can wonder why between 5 min and 50 min following

injection the green fluorescence in the nucleus remains more or less constant. Probably the loss of

intranuclear fluorescence due to nucleocytoplasmic redistribution of (intact and already degraded)

siRNA is countered by the increase in green fluorescence as a result of degradation. Approximately

60 min after injection, the FA/FD does no longer drop (Figure 2.9c), indicating that no further

degradation occurs. The slight decrease in red and green fluorescence is now completely attributed

to diffusion of siRNA fragments from the nucleus into the cytoplasm.

DISCUSSION

Many disease‐causing genes can be silenced efficiently in cell culture with rationally designed small

interfering RNAs under optimized transfection conditions. However, application of siRNA

therapeutics is not yet conceivable as siRNA delivery, biodistribution, target cell selectivity etc.

remain major hurdles to be overcome. Obviously, the therapeutic success of siRNA will also greatly

depend on the biostability of siRNA. Indeed, the siRNA molecules should be sufficiently stable in the

extra‐ and intracellular matrices to be able to exert a biological effect.[1,2,6]

Previous reports in the literature have studied degradation of synthetic RNA/DNA chimera [27] and

antisense oligonucleotides by FRET.[18] Also inside living cells, FRET has been used to evaluate e.g.

ribozyme cleavage of a labelled oligoribonucleotide substrate [28] and for detection of intact and

degraded oligonucleotides.[18,29] In a recent report, FRET was applied to track the degradation of

double‐labelled pDNA in the intracellular environment through confocal time‐lapse imaging.[30] Modi

et al. even describe the design of so called ‘DNA nanomachines’, bearing an appropriate FRET pair, as

intracellular pH sensors that are able to map pH changes during intracellular trafficking.[31] FRET as a

tool to evaluate siRNA integrity in particular is described in three independent reports in the

literature.[17,32,33]

In the presented study we first wanted to evaluate whether our dual‐colour FFS setup (as depicted in

Figure 2.1b) is suited to probe nucleolytic decay of double fluorescently labelled siRNAs which are

subject to FRET. Figures 2.3 to 2.7 clearly show that the FA/FD ratio, as obtained from the

fluorescence fluctuation data (measured through the FFS setup), can be applied to study the

transition from intact to degraded siRNA. As single‐stranded siRNA shows much higher FRET

efficiency than duplex siRNA (Table 2.1), this also allows us to distinguish between dsRNA

degradation and duplex strand denaturation. While dsRNA degradation results in a decrease of the

76


FA/FD ratio, destabilization of the intact siRNA duplex will eventually lead to an increase of the FA/FD

ratio. Indeed, release of the more flexible siRNA sense strand from the duplex structure will induce a

higher FRET signal.

Correlation analysis on the fluorescence fluctuation profiles would allow the calculation of the

diffusion coefficient of the fluorescently tagged molecules passing through the observation volume

of the confocal microscope.[19,21,22] However, this strategy is not recommended to monitor the

degradation of siRNA as a function of time.[20] Strictly, FFS can only distinguish between fluorescent

molecules with sufficiently different mobility i.e. when their diffusion coefficients (D) differ for at

least factor 1.6.[34] Considering D ~ 1/(MW)1/3, this implies that FFS could only differentiate between

intact and fully degraded siRNA. In contrast, FRET is sensitive to a single cleavage, irrespective of the

molecular weight (MW) of the degradation products formed.[20] It is also worth highlighting that in

this report end‐labelled siRNAs are used. It could be argued that the labels sterically interfere with

(primarily) exoribonuclease binding to the siRNA substrate thereby limiting the enzymatic activity

which would lead to an underestimation of the degradation rate.

Using a confocal FFS setup to register the fluorescence intensities is in particular advantageous for

quantifying the intracellular fate of siRNAs. Indeed, the confocal excitation volume can easily be

placed inside a defined cellular compartment. As the fluorescence photons emanating from this focal

spot are registered by ultrasensitive avalanche photodiode detectors this enables us to measure low

concentrations of fluorescent molecules which could provide a suitable alternative for confocal

imaging in the case the fluorescence signal becomes too dim to provide satisfactory CLSM images.[35]

Especially for future siRNA research this could be attractive as only low concentrations of highly

active siRNA are needed to show a biological effect. It is also known that fluorescence from

endogenous cellular components (autofluorescence) may interfere with FFS measurements in

mammalian cells, especially when excitation wavelengths in the blue spectral range (< 500 nm) are

used.[36] In particular, the autofluorescence signal may render the interpretation of intracellular data

challenging [37] especially when the fluorescently labelled molecules of interest are present at low

concentration. In this report, the intracellular fluorescence exceeded the autofluorescence at least 5‐

fold at every data point, making sure that the obtained FA/FD ratios really reflect the fate of the

injected siRNAs.

Confocal images showed nuclear accumulation of dsRNA when microinjected in the cytoplasm

(Figure 2.8a). Several reports, however, present evidence for the cytoplasmic nature of RNAi as

activated RISC complexes localize and function in the cytosol where translation of target mRNA takes

place.[38] Transfection of Hela cells with TAT47‐57‐siRNA conjugates, siRNA/lipofectamine nanoparticles

or siRNA/polyamidoamine nanoparticles revealed that the siRNAs preferentially locate in the

77


perinuclear region, thought to contain focal points of RNAi machinery.[39] This specific subcellular

localization showed strong correlation with RNAi activity. A recent report, however, declared that

RNAi activity is also present in the nucleus of HeLa cells as functional RISC complexes were found in

both nuclear and cytoplasmic extracts.[40] Thus, upon transfection, small interfering RNA that is

released from its carrier presumably can enter the nucleus and form RISC* complexes on site to

cleave homologous nuclear RNA transcripts. It has also been shown in the literature that siRNAs can

induce transcriptional gene silencing in mammalian cells through DNA methylation, a process that

occurs in the nucleus, which is explained in more detail in the previous chapter.[41] Järve et al. further

substantiate our observations since these authors also mention the intranuclear transport of

microinjected siRNAs. Moreover, nuclear siRNA was detected following liposomal transfection as

well.[17] The siRNA duplexes that reach the nucleus are however quickly (~hours) relocated to the

cytoplasm through the RanGTP/Exportin 5 pathway [17,42], where they can localize to cytoplasmic P‐

bodies.[33,39] Many uncertainties on the intracellular fate of siRNA remain to exist and further

underline the need for advanced methods to investigate them. On the other hand, one should keep

in mind that the intracellular distribution of siRNAs will probably depend on how they are delivered

to the cytosol. It is indeed possible that the intranuclear accumulation of siRNA is influenced by

microinjection or transfection procedures.[17]

Intracellular FFS measurements showed substantial degradation of the single‐stranded RNA. In

contrast, the FA/FD ratio as measured for dsRNA remained unchanged for up to 2 h, emphasizing the

higher intrinsic stability of dsRNA (Figure 2.9c). Probably the low dsRNase activity in both the

cytoplasmic and nuclear compartment, as compared to the presence of single‐strand specific

nucleases, explains the discrepancy in the stability of ssRNA and dsRNA. A recent article on the

stability of native siRNAs in serum revealed that duplex siRNA is mainly degraded by enzymes of the

(ss)RNase A family.[43] Some of these enzymes can be termed single‐strand preferring, rather than

single‐strand specific, because they are also capable of degrading double‐stranded RNA, although

through a much lesser extent. The biochemical mechanism of this process has already been

described in the literature.[44] Inside living cells, there are numerous types of ribonucleases (exo‐ and

endoribonucleases) with different structures and catalytic function.[45] Because of the high

complexity and diversity of this class of molecules, it is difficult to predict the predominant

ribonuclease pathway involved in the intracellular degradation of siRNA. Because only one siRNA

sequence was studied in this chapter, it is not possible to draw any conclusions with regard to the

sequence‐dependent degradation of siRNA. However, on the basis of the fact that different articles in

literature suggest that single‐stranded siRNA degrades substantially faster than a siRNA

duplex,[3,10,11,46] we are quite confident that in general for native siRNAs the duplex structure will

78


show a higher stability in a biological environment compared to the stability of the single‐stranded

counterpart.

The observations in Figure 2.9 also raise the interesting question of how RISC incorporation of the

double‐labelled siRNA duplex would influence the observed FA/FD profile. Assembly of the

(unlabelled) guide strand into RISC, followed by unwinding of the duplex structure and release of the

(double labelled) passenger strand, would imply degradation of the latter.[47,48] As a consequence,

single labelled RNA fragments would emerge and cause a decrease in the FA/FD ratio, which is not

observed here. As blocking of the 5’‐end of the antisense strand can abolish RNAi activity and a free

terminal 5’‐phosphate is believed to be essential for RNAi activity,[42,49‐52] it is very unlikely that the 3’‐

5’ double labelled siRNA strand will act as guide strand. From this we eventually conclude that RISC

activation does not occur under these experimental conditions, thereby also partially explaining the

significant nuclear accumulation as we observed in our confocal microscopy experiments. It is a

possibility that the relatively high intracellular siRNA concentrations can lead to saturation of the

cellular compartments and the RNAi machinery, thereby masking the interaction of siRNA with RISC.

It could be hypothesized that the lower stability of ssRNA is partially responsible for less effective

gene silencing activity of antisense siRNA compared to duplex siRNA.[3,46] It has been suggested that

single‐stranded antisense siRNA can also directly enter the mammalian RNAi pathway with

subsequent depletion of endogenous gene expression [9,46,52] but still the duplex siRNA potency

generally exceeds the single‐stranded RNAi activity. Also, in most cases duplex siRNA shows a more

persistent gene silencing than antisense DNA oligonucleotides [1,2,12] which is, according to Bertrand

et al.,[53] primarily attributed to their difference in susceptibility for nuclease attack.

The stability data we present here have been gathered inside living cells after microinjection of the

siRNA molecules in the cytosol. Because no transfection procedure was applied, the intracellular

environment to which the microinjected siRNAs are exposed resembles that of siRNAs released from

a pharmaceutical nanocarrier (e.g. liposomes and polymeric nanoparticles) into the cytoplasm. As the

applied delivery system, together with incorporation of the antisense siRNA strand into RISC, could

provide extra protection of the siRNA against nuclease attack, especially the time frame between

release from its carrier and interaction with RISC will be crucial for the siRNA integrity inside the

intracellular matrix.[46] Our results suggest that intracellular degradation is much more critical for

ssRNA than for duplex siRNA. The observation that modified siRNA duplexes with enhanced nuclease

resistance do not always show longer lasting RNAi effect [13] supports the hypothesis that intracellular

stability is not the key issue for effective application of siRNA duplexes. It can be postulated that

upon delivery to the cytosol, non‐modified duplex siRNA is able to induce a response before

significant degradation occurs.[13] Bartlett et al. provided data on the anti‐luciferase activity of

79


unmodified and nuclease‐stabilized siRNA duplexes after electroporation and observed no

differences in the amplitude nor the duration of the gene silencing effect.[54] These results strongly

suggest that once the siRNA duplexes reach the cell cytoplasm, nuclease degradation is absent or at

least not the limiting factor. Hence, further research into functional in vivo delivery systems that can

safely guide active siRNA to the cytosol of the target cell should be stimulated.

ACKNOWLEDGMENT

N. Opitz (Max Planck Institute for Molecular Physiology, Dortmund, Germany) is thanked with

gratitude for the installation of the FFS‐module on the MRC‐1024.

80


REFERENCE LIST

1. Dorsett Y. and Tuschl T., siRNAs: applications in functional genomics and potential as therapeutics, Nat. Rev. Drug Discov. 2004, 3: 318‐329.

2. Dykxhoorn D.M. and Lieberman J., The silent revolution: RNA interference as basic biology, research tool, and therapeutic, Annu. Rev. Med. 2005, 56: 401‐423.

3. Elbashir S.M., Lendeckel W., and Tuschl T., RNA interference is mediated by 21‐and 22‐nucleotide RNAs, Genes & Development 2001, 15: 188‐200.

4. Zamore P.D., Tuschl T., Sharp P.A., and Bartel D.P., RNAi: Double‐stranded RNA directs the ATP‐dependent cleavage of mRNA at 21 to 23 nucleotide intervals, Cell 2000, 101: 25‐33.


6. Schiffelers R.M., Woodle M.C., and Scaria P., Pharmaceutical prospects for RNA interference, Pharm. Res. 2004, 21: 1‐7.

7. Elmen J., Thonberg H., Ljungberg K., Frieden M. et al., Locked nucleic acid (LNA) mediated improvements in siRNA stability and functionality, Nucleic Acids Res. 2005, 33: 439‐447.

8. Urban‐Klein B., Werth S., Abuharbeid S., Czubayko F. et al., RNAi‐mediated gene‐targeting through systemic application of polyethylenimine (PEI)‐complexed siRNA in vivo, Gene Ther. 2005, 12: 461‐466.

9. Amarzguioui M., Holen T., Babaie E., and Prydz H., Tolerance for mutations and chemical modifications in a siRNA, Nucleic Acids Res. 2003, 31: 589‐595.

10. Braasch D.A., Jensen S., Liu Y., Kaur K. et al., RNA interference in mammalian cells by chemically‐modified RNA, Biochemistry 2003, 42: 7967‐7975.

11. Chiu Y.L. and Rana T.M., siRNA function in RNAi: a chemical modification analysis, RNA 2003, 9: 1034‐1048.

12. Czauderna F., Fechtner M., Dames S., Aygun H. et al., Structural variations and stabilising modifications of synthetic siRNAs in mammalian cells, Nucleic Acids Res. 2003, 31: 2705‐2716.

13. Layzer J.M., McCaffrey A.P., Tanner A.K., Huang Z. et al., In vivo activity of nuclease‐resistant siRNAs, RNA‐A Publication of the Rna Society 2004, 10: 766‐771.

14. Volkov A.A., Kruglova N.S., Meschaninova M.I., Venyaminova A.G. et al., Selective protection of nuclease‐sensitive sites in siRNA prolongs silencing effect, Oligonucleotides. 2009, 19: 191‐202.

15. Mantei A., Rutz S., Janke M., Kirchhoff D. et al., siRNA stabilization prolongs gene knockdown in primary T lymphocytes, Eur. J. Immunol. 2008, 38: 2616‐2625.

16. Remaut K., Lucas B., Braeckmans K., Sanders N.N. et al., Protection of oligonucleotides against nucleases by pegylated and non‐pegylated liposomes as studied by fluorescence correlation spectroscopy, J. Control Release 2005, 110: 212‐226.

17. Jarve A., Muller J., Kim I.H., Rohr K. et al., Surveillance of siRNA integrity by FRET imaging, Nucleic Acids Res. 2007, 35: e124.

18. De Smedt S.C., Remaut K., Lucas B., Braeckmans K. et al., Studying biophysical barriers to DNA delivery by advanced light microscopy, Adv. Drug Deliv. Rev. 2005, 57: 191‐210.

19. Schwille P., Meyer‐Almes F.J., and Rigler R., Dual‐color fluorescence cross‐correlation spectroscopy for multicomponent diffusional analysis in solution, Biophys. J. 1997, 72: 1878‐1886.

20. Remaut K., Lucas B., Braeckmans K., Sanders N.N. et al., FRET‐FCS as a tool to evaluate the stability of oligonucleotide drugs after intracellular delivery, J. Control Release 2005, 103: 259‐271.

21. Elson E.S. and Magde D., Fluorescence correlation spectroscopy: I. Conceptual basis and theory, Biopolymers 1974, 13: 1‐27.

22. Fluorescence Correlation Spectroscopy: Theory and Applications (2001) (Rigler, R. and Elson, E.L., eds.), Springer‐Verlag Berlin‐Heidelberg, Germany.

23. Clegg R.M., Fluorescence resonance energy transfer and nucleic acids, Methods Enzymol. 1992, 211: 353‐388.

24. Selvin P.R., The renaissance of fluorescence resonance energy transfer, Nature Structural Biology 2000, 7: 730‐734.

25. Leonetti J.P., Mechti N., Degols G., Gagnor C. et al., Intracellular‐Distribution of Microinjected Antisense Oligonucleotides, Proceedings of the National Academy of Sciences of the United States of America 1991, 88: 2702‐2706.

26. Lorenz P., Misteli T., Baker B.F., Bennett C.F. et al., Nucleocytoplasmic shuttling: a novel in vivo property of antisense phosphorothioate oligodeoxynucleotides, Nucleic Acids Research 2000, 28: 582‐592.

81


27. Uhler S.A., Cai D., Man Y., Figge C. et al., RNA degradation in cell extracts: real‐time monitoring by fluorescence resonance energy transfer, J. AM. CHEM. SOC. 2003, 125: 14230‐14231.

28. Vitiello D., Pecchia D.B., and Burke J.M., Intracellular ribozyme‐catalyzed trans‐cleavage of RNA monitored by fluorescence resonance energy transfer, RNA 2000, 6: 628‐637.

29. Uchiyama H., Hirano K., Kashiwasake‐Jibu M., and Taira K., Detection of undegraded oligonucleotides in vivo by fluorescence resonance energy transfer. Nuclease activities in living sea urchin eggs, J. Biol. Chem. 1996, 271: 380‐384.

30. Srinivasan C., Siddiqui S., Silbart L.K., Papadimitrakopoulos F. et al., Dual fluorescent labeling method to visualize plasmid DNA degradation, Bioconjug. Chem. 2009, 20: 163‐169.

31. Modi S., M G S, Goswami D., Gupta G.D. et al., A DNA nanomachine that maps spatial and temporal pH changes inside living cells, Nat. Nanotechnol. 2009, 4: 325‐330.

32. Chiu Y.L., Dinesh C.U., Chu C.Y., Ali A. et al., Dissecting RNA‐interference pathway with small molecules, Chem. Biol. 2005, 12: 643‐648.

33. Jagannath A. and Wood M.J., Localization of double‐stranded small interfering RNA to cytoplasmic processing bodies is Ago2 dependent and results in up‐regulation of GW182 and Argonaute‐2, Mol. Biol. Cell 2009, 20: 521‐529.

34. Meseth U., Wohland T., Rigler R., and Vogel H., Resolution of fluorescence correlation measurements, Biophys. J. 1999, 76: 1619‐1631.

35. Remaut K., Lucas B., Raemdonck K., Braeckmans K. et al., Can we better understand the intracellular behavior of DNA nanoparticles by fluorescence correlation spectroscopy?, J. Control Release 2007, 121: 49‐63.

36. Schwille P., Fluorescence correlation spectroscopy and its potential for intracellular applications, Cell Biochem. Biophys. 2001, 34: 383‐408.

37. Brock R., Hink M.A., and Jovin T.M., Fluorescence correlation microscopy of cells in the presence of autofluorescence, Biophys. J. 1998, 75: 2547‐2557.

38. Zeng Y. and Cullen B.R., RNA interference in human cells is restricted to the cytoplasm, RNA. 2002, 8: 855‐860.

39. Chiu Y.L., Ali A., Chu C.Y., Cao H. et al., Visualizing a correlation between siRNA localization, cellular uptake, and RNAi in living cells, Chem. Biol. 2004, 11: 1165‐1175.

40. Robb G.B., Brown K.M., Khurana J., and Rana T.M., Specific and potent RNAi in the nucleus of human cells, Nat. Struct. Mol. Biol. 2005, 12: 133‐137.

41. Matzke M.A. and Birchler J.A., RNAi‐mediated pathways in the nucleus, Nature Reviews Genetics 2005, 6: 24‐35.

42. Ohrt T., Merkle D., Birkenfeld K., Echeverri C.J. et al., In situ fluorescence analysis demonstrates active siRNA exclusion from the nucleus by Exportin 5, Nucleic Acids Research 2006, 34: 1369‐1380.

43. Haupenthal J., Baehr C., Kiermayer S., Zeuzem S. et al., Inhibition of RNAse A family enzymes prevents degradation and loss of silencing activity of siRNAs in serum, Biochemical Pharmacology 2006, 71: 702‐710.

44. Sorrentino S., Naddeo M., Russo A., and D'Alessio G., Degradation of double‐stranded RNA by human pancreatic ribonuclease: Crucial role of noncatalytic basic amino acid residues, Biochemistry 2003, 42: 10182‐10190.

45. Deutscher M.P. and Li Z.W., Exoribonucleases and their multiple roles in RNA metabolism, Prog. Nucleic Acid Res. Mol. Biol. 2001, 67‐105.

46. Holen T., Amarzguioui M., Babaie E., and Prydz H., Similar behaviour of single‐strand and double‐strand siRNAs suggests they act through a common RNAi pathway, Nucleic Acids Res. 2003, 31: 2401‐2407.

47. Leuschner P.J.F., Ameres S.L., Kueng S., and Martinez J., Cleavage of the siRNA passenger strand during RISC assembly in human cells, EMBO Reports 2006, 7: 314‐320.

48. Schwarz D.S., Hutvagner G., Du T., Xu Z.S. et al., Asymmetry in the assembly of the RNAi enzyme complex, Cell 2003, 115: 199‐208.

49. Nykanen A., Haley B., and Zamore P.D., ATP requirements and small interfering RNA structure in the RNA interference pathway, Cell 2001, 107: 309‐321.

50. Harborth J., Elbashir S.M., Vandenburgh K., Manninga H. et al., Sequence, chemical, and structural variation of small interfering RNAs and short hairpin RNAs and the effect on mammalian gene silencing, Antisense & Nucleic Acid Drug Development 2003, 13: 83‐105.

51. Chiu Y.L. and Rana T.M., RNAi in human cells: Basic structural and functional features of small interfering RNA, Molecular Cell 2002, 10: 549‐561.

82


52. Martinez J., Patkaniowska A., Urlaub H., Luhrmann R. et al., Single‐stranded antisense siRNAs guide target RNA cleavage in RNAi, Cell 2002, 110: 563‐574.

53. Bertrand J.R., Pottier M., Vekris A., Opolon P. et al., Comparison of antisense oligonucleotides and siRNAs in cell culture and in vivo, Biochemical and Biophysical Research Communications 2002, 296: 1000‐1004.

54. Bartlett D.W. and Davis M.E., Effect of siRNA nuclease stability on the in vitro and in vivo kinetics of siRNA‐mediated gene silencing, Biotechnol. Bioeng. 2007, 97: 909‐921.

83


84

dextran microgels for time‐controlled delivery of siRNA | CHAPTER 3

CHAPTER 3 | Dextran microgels for time‐

controlled delivery of siRNA


Koen Raemdonck1, Tinneke G. Van Thienen1, Roosmarijn E. Vandenbroucke1, Niek N. Sanders1,2,

Joseph Demeester1 and Stefaan C. De Smedt1, Advanced Functional Materials 2008, 18: 993‐1001.



2Laboratory of Gene Therapy, Department of Nutrition, Genetics and Ethology, Faculty of Veterinary Medicine, Ghent


85

CHAPTER 3 | dextran microgels for time‐controlled delivery of siRNA

86

ABSTRACT

To apply siRNA as a therapeutic agent, appropriate attention should be paid to the optimization of

the siRNA gene silencing effect, both in terms of magnitude and duration. Intracellular time‐

controlled siRNA delivery could aid in tailoring the kinetics of siRNA gene knockdown. However,

materials with easily tunable siRNA release properties have not been subjected to thorough

investigation thus far. This report describes cationic biodegradable dextran microgels which can be

loaded with siRNA posterior to gel formation. Even though the siRNAs are incorporated in the

hydrogel network based on electrostatic interaction, still a time‐controlled release can be achieved

by varying the initial network density of the microgels. To demonstrate the biological functionality of

the siRNA loaded gels, we studied their cellular internalization and enhanced green fluorescent

protein (EGFP) gene silencing potential in Huh‐7 human hepatoma cells.



Chapter 3 provides:

• details on the formulation of biodegradable dextran hydrogel particles based on

methacrylate functionalized dextran

• information on the applicability of dextran microgels for siRNA encapsulation and siRNA

release in a degradation‐controlled manner

• a first indication that siRNA loaded dextran hydrogel particles show promise towards gene

silencing via the process of RNAi

Key terms:

Microgel – dextran – cationic – siRNA – controlled release – biodegradation – EGFP

INTRODUCTION

Since its discovery in C. elegans [1] and the confirmation of its applicability in mammalian cells,[2] RNA

interference (RNAi) has become the method of choice to suppress gene expression for therapeutic

intervention.[3] RNAi comprises a complex, evolutionary conserved pathway, where ~21 bp RNA

duplexes (termed small interfering RNAs or siRNAs) function as the effector molecules for sequence

specific downregulation of gene expression.[4] Although siRNAs are considered to be more powerful

when compared to antisense oligonucleotides,[2,5] the same barriers exist for successful in vivo

application.[6,7] Indeed, small interfering RNAs have already demonstrated their in vivo potential as

therapeutic agents,[8] but there still is a great need for efficient delivery vehicles that can safely guide

their siRNA payload to the desired target tissue.[3,4,7] Recent reviews critically discuss therapeutic

siRNA delivery, which is still regarded as the main obstacle towards clinical application of siRNA.

87


Several strategies for in vivo siRNA delivery have already been proposed, including the application of

siRNA bioconjugates or the incorporation of siRNA into multifunctional polymeric and lipidic

nanodevices, which are also briefly discussed in chapter 1.[4,7‐11]

To apply siRNA as a therapeutic, both the intensity and duration of the gene silencing effect will have

to be controlled and optimized according to the selected disease target. The kinetics of gene

silencing by (unmodified and nuclease stabilized) siRNAs has been recently described in both an in

vitro and an in vivo experimental setting by Bartlett et al.. [12,13] In fast dividing cells, the transient

gene silencing is mainly due to the intracellular dilution of siRNA and activated RISC (RNA Induced

Silencing Complex) upon cell division,[12] while the intracellular siRNA stability has little or no

influence on the magnitude and duration of the gene knockdown effect once the siRNA has reached

the cytosol.[13]

We anticipate that an intracellular time‐controlled release of siRNA could be an interesting approach

to maintain the cytosolic siRNA concentration above a critical threshold for a longer period of time.

The packaging of siRNA in carriers that shield the siRNA from its direct environment and slowly

releases it into the cytosol could be advantageous to prolong their silencing effect. Moreover,

adverse effects originating from saturation of the RNAi pathway or unwanted immune activation may

be minimized when obtaining better control over the intracellular siRNA concentrations. To achieve

such goals it is evident that carriers with ‘flexible’ siRNA release properties are needed. However, the

literature to date contains only few reports on carriers which should be able to deliver siRNA

intracellularly in a time‐controlled manner.[14] Several papers describe the cytosolic release of siRNA

based on an intracellular trigger,[15‐17] but detailed studies on intracellular time‐controlled siRNA

delivery based on the characteristics of the siRNA carrier are not yet available at present.

The siRNA carrier that is envisioned in the present chapter is hydrogel based. Hydrogels can be

defined as three‐dimensional, hydrophilic matrices that are able to retain large amounts of water or

biological fluid.[18] In general, as a consequence of this specific feature, hydrogels can be regarded as

biocompatible and are therefore suitable for a wide range of pharmaceutical and biomedical

applications.[19] In particular, hydrogels offer excellent opportunities for development of controlled

drug delivery devices for different types of drugs.[18‐21] Our research is particularly directed towards

biodegradable hydrogels for controlled nucleic acid delivery. In this application, the ability to tailor

the drug release time depending on the physicochemical properties of the hydrogel matrix will be the

main focus.

This chapter reports on the preparation of siRNA loaded (bio)degradable microgels based on dextran.

Dextran is one the most promising polysaccharides, produced either chemically or by bacteria from

88


sucrose, for the preparation of hydrogels for controlled drug delivery purposes.[20,22] Bacterial dextran

mainly consists of linear α‐1,6‐linked glucopyranose units, with some degree of 1,3‐branching.

Several coupling reactions have already been described for the modification of the numerous OH‐

groups present along the dextran chain.[22]

We especially wished to know whether the composition of the dextran gel network allows controlling

the release kinetics of the loaded siRNA. We also describe the collection of smaller sized microgels

with the aim to conduct a series of preliminary cell culture experiments. We verified their cellular

uptake in a human hepatoma cell line (Huh‐7) with confocal microscopy. Finally, the gene silencing

potential of these microgels, loaded with siRNA targeted against enhanced green fluorescent protein

(EGFP) is demonstrated in Huh‐7 cells that stably express the EGFP gene (Huh‐7_EGFP cells).

EXPERIMENTAL SECTION

siRNA oligonucleotides.

Twenty‐one nucleotide siRNA duplexes, targeted against enhanced green fluorescent protein

(siEGFP), were purchased from Dharmacon (Lafayette, CO). siCONTROL is a negative control

sequence, synthesized by Eurogentec (Seraing, Belgium), which does not match to any known

sequence in the human genome. siRNA batches were diluted in RNase free water to a final

concentration of 20‐50 µM and stored at ‐85°C. For fluorescence experiments, a siCONTROL duplex

5’‐labeled (sense strand) with atto488 or atto647 dye was used. Fluorescent modifications were

done by Eurogentec.

Synthesis of dex‐HEMA and dex‐MA.

Dextran hydroxyethyl methacrylate (dex‐HEMA; Figure 3.1a) and dextran methacrylate (dex‐MA;

Figure 3.1a) were prepared and characterized according to a method described previously.[23]

Dextran with a number average molecular weight of 19 kDa was used. The degree of substitution

(DS, i.e. the number of HEMA (or MA) groups per 100 glucopyranose residues of dextran) was

determined by proton nuclear magnetic resonance spectroscopy (1H NMR) in D2O with a Gemini 300

spectrometer (Varian). As reported before, the carbonate ester in dex‐HEMA (Figure 3.1a) hydrolyzes

at 37°C, pH 7.4 (further referred to as physiological conditions).

89


Preparation of dextran microgels.

To prepare dex‐HEMA and dex‐MA microgels, an emulsion photopolymerization method in mineral

oil was applied. A dex‐HEMA (or dex‐MA) solution (20% w/w) with ~0.1% (w/v) photoinitiator

(irgacure 2959, Ciba Specialty Chemicals) was briefly flushed with N2 and subsequently emulsified in

mineral oil (Sigma) supplemented with 10% (v/v) of surfactant (ABIL EM 90, Goldschmidt) by

vortexing (2 min). To obtain cationic microgels, 2‐(dimethylamino)ethyl methacrylate (DMAEMA, pKa

8.4, Figure 3.1) was added to the dex‐HEMA (or dex‐MA) solution. Within 60 s after emulsification,

the emulsion droplets were polymerized by UV irradiation (450 s) at room temperature (Bluepoint

2.1 UV source, Dr. Hönle UV technology). Subsequently, the continuous oil phase was diluted in

hexane (HPLC grade), followed by short centrifugation (525 g, 5 min) to collect the dextran microgels.

The supernatant was discarded and the washing step was repeated three consecutive times with an

excess of hexane before redispersing the dextran microgels in 20 mM N‐2‐hydroxyethylpiperazine‐

N’‐2‐ethanesulfonic acid (HEPES) buffer (adjusted to pH 7.4 with 1 N NaOH) or in RNase free water.

Remaining hexane was removed by vacuum evaporation. Contrary to dex‐HEMA hydrogels,

hydrogels prepared from dex‐MA are essentially non‐degradable at physiological pH as they lack the

presence of a labile carbonate ester bond.[24,25]

Penetration of siRNA into dextran microgels.

Neutral and cationic dextran microgels were prepared as described above. Intact dex‐HEMA

microgels were immersed in a 2 µM solution of atto488‐siRNA (24 h at 4°C) or a 5 µM solution of

carboxyfluorescein (CaF) (10 min at room temperature) and subsequently confocal micrographs were

taken. Confocal time lapse microscopy (1 frame/25 s) was applied to visualize siRNA penetration in

degrading neutral dextran microgels, following addition of 1 µL of NaOH (0.05 M) to 5 µL of the

siRNA‐microgel dispersion. Non‐degradable dex‐MA microgels were used as a control. To investigate

the interaction of siRNA with cationic gels, dex‐HEMA‐co‐DMAEMA microgels were incubated with

increasing atto488‐siRNA concentrations and subsequently visualized with confocal microscopy.

siRNA release measurements from cationic dextran microgels.

The dextran microgels used in the siRNA release experiments were prepared as described below, this

time using a water‐in‐water emulsion method based on the immiscibility of aqueous solutions of

dextran and PEG. Briefly, 0.025 g dex‐HEMA (or dex‐MA), 0.012 g dextran (19 kDa) and 25 µL

DMAEMA were dissolved in 2.167 mL RNase free water and emulsified in 2.7 mL of a 30% (w/w)

aqueous polyethyleneglycol (PEG; 20 kDa) solution by vortexing (2 min).

90


FIGURE 3.1 | (a) Chemical structures of dextran hydroxyethyl methacrylate (dex‐HEMA), dextran methacrylate (dex‐MA) and 2‐(dimethylamino)ethyl methacrylate (DMAEMA). (b) Radical polymerization of dex‐HEMA yields a 3D hydrogel network. Other methacrylate monomers (e.g. DMAEMA) can be copolymerized with dex‐HEMA and in this way cationic charges can be incorporated. Dex‐HEMA hydrogels are biodegradable due to the hydrolysis of carbonate esters (indicated by the arrows) at physiological conditions. Contrary to dex‐HEMA hydrogels, dex‐MA hydrogels are essentially non‐degradable under physiological conditions since they lack the presence of this labile ester bond.

91


Radical polymerization of the dextran/dex‐HEMA (or dextran/dex‐MA) droplets was initiated by 100

µL of a 20% (v/v) N,N,N’,N’‐tetramethylethylenediamine (TEMED) solution (neutralized with 2N HCl)

and 180 µL potassium peroxodisulphate (KPS, 50 mg mL‐1). The reaction mixture was left to

polymerize at room temperature for at least 30 minutes. Microgels of lower crosslink density were

obtained by decreasing the amount of polymerizable dextran while keeping the total amount of

dextran constant (0.0185 g dex‐HEMA (or dex‐MA) + 0.0185 g dextran). To remove remaining KPS,

TEMED, PEG and unpolymerized DMAEMA, the microgels were washed three times with RNase free

water, collected through centrifugation (750 g, 5 min) and finally redispersed in 5 mL HEPES buffer.

To load the microgels with siRNA, a 0.9 mL aliquot of the microgel dispersion was incubated with 100

µL siRNA (50 µM) during 30 min at room temperature in non‐stick RNase free 2 mL eppendorfs

(Ambion). Subsequently, the samples were kept at 37°C under continuous shaking. At defined time

points, the microgels were pelleted down by centrifugation (750g, 5 min). A 50 µL aliquot of the

supernatant was then withdrawn and replaced by 50 µL fresh HEPES buffer. The siRNA concentration

in each aliquot was quantified by measuring the absorbance at 260 nm (Nanodrop ND1000

Spectrophotometer) or with the Quant‐it™ Ribogreen® assay (Molecular Probes). It was verified that

the absorbance and fluorescence values lied within the linear range of detection. The siRNA

concentration measured when the microgels were completely degraded (as verified through optical

microscopy) was set to 100% in the release graphs.

Collection of smaller sized microgels.

Microgels were prepared as described in the previous paragraph. After polymerization of the

microgels, the dispersion (~5 mL) was diluted 1:3 with RNase free water and centrifuged during 30

min at 1000 g. The supernatant was collected and smaller sized microgels were obtained through a

second centrifugation (12000 g, 10 min). The microgels were washed three times with an excess of

RNase free water to remove any impurities and finally redispersed in 667 µL of HEPES buffer. The size

distribution of the microgels was measured by fluorescence microscopy. Therefore, fluorescein O‐

methacrylate (Sigma) was copolymerized in the microgels. Fluorescence images at 488 nm excitation

were recorded with a SPOT Insight CCD camera (Diagnostic Instruments, Sterling Heigths, MI)

attached to a Nikon TE300 inverted epi‐fluorescence microscope with a Nikon Plan Fluor 20x

objective lens. To perform accurate size measurements, the optical system was calibrated using a

stage grid consisting of 10 µm by 10 µm squares (Edmund Optics, York, UK). The fluorescence images

were analyzed with custom built software to obtain size distributions of the microgels.

92


Scanning electron microscopy (SEM).

Scanning electron microscopy was carried out using a Quanta 200F FEG (Field Emission Gun) SEM (FEI

company) operating in High VacMode at an accelerating voltage of 25 kV. For sample preparation, a

drop of the microgel dispersion was dried under low vacuum on a silica wafer and sputtered with

gold/palladium before visualization.

Atomic Force Microscopy (AFM).

Atomic Force Microscopy images in air were acquired using a PicoPlus instrument (Molecular

Imaging) in tapping mode at room temperature. For sample preparation, a drop of the microgel

dispersion was left to dry overnight on a silica wafer. Height images were flattened for visualization

using WSxM scanning probe microscopy software (Nanotec Electronica).

Confocal Laser Scanning Microscopy.

Confocal micrographs of the dex‐HEMA and dex‐HEMA‐co‐DMAEMA microgels incubated with

atto488‐siRNA were taken using a Nikon EZC1‐si confocal scanning module installed on a motorized

Nikon TE2000‐E inverted microscope (Nikon Benelux, Brussels, Belgium), equipped with an Argon ion

laser (488 nm) and a 60x water‐immersion objective lens (Plan Apo 60x, numerical aperture 1.2,

collar rim correction, Nikon).

Cellular uptake of dextran microgels in Huh‐7 cells.

The human hepatoma cell line Huh‐7 was cultured (37°C, 5% CO2) in DMEM:F12, supplemented with

2 mM glutamine, 10% heat‐inactivated Fetal Bovine Serum (FBS) and 100 U mL‐1 penicillin‐

streptomycin. Huh‐7 cells were seeded in sterile glass‐bottomed culture dishes (MatTek Corporation,

Ashland, MA) at a density of 3 x 104 cells per cm2 and allowed to attach overnight. The cells were

incubated (37°C, 5% CO2) with 50 µL of a dex‐HEMA‐co‐DMAEMA microgel dispersion (18.5 mg

dextran, DS 2.5), loaded with atto647‐siRNA, in a total volume of 250 µL OptiMEM (Invitrogen,

Merelbeke, Belgium) during 20 h (37°C). Subsequently, cells were washed with Phosphate Buffered

Saline (PBS) and prewarmed culture medium was added. The intracellular distribution of the

microgels was visualized with a confocal microscope (see previous section). For lysosomal staining,

cells were incubated for 15 min with 1:104 diluted Lysotracker Green® (Molecular Probes) prior to

image recording. Images were captured with a Nikon Plan Apochromat 60x oil immersion objective

(numerical aperture 1.4) using a 488 nm and 639 nm laser line for the excitation of Lysotracker

Green® and atto647‐siRNA respectively.

93


EGFP silencing in Huh‐7_EGFP cells by siRNA loaded dextran microgels.

Huh‐7_EGFP cells stably expressing EGFP were generated by transfecting Huh‐7 cells with the pEGFP‐

N2 plasmid (Clontech, Palo Alto, USA). The pDNA was linearized using the restriction enzyme XbaI

and transfected by complexation with linear polyethylenimine (PEI) 22 kDa. Transfected cells were

incubated in fresh medium for 48 h and then selected with 400 µg mL‐1 G418 (geneticine, Invitrogen).

After two weeks, clones were isolated and expanded. Subsequently, the generated clones were

analyzed by CLSM and a 100% EGFP positive clone was selected. Huh‐7_EGFP cells and Huh‐7 cells

were cultured under identical conditions. To assess the EGFP gene silencing potential of siRNA loaded

dextran microgels, Huh‐7_EGFP cells were seeded at a density of 3 x 104 cells per cm2 in Lab‐Tek™ II

chambered coverglasses (Nalge Nunc International, Rochester, NY). Two days after seeding, cells

were washed with PBS and 30 µL dex‐HEMA‐co‐DMAEMA microgel dispersion (18.5 mg dex‐HEMA,

DS 2.5), loaded with 0.2 nmol siEGFP or siCONTROL, was added per well in a total volume of 400 µL

OptiMEM. After 20 h incubation (37°C, 5% CO2), the transfection medium was removed and replaced

with complete cell culture medium. Cells were left to grow to 100% confluence (~48 h).

Subsequently, the cells were trypsinized and replated (1:3 diluted in culture medium) in a glass‐

bottomed 96‐well plate (Greiner Bio‐One). Huh‐7_EGFP cells were visualized 8 days after transfection

with confocal microscopy using a Nikon Plan Fluor 20x objective (numerical aperture 0.5). The 488

nm laser line was selected for EGFP excitation.

RESULTS AND DISCUSSION

Interaction of siRNA with neutral dextran microgels

Dextran microgels were prepared by emulsion photopolymerization of dextran hydroxyethyl

methacrylate (dex‐HEMA, Figure 3.1a) in mineral oil. In this way, a microscopic gel network is

obtained, consisting of dextran polymer chains coupled by inter‐ and intramolecular

oligomethacrylate crosslinks.[23,25] First, we aimed to investigate to which extent siRNA can diffuse

into intact neutral dextran hydrogel networks with variable crosslinking density. The microgel

crosslink density can be altered by varying the concentration of methacrylated dextran or by using

dex‐HEMA with a different degree of substitution (DS, i.e. the number of HEMA groups per 100

glucopyranose residues of dextran). In the following experiment, dex‐HEMA microgels (20% w/w)

were prepared with different DS values. The microgels were incubated with green fluorescent siRNA

(atto488‐siRNA) during 24 h at 4°C (to avoid excessive hydrolysis of the carbonate esters) and the

samples were analyzed by confocal microscopy. As shown in the left and middle image of Figure 3.2,

94


siRNA did not significantly penetrate into loosely crosslinked dex‐HEMA microgels (DS 3.2). In

contrast, as illustrated in Figure 3.2 (right image), carboxyfluorescein (CaF) immediately diffused into

the network of even highly crosslinked (DS 21) dex‐HEMA microgels. These results suggest (a) that

siRNA molecules do not show any affinity for neutral dex‐HEMA networks and (b) that they are

sterically excluded from intact neutral dex‐HEMA microgels.

FIGURE 3.2 | Confocal images of dex‐HEMA microgels (DS 3.2 and DS 21, 20% w/w) immersed in an atto488‐siRNA solution (left, middle) and dex‐HEMA microgels (DS 21, 20% w/w) incubated in a carboxyfluorescein solution (CaF, right). Scale bar = 20 µm.

In a next approach we investigated whether siRNA penetrates into degrading neutral dextran

hydrogel networks. As reported before, dex‐HEMA networks are biodegradable by virtue of the

carbonate esters that link the polymerized methacrylate groups to the dextran backbone (Figure

3.1b). Hydrolysis of these ester linkages occurs under physiological conditions (37°C, HEPES buffer,

pH 7.4), thereby yielding both the original dextran chains and HEMA oligomers.[23‐25] Figure 3.3a

shows confocal images of degrading dex‐HEMA microgels immersed in an atto488‐siRNA solution. To

accelerate the hydrolysis of the crosslinks, the experiment was conducted at alkaline pH. As a

function of time we observed a clear swelling of the microgels due to degradation of the gel network.

When a fraction of the carbonate esters is broken, the distance between two consecutive crosslinks

increases, resulting in a larger pore size and higher water influx.[25,26] After 175 s, fluorescence started

to appear in the microgel core, indicating the infiltration of siRNA. As a control, the experiment in

Figure 3.3a was repeated with non‐degradable dex‐MA microgels of comparable crosslink density.

Note that in contrast to dex‐HEMA, dex‐MA hydrogels do not degrade under physiological conditions

as the methacrylate ester in the crosslinks is hydrolytically stable under the experimental

conditions.[23‐25] As expected, the dex‐MA microgels did not swell and did not show any significant

siRNA penetration (Figure 3.3b). All together, Figures 3.3a and 3.3b support the conclusion that

degradation of the dex‐HEMA crosslinks sufficiently enlarges the network pores in dex‐HEMA gels

95


thereby lifting sterical hindrance and allowing the diffusion of siRNA molecules. Importantly, from

these experiments we may assume that if siRNA molecules would be incorporated inside intact dex‐

HEMA microgels during microgel formation, initially they would remain physically entrapped in the

dextran network and subsequently released upon network degradation.

FIGURE 3.3 | (a) Confocal snapshots of a degrading dex‐HEMA microgel (DS 8.9, 20% w/w) dispersed in an atto488‐siRNA solution at ambient temperature. Scale bar = 10 µm. (b) Confocal snapshots of a (non‐degradable) dex‐MA microgel (DS 9.0, 20% w/w) dispersed in an atto488‐siRNA solution at ambient temperature. Scale bar = 5 µm.

96


In the experiments described above, the dextran microgels were prepared by emulsion

polymerization in mineral oil. As siRNA is not soluble in the continuous phase we anticipate a high

loading efficiency when the siRNA is added to the dextran phase, prior to emulsification. However,

this preparation protocol implies that the siRNA molecules would be brought into contact with free

radicals and UV light during the photopolymerization step. To avoid this, we sought for an alternative

preparation protocol which allows (a) high siRNA loading of the dextran microgels and (b) siRNA

loading of the microgels after their preparation.

Penetration of siRNA into cationic dextran microgels

Cationic microgels were prepared by copolymerization of 2‐(dimethylamino)ethyl methacrylate

(DMAEMA, Figure 3.1a, pKa 8.4) with dex‐HEMA or dex‐MA.[27,28] The incorporation of positively

charged groups in the microgel network was verified by measuring the zeta‐potential of the

microgels, which was indeed positive (~20‐30 mV) when DMAEMA was added during microgel

preparation.

Figure 3.4 shows confocal fluorescence images of cationic dextran microgels which were incubated

with increasing amounts of atto488‐siRNA. When the microgels were incubated with a low amount

of siRNA (Figure 3.4a), a bright fluorescent ring was observed, suggesting the selective binding of

negatively charged siRNAs to the cationic DMAEMA groups near the surface of the microgels.

However, in more concentrated siRNA solutions (Figures 3.4b and 3.4c), not only the surface but also

the interior of the microgels became loaded with siRNA. It seems that the siRNA molecules first

neutralize the charges at the surface of the microgels after which remaining siRNA molecules are

able to penetrate deeper into the microgel network. When a sufficient amount of siRNA was added,

a homogeneous loading of the cationic dextran microgels could be achieved (Figure 3.4c and 3.4f).

Interestingly, while the pores in neutral dextran gels (DS 3.2, 20% w/w) are too small to allow siRNA

diffusion into the network (Figure 3.2), the incorporation of cationic groups in these gels seems to aid

the diffusion (Figure 3.4), thereby loading the microgels with siRNA based on electrostatic attraction.

The exact reason for this remains unclear. Possibly, the incorporation of protonable amines into the

gel network leads to electrostatic repulsion between positive charges which may expand the pores in

the network. For this reason, a direct comparison between microgels composed of dex‐HEMA alone

and dex‐HEMA microgels with copolymerization of DMAEMA is not feasible.

The previous experiments demonstrate that loading of positively charged dextran‐co‐DMAEMA gels

with siRNA after gel formation (“post‐loading”) is possible. Post‐loading of siRNA may have major

advantages as (1) exposure of siRNA to UV light and/or free radicals during the gelation process is

97


avoided and (2) applying a sufficient amount of cationic groups in the microgels can result in a high

encapsulation efficiency of the siRNA.

FIGURE 3.4 | (a‐c) Confocal fluorescence images of cationic dextran microgels (DS 3.2, 20% w/w, molar ratio HEMA/DMAEMA of 0.5) incubated in a solution containing increasing amounts of atto488‐siRNA. Scale bar = 20 µm. (d‐f) Corresponding fluorescence intensity profiles, plotted along a microgel cross‐section indicated by the dotted line in the confocal fluorescence micrographs.

Controlled release of siRNA from cationic dextran microgels

Figure 3.5 shows the cumulative release of siRNA under physiological conditions from respectively

lowly (~DS 3) and highly crosslinked (~DS 9) cationic dex‐HEMA‐co‐DMAEMA microgels. Also the

siRNA release from non‐degradable dex‐MA‐co‐DMAEMA microgels was evaluated. The release

profile of siRNA from the lowly and highly crosslinked dex‐HEMA‐co‐DMAEMA microgels differs

significantly. Lowly crosslinked cationic dex‐HEMA microgels show a small burst release of siRNA (<

10%) after which siRNA is continuously released until the microgels are completely degraded (after

~6 days of incubation). Non‐degradable lowly crosslinked cationic dex‐MA microgels however do not

98


release the encapsulated siRNA. In contrast, the siRNA release from highly crosslinked cationic dex‐

HEMA microgels shows a sustained profile after an initial lag time of about two days. Where the

lowly crosslinked cationic dex‐HEMA microgels have already released 50% of the loaded siRNA after

~3 days of incubation, the highly crosslinked microgels need almost 9 days.

FIGURE 3.5 | Cumulative release of siRNA from (degradable) dex‐HEMA‐co‐DMAEMA and (non‐degradable) dex‐MA‐co‐DMAEMA microgels with high (~DS 9) and low (~DS 3) crosslink density. The microgels were incubated at 37°C, pH 7.4. In each sample preparation 25 mg dex‐HEMA (dex‐MA) was used (see experimental section). The sample aliquots were measured in triplicate with Quant‐iT™ RiboGreen® assay reagent (Molecular Probes).

It has been shown previously that dex‐HEMA hydrogels are suited for the controlled release of

proteins.[22,29,30] Due to hydrolysis of the crosslinks in dex‐HEMA gels, the pores in the network

enlarge which eventually results in the release of the protein molecules which were physically

entrapped before degradation of the network. It was shown that varying the initial crosslink density

of dex‐HEMA hydrogels (by varying the DS and/or the dex‐HEMA content of the hydrogels) influences

the protein release. In gels with a higher network density, indeed more crosslinks need to be

degraded before the hydrogel mesh size exceeds the hydrodynamic diameter of the entrapped

protein.[31]

In this study, the spontaneous uptake of siRNA in the cationic microgels is mainly based on

electrostatic interactions between the negatively charged siRNA molecules and the positively

charged microgel network. Figure 3.5 clearly proves that even though the siRNA molecules are

electrostatically incorporated (and not sterically entrapped) in the dextran gel, the siRNA release is

still governed by the degradation kinetics of the dextran network (as illustrated in Figure 3.6).

99


FIGURE 3.6 | (left) Consecutive snapshots of a degrading cationic microgel sample, where the hydrogel particles are loaded with atto647‐siRNA. (right) Schematic overview of dex‐HEMA microgel degradation. Hydrolysis of the carbonate esters in HEMA crosslinks of cationic dex‐HEMA‐co‐DMAEMA microgels results in a swelling of the gelparticles accompanied by a release of free dextran, HEMA‐co‐DMAEMA oligomers and complexed siRNA.

Thus, importantly, by varying the crosslink density of the hydrogels, the onset of siRNA release and

the release kinetics of the encapsulated siRNA can be easily controlled. Moreover, as the siRNA

molecules are electrostatically bound to the hydrogel network rather than sterically entrapped in the

100


network, also siRNA release from loosely crosslinked gels is feasible (without the risk of siRNA

leakage due to diffusion) which may be of interest when a faster release is required.

Smaller sized dextran microgels for intracellular siRNA delivery

As the biological targets of siRNA are located intracellularly,[10] it is evident that siRNA molecules have

to be carried by and released from dextran gel particles which should be able to cross the cellular

membrane barrier. The average size of the dextran microgels used in Figures 3.2 till 3.5 was 7‐8 µm

(data not shown), which is too large for efficient intracellular uptake. To analyze if siRNA loaded

dextran hydrogels are suitable for intracellular siRNA delivery we therefore collected smaller sized

microgels through centrifugation (see experimental section). Figure 3.7a shows a fluorescence image

of the thus obtained microgels which have a mean diameter between 2 and 3 µm (Figure 3.7b).

FIGURE 3.7 | (a) Fluorescence micrograph of smaller sized microgels (mean diameter of 2‐3 µm). Microgels were fluorescently tagged by copolymerization with fluorescein methacrylate. Scale bar = 20 µm. (b) Corresponding size distribution as measured with fluorescence microscopy (12 x 103 particles were analyzed).

The cellular internalization of several types of particles with comparable size has already been

evaluated in the literature for the intracellular delivery of macromolecules or for the application as

biomarkers, also in non‐phagocytic cells.[32‐36] In this chapter, we preferred to evaluate the cellular

uptake and biological functionality of microgels with a lower crosslink density and, consequently, a

relatively faster siRNA release. Therefore, the amount of dex‐HEMA used during the preparation of

the microgels was decreased with ~25% (see experimental section). As proven in Figure 3.8a,

microgels can be produced which display maximal release of the encapsulated siRNA in less than 3

days under physiological conditions. Figure 3.8b shows a SEM and AFM image (graph inset) of dex‐

MA microgels (DS 3.1). These images clearly underline the loosely crosslinked nature of the gels as

they tend to spread out on a flat surface. This behaviour was not observed on dextran gels with a

101


higher density (data not shown). Also note the scale bar on the AFM images, showing a high

concentration of particles < 2 µm in diameter.

FIGURE 3.8 | (a) Cumulative release (%) of siRNA from (degradable) dex‐HEMA‐co‐DMAEMA and (non‐degradable) dex‐MA‐co‐DMAEMA microgels with different crosslink density. In each sample preparation 18.5 mg dex‐HEMA was used (see experimental section). The data points are interconnected with a B‐spline (Origin 5.0 software). (b) Scanning electron micrograph of dex‐MA‐co‐DMAEMA microgels (DS 3.1). Graph inset displays AFM images obtained on the same sample.

When presented to living (human hepatoma) cells, siRNA loaded dex‐HEMA‐co‐DMAEMA microgels

(<3µm) were taken up in large numbers, as depicted in Figure 3.9. Confocal scanning along the z‐axis

proved that the microgels were effectively internalized by the cells and were not merely bound to

the cell surface (data not shown). It seems that some of the microgels show colocalization with

Lysotracker Green®, an endolysosomal marker, indicating their presence inside acidified vesicles

(yellow colour in Figure 3.9c). Nonetheless, also a substantial fraction of the microgels appears to

reside in other cytosolic compartments as they do not show this colocalization pattern. The uptake

mechanism of large particles (>1 µm) in the Huh‐7 hepatoma cell line was found to be mediated by a

clathrin‐ and caveolae/raft‐ independent pathway.[37] However, the different nature of the particles

described in this report and their high polydispersity makes it very challenging to clarify the exact

mechanism of internalisation.

102


FIGURE 3.9 | Confocal images (a‐c) and corresponding transmission image (d) of siRNA loaded dex‐HEMA‐co‐DMAEMA microgels (DS 2.5) in Huh‐7 hepatoma cells. The microgels are red fluorescent as they are loaded with atto647‐siRNA. Endolysosomes appear green by staining with Lysotracker Green®. Image (c) displays the overlay of red and green fluorescence. Yellow colour originates from green and red colocalization. Graph insets (bottom left, scale bar = 20 µm) show images in more detail.

EGFP silencing with siRNA loaded dextran microgels

To investigate the biological activity of the siRNA loaded dextran microgels, a gene silencing

experiment was conducted. Microgels (mean diameter of 2‐3 µm, DS 2.5) were loaded with control

siRNA (siCONTROL) or siRNA targeted against enhanced green fluorescent protein (siEGFP) and

incubated with Huh‐7 hepatoma cells stably expressing the EGFP gene (Huh‐7_EGFP cells). The Huh‐

7_EGFP cells were imaged with confocal microscopy approximately 8 days after transfection under

identical conditions. The results shown in Figure 3.10 indicate that the green fluorescent signal is

significantly weaker for the cells which were transfected with siEGFP‐microgels, when compared with

the cells which were incubated with siCONTROL‐microgels. In both experiments a comparable

number of cells was imaged. Consistent with the observations made in Figure 3.9, these results

103


suggest that siRNA loaded dextran microgels can pass the cellular barrier and release active siRNA

into the cell cytoplasm, leading to suppressed EGFP expression.

FIGURE 3.10 | EGFP expression in Huh‐7_EGFP cells transfected with dex‐HEMA‐co‐DMAEMA microgels (DS 2.5), loaded with respectively negative control siRNA (siCONTROL, left image) or siRNA targeting the EGFP gene (siEGFP, right image). Both images were captured under identical settings of the confocal microscope. Scale bar = 100 µm.

We would like to emphasize that there is room for further improvement of the biological activity of

the siRNA loaded dex‐HEMA microgels described in this study. Upon cellular internalization, dex‐

HEMA gelparticles will reside in endocytic vesicles and we may assume that a substantial fraction of

those particles will eventually end up in the degradative lysosomal pathway. More than likely,

modifications or delivery techniques leading to enhanced endosomal escape will be of vital

importance. From literature data we can expect slower dex‐HEMA degradation and consequently a

slower siRNA release at the lower pH values which can be found in late endosomes and

lysosomes.[24,37] We anticipate that this will have a significant influence on the eventual outcome of

the gene expression level in a knockdown experiment. Nonetheless, the data presented in Figure 3.9

and Figure 3.10 prove to be encouraging towards the application of cationic dextran microgels for

intracellular time‐controlled siRNA release and subsequent knockdown of a target gene. The

following chapters in this thesis, with the emphasis on chapters 5 to 7, aim to optimize the

preparation of (cationic) hydrogel particles (e.g. downscaling from micro‐ to nanoscopic dimensions)

and provide a more detailed investigation on their intracellular behaviour.

104


CONCLUSIONS

The results presented here indicate that cationic biodegradable dextran microgels can be produced

through emulsion polymerization and that they can be subsequently loaded with siRNA. The uptake

of siRNA by the microgels is mainly based on electrostatic interaction. This enables the ‘post‐loading’

of siRNA in preformed cationic gels, which offers several advantages: (1) the siRNA integrity is not

affected by the presence of free radicals during the emulsion polymerization, (2) depending on the

number of cationic groups incorporated in the gel network, a high siRNA encapsulation efficiency can

be achieved and (3) loosely crosslinked gels can be loaded with siRNA, without unwanted leakage of

siRNA due to diffusion. Irrespective of the electrostatic nature of the siRNA incorporation in the

microgel network, the siRNA release is still governed by the degradation of the crosslinks in the gel

network. Using a Huh‐7 cell line that stably expresses the EGFP gene, we were able to demonstrate

the biological functionality of the siRNA loaded microgels. We would like to remark that the siRNA

microgels show promise to investigate the influence of siRNA release kinetics on both the magnitude

and duration of gene knockdown, which has not been studied so far.

ACKNOWLEDGMENTS

Financial support from IWT (Institute for the Promotion of Innovation through Science and

Technology in Flanders) and FWO‐Flanders (Fund for Scientific Research‐Flanders) is acknowledged

with gratitude. Prof. Dr. K. Braeckmans is thanked for his help with the size distribution

measurements. Olivier Janssens and Karel Lambert are acknowledged for conducting the SEM and

AFM measurements respectively. The EU (FP6) is acknowledged for financially supporting MediTrans,

an Integrated Project on Targeted Delivery of Nanomedicines.

105


REFERENCE LIST

1. Fire A., Xu S.Q., Montgomery M.K., Kostas S.A. et al., Potent and specific genetic interference by double‐stranded RNA in Caenorhabditis elegans, Nature 1998, 391: 806‐811.


3. Bumcrot D., Manoharan M., Koteliansky V., and Sah D.W.Y., RNAi therapeutics: a potential new class of pharmaceutical drugs, Nature Chemical Biology 2006, 2: 711‐719.

4. Kim D.H. and Rossi J.J., Strategies for silencing human disease using RNA interference, Nat. Rev. Genet. 2007, 8: 173‐184.

5. Bertrand J.R., Pottier M., Vekris A., Opolon P. et al., Comparison of antisense oligonucleotides and siRNAs in cell culture and in vivo, Biochemical and Biophysical Research Communications 2002, 296: 1000‐1004.

6. Corey D.R., RNA learns from antisense, Nature Chemical Biology 2007, 3: 8‐11. 7. de Fougerolles A., Vornlocher H.P., Maraganore J., and Lieberman J., Interfering with disease: a progress

report on siRNA‐based therapeutics, Nature Reviews Drug Discovery 2007, 6: 443‐453. 8. Behlke M.A., Progress towards in vivo use of siRNAs, Molecular Therapy 2006, 13: 644‐670. 9. De Paula D., Bentley M.V., and Mahato R.I., Hydrophobization and bioconjugation for enhanced siRNA

delivery and targeting, RNA 2007, 13: 431‐456. 10. Dykxhoorn D.M. and Lieberman J., Knocking down disease with siRNAs, Cell 2006, 126: 231‐235. 11. Oliveira S., Storm G., and Schiffelers R.M., Targeted delivery of siRNA, Journal of Biomedicine and

Biotechnology 2006, 12. Bartlett D.W. and Davis M.E., Insights into the kinetics of siRNA‐mediated gene silencing from live‐cell and

live‐animal bioluminescent imaging, Nucleic Acids Research 2006, 34: 322‐333. 13. Bartlett D.W. and Davis M.E., Effect of siRNA nuclease stability on the in vitro and in vivo kinetics of siRNA‐

mediated gene silencing, Biotechnol. Bioeng. 2007, 97: 909‐921. 14. Kunisawa J., Masuda T., Katayama K., Yoshikawa T. et al., Fusogenic liposome delivers encapsulated

nanoparticles for cytosolic controlled gene release, Journal of controlled release 2005, 105: 344‐353. 15. Jeong J.H., Christensen L.V., Yockman J.W., Zhong Z.Y. et al., Reducible poly(amido ethylenimine) directed

to enhance RNA interference, Biomaterials 2007, 28: 1912‐1917. 16. Kakizawa Y., Furukawa S., and Kataoka K., Block copolymer‐coated calcium phosphate nanoparticles

sensing intracellular environment for oligodeoxynucleotide and siRNA delivery, Journal of controlled release 2004, 97: 345‐356.

17. Read M.L., Singh S., Ahmed Z., Stevenson M. et al., A versatile reducible polycation‐based system for efficient delivery of a broad range of nucleic acids, Nucleic Acids Research 2005, 33: e86.

18. Peppas N.A., Bures P., Leobandung W., and Ichikawa H., Hydrogels in pharmaceutical formulations, Eur. J. Pharm. Biopharm. 2000, 50: 27‐46.

19. Peppas N.A., Hilt J.Z., Khademhosseini A., and Langer R., Hydrogels in biology and medicine: From molecular principles to bionanotechnology, Advanced Materials 2006, 18: 1345‐1360.

20. Coviello T., Matricardi P., Marianecci C., and Alhaique F., Polysaccharide hydrogels for modified release formulations, Journal of controlled release 2007, 119: 5‐24.

21. Lin C.C. and Metters A.T., Hydrogels in controlled release formulations: Network design and mathematical modeling, Advanced Drug Delivery Reviews 2006, 58: 1379‐1408.

22. Van Tomme S.R. and Hennink W.E., Biodegradable dextran hydrogels for protein delivery applications, Expert Review of Medical Devices 2007, 4: 147‐164.

23. van Dijk‐Wolthuis W.N.E., Tsang S.K.Y., Kettenes‐vandenBosch J.J., and Hennink W.E., A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer, Polymer 1997, 38: 6235‐6242.

24. van Dijk‐Wolthuis W.N.E., van Steenbergen M.J., Underberg W.J.M., and Hennink W.E., Degradation kinetics of methacrylated dextrans in aqueous solution, Journal of Pharmaceutical Sciences 1997, 86: 413‐417.

25. van Dijk‐Wolthuis W.N.E., Hoogeboom J.A.M., van Steenbergen M.J., Tsang S.K.Y. et al., Degradation and release behavior of dextran‐based hydrogels, Macromolecules 1997, 30: 4639‐4645.

26. Stubbe B.G., Horkay F., Amsden B., Hennink W.E. et al., Tailoring the swelling pressure of degrading dextran hydroxyethyl methacrylate hydrogels, Biomacromolecules 2003, 4: 691‐695.

27. De Geest B.G., Dejugnat C., Verhoeven E., Sukhorukov G.B. et al., Layer‐by‐layer coating of degradable microgels for pulsed drug delivery, Journal of controlled release 2006, 116: 159‐169.

106


28. Van Tomme S.R., van Steenbergen M.J., De Smedt S.C., van Nostrum C.F. et al., Self‐gelling hydrogels based on oppositely charged dextran microspheres, Biomaterials 2005, 26: 2129‐2135.

29. Cadee J.A., de Groot C.J., Jiskoot W., den Otter W. et al., Release of recombinant human interleukin‐2 from dextran‐based hydrogels, Journal of controlled release 2002, 78: 1‐13.

30. Vlugt‐Wensink K.D.F., Vlugt T.J.H., Jiskoot W., Crommelin D.J.A. et al., Modeling the release of proteins from degrading crosslinked dextran microspheres using kinetic Monte Carlo simulations, Journal of controlled release 2006, 111: 117‐127.

31. Franssen O., Vandervennet L., Roders P., and Hennink W.E., Degradable dextran hydrogels: controlled release of a model protein from cylinders and microspheres, Journal of controlled release 1999, 60: 211‐221.

32. De Geest B.G., Vandenbroucke R.E., Guenther A.M., Sukhorukov G.B. et al., Intracellularly degradable polyelectrolyte microcapsules, Advanced Materials 2006, 18: 1005.

33. Javier A.M., Kreft O., Alberola A.P., Kirchner C. et al., Combined atomic force microscopy and optical microscopy measurements as a method to investigate particle uptake by cells, Small 2006, 2: 394‐400.

34. Sanchez‐Martin R.M., Muzerelle M., Chitkul N., How S.E. et al., Bead‐based cellular analysis, sorting and multiplexing, Chembiochem 2005, 6: 1341‐1345.

35. Skirtach A.G., Javier A.M., Kreft O., Kohler K. et al., Laser‐induced release of encapsulated materials inside living cells, Angewandte Chemie‐International Edition 2006, 45: 4612‐4617.

36. Sukhorukov G.B., Rogach A.L., Zebli B., Liedl T. et al., Nanoengineered polymer capsules: Tools for detection, controlled delivery, and site‐specific manipulation, Small 2005, 1: 194‐200.

37. von Gersdorff K., Sanders N.N., Vandenbroucke R., De Smedt S.C. et al., The internalization route resulting in successful gene expression depends on polyethylenimine both cell line and polyplex type, Molecular Therapy 2006, 14: 745‐753.

107


108

advanced nanogel engineering for drug delivery | CHAPTER 4

109

CHAPTER 4 | Advanced nanogel engineering for drug delivery


Koen Raemdonck, Joseph Demeester and Stefaan C. De Smedt, Soft Matter 2009, 5: 707‐715



CHAPTER 4 | advanced nanogel engineering for drug delivery

110

ABSTRACT

Nanosized hydrogels (nanogels) have attracted considerable attention as multifunctional polymer‐

based drug delivery systems. Their versatility is demonstrated both in drug encapsulation and drug

release. Nanogels can be designed to facilitate the encapsulation of diverse classes of bioactive

compounds. With optimization of their molecular composition, size and morphology, nanogels can

be tailor‐made to sense and respond to environmental changes in order to ensure spatial and

stimuli‐controlled drug release in vivo. This chapter aims to highlight recent advances on the

interface between biology and nanomedicine with the emphasis on nanogels as carriers for

controlled drug delivery.



Chapter 4 provides:

• information on the design of nanogels and highlights on recent advanced preparation

protocols

• details on how nanogels can be used for the encapsulation and subsequent controlled

release of different classes of therapeutics, with the emphasis on the intracellular

environment

• insights in the possible applications of nanogels as drug delivery carriers and their ‘fine‐

tuning’ for in vivo delivery purposes

Key terms:

Crosslinking – hydrogel – dextran – stimuli‐responsive – PEGylation – controlled release

INTRODUCTION

In the past decade the quest for intelligent biomaterials in pharmaceutical applications has moved

into second gear. More than ever, the design of efficient drug delivery carriers has evolved as a

multidisciplinary effort where the interplay between material science, medicine and biology provides

innovative biofunctional materials.[1] Amongst the available biomaterials already described in the

literature, hydrogels have already proved their value in diverse biomedical applications. Hydrogels

are described as hydrophilic three‐dimensional polymer networks that are able to take up large

amounts of water or physiological fluid, while maintaining their internal network structure.[2,3] Their

high water content and low surface tension contribute to their biocompatibility. Some hydrogels

encompass a high loading capacity for bioactive compounds and are able to release their therapeutic

111


payload in a controlled fashion. These unique features explain their widespread application in tissue

engineering, drug delivery and diagnostics.[4]

Although research has mainly focused on macroscopic hydrogels, there is now an increasing interest

in hydrogels confined to micro‐ and especially nanoscopic dimensions (termed micro‐ and

nanogels).[5,6] This is easily proven by a simple literature search on ‘nanogels’ and ‘drug delivery’ that

reveals an exponential growth in the number of publications during the last 10 years (Figure 4.1).

Colloidally stable nanogel particles have properties similar to those of their macrogel counterparts,

but they have the added value that e.g. upon intravenous injection, they can be deployed in areas of

the body that are not easily accessed by macroscopic hydrogels and microgels.[7] As most nanogels

are able to be taken up by cells they are ideal candidates for intracellular drug delivery. This is of

particular interest for therapeutic agents that need to be safely delivered into the cytoplasm of the

target cell, such as antisense oligonucleotides, siRNAs, peptides and various low molecular weight

chemotherapeutics. In addition, nanogel dispersions have a large surface area that can function as a

template for multivalent conjugations which is important to optimize the particles towards in vivo

applications. Moreover, their nanoscaled dimensions ensure that they respond very rapidly to

environmental stimuli, which is attractive when aiming for triggered drug delivery. Hydrogels, and

nanogels in particular, hold all the cards to be versatile drug delivery carriers for exploitation in

biomedicine, as already discussed in many review articles.[2‐6,8,9] In this paper, we especially aim to

highlight some of the most recent advances made in this emerging field of nanomaterials, with the

emphasis on nanogel design towards efficient application for drug delivery purposes.

DESIGNING NANOSIZED HYDROGELS

Hydrogel structures can be formed by introducing chemical or physical crosslinks between

hydrophilic natural or synthetic polymers.[10] These crosslinks are essential for the structural stability

of the hydrogel as they prevent dissolution of the polymer chains in the aqueous environment.

Chemical crosslinking involves the formation of covalent bonds, leading to an insoluble polymer

network. To obtain nanogels instead of macro‐ or microgels by chemical crosslinking, mostly (inverse)

emulsion polymerization[11] and precipitation polymerization[12] methods are applied. Alternatively,

also physical interactions between polymer chains (e.g. hydrogen bonds, chain entanglements, Van

der Waals forces, electrostatic interactions,…) can result in stable hydrogels. Physical self‐assembly

implies the controlled formation of nanosized aggregates as a function of polymer concentration[13‐15]

or environmental parameters such as temperature and pH.[9,16]

112


FIGURE 4.1 | Number of publications on ‘nanogels’ in the field of ‘drug delivery’ during the last decade. Literature search was performed in Web of Science. (*) Data of 2009 are extrapolated based on the number of publications during the first trimester of that year.

We illustrate these different modes of crosslinking taking dextran biopolymer hydrogels as an

example.[17] Dextran chains (dex, Figure 4.2a) can be derivatized with different moieties such as 2‐

hydroxyethyl methacrylate (dex‐HEMA, Figure 4.2b), lactate‐2‐hydroxyethyl methacrylate (dex‐

lactate‐HEMA, Figure 4.2d) or allyl isocyanate (dex‐AI, Figure 4.2c) for radical polymerization to form

chemically crosslinked hydrogel particles.[17‐19] Alternatively, dextran nanogels were also obtained

through respectively stereocomplexation of dextran grafted with poly(L‐lactide) and poly(D‐lactide)

(Figure 4.2e)[15] and inclusion complexation of lauryl‐modified dextran (Figure 4.2f) with β‐

cyclodextrin polymers.[20] Jean Fréchet’s group recently reported on acid‐degradable dextran

nanogels based on acetalated dextran (Ac‐dextran, Figure 4.2g) which are formed using a double

emulsion (w/o/w) evaporation method in CH2Cl2.[21] A plethora of different protocols for colloidally

stable micro‐and nanogels can be found in the literature and many examples hereof are detailed by

Oh et al. in a recent review article[6] and in Yallapu et al.[9] Liu et al. published a review article on

polysaccharide based nanoparticles applied as drug delivery vehicles.[22]

Recently, new technologies have arisen for the preparation of nanogels with a better defined

molecular architecture, which is important to optimize them towards in vivo drug delivery

applications.

113


FIGURE 4.2 | Chemical structure of various monomers for dextran‐based hydrogel particles, (a) dextran, (b) dex‐HEMA: dextran hydroxyethyl methacrylate, (c) dex‐AI: dextran allyl isocyanate, (d) dex‐lactate‐HEMA, (e) dex‐g‐PLA: dextran poly(lactic acid), (f) lauryl modified dextran – R = H or CH3‐(CH2)10‐CO‐, (g) Ac‐DEX: acetalated dextran

One example is the application of liposomes as nanosized reactors where hydrogel formation occurs

in the aqueous lumen of the lipid vesicles. In this way, one can achieve better control over the

particle size since the dimensions of the ‘naked’ nanogels, obtained after lipid removal, closely match

the size of the liposomal template (Figure 4.3).[23‐28]

Secondly, controlled radical polymerization (CRP) techniques, which offer great opportunities for the

synthesis of (co)polymers and bioconjugates with controlled molecular weight and low

polydispersity[29], have recently been explored for the synthesis of crosslinked nanogels.[30]

Matyjaszewski’s group prepared well‐defined and uniformly crosslinked biodegradable nanogels

from oligo(ethylene glycol) monomethyl ether methacrylate (OEOMA) with a disulfide functionalized

114


crosslinker using inverse mini‐emulsion atom transfer radical polymerization (ATRP).[31] These

nanogels showed potential as drug delivery carriers in their studies.[32,33] Another interesting example

is the photolithographic fabrication of sub‐micrometer particles using a technique called PRINT

(Particle Replication in Nonwetting Templates). PRINT applies photocurable perfluoropolyether

molds for imprint lithography to form uniform nanoparticles of well‐controlled size, shape and

composition.[34‐36] Gratton and coworkers used PRINT nano‐ and microgels to investigate the

influence of particle size, shape and surface chemistry on their cellular internalization. The

experimental data point out that cationic rod‐shaped particles are internalized at higher rates,

underlining the importance of particle design when intracellular targets are envisioned.[37] An

alternative nanoimprint photo‐lithographic approach (Step and Flash Imprint Lithography or S‐FIL) for

the preparation of crosslinked peptide nanogels was recently presented by Glangchai et al. (Figure

4.4).[38]

FIGURE 4.3 | Synthesis of nanogels using a liposomal template. Liposomes are prepared in the presence of a polymer solution. Non‐encapsulated polymer is removed or diluted below the critical polymerization concentration. Crosslinking of the polymers inside the lipid vesicles leads to the formation of lipid coated nanogels. Removal of the lipid coat by e.g. the addition of a detergent, finally results in naked nanogels with a size comparable to that of the liposomal template.

Rigorous control over the size of nanocolloids is essential for their in vivo biodistribution. While

microgels are often designed to release their payload in the extracellular space or in the cytoplasm of

dendritic cells following subcutaneous or intramuscular injection, nanogels may also be applied for

intravenous administration where they need to extravasate through the capillary endothelium to

reach the target tissue. The morphology of the capillary wall will set the size limits for particle

extravasation into the surrounding tissue. Discontinuous or sinusoidal capillaries can be found in

liver, spleen and bone marrow. This ‘leaky’ vasculature is also observed in inflammation sites and

tumours which allows extravasation of particles even up to 400 nm in diameter.[1,39] Since size is a

critical determinant for nanogels to reach the intended target following intravenous administration,

115


new nanofabrication methods with good control over particle dimensions may be promising for in

vivo applications of drug loaded nanogels.

FIGURE 4.4 | (Top) Schematic representation of the Step and Flash Imprint Lithography (S‐FIL) method. Details can be found in Glangchai et al.[38], (Bottom) Scanning electron micrographs of S‐FIL imprinted poly(ethylene glycol) diacrylate (PEGDA) nanogels. Reprinted from reference [38].

Gao et al. recently published on ultrafine polyacrylamide‐based hydrogels with an average size ~2 nm

for the encapsulation of tetra(m‐hydroxyphenyl)chlorin (mTHPC) as a photosensitizer in

photodynamic therapy.[40] The authors state that owing to their small size, these hydrophilic particles

will evade the reticulo‐endothelial system (RES). Moreover, risk of accumulation of the mTHPC

loaded nanogels is also reduced since these non‐degradable particles should be cleared from the

bloodstream by renal filtration.[40]

NANOGELS AS VERSATILE DRUG DELIVERY VEHICLES

Drug encapsulation

Many drug molecules suffer severely from several impediments such as low solubility, off‐target

toxicity, instability or inefficient transfer across biological barriers, all of which significantly hamper

their in vivo use. Nanogels may offer a solution to many of these drug delivery issues, involving

various types of bioactive compounds. Both hydrophilic and lipophilic low molecular weight drugs

(e.g. certain chemotherapeutics[41]) and macromolecular therapeutics (DNA[42], siRNA[43‐45], peptides

and proteins[46]) may become incorporated in the nanogel network. The loading is generally based on

(a) physical entrapment and/or (b) non‐covalent interactions. The sterical entrapment of

116


therapeutics inside a hydrogel network implies in most cases that the compounds are present in the

polymer solution during the gelation process. However, it is conceivable that polymerization may

occur under conditions that are possibly detrimental for the therapeutic cargo. Therefore this

encapsulation technique obviously requires rigorous control over the gelation process to ensure drug

stability.[38,42,47] When the dimensions of the encapsulated therapeutics exceed the mesh size of the

hydrogel network, their diffusional leaching can be prevented. This encapsulation method is

therefore most feasible for macromolecular therapeutics,[38,42,43,48,49] especially when loosely

crosslinked or porous hydrogels are formed.

In contrast, non‐covalent interactions between the biological agent and the polymer matrix are the

basis of the research of Akiyoshi and co‐workers who focus on amphiphilic cholesterol‐modified

pullulan (CHP) nanogels of typically 20‐50 nm in diameter, mainly for the encapsulation of different

proteins.[50] Already in 1993, Akiyoshi disclosed that polysaccharides partially modified with

hydrophobic moieties could form nanoparticles through a self‐assembly process in an aqueous

environment.[51] The main driving force for protein encapsulation is the hydrophobic attraction

between the cholesteryl moieties in the nanogels and hydrophobic nanodomains in the protein of

interest. Recently, phase I clinical trials were conducted with CHP nanogels encapsulating truncated

HER2 protein for cancer vaccination. Patients received repetitive subcutaneous injections of the CHP‐

HER2 vaccine and results showed HER2 specific antibody and T‐cell immune responses.[52,53] A

drawback of this nanogel platform is their limitation for intravenous administration since abundant

serum proteins, like albumin, can destabilize the nanogel complex.[50] Especially when an intracellular

drug target is envisioned, the stability of drug delivery systems in the extracellular matrix should be

routinely tested to avoid premature drug release before the target cells have been reached.[54] For

the intracellular delivery of membrane‐impermeable compounds, amino‐functionalized CHP

nanogels were proposed to improve the efficiency of interactions with the cellular membrane and

their subsequent internalization. Once the cellular barrier is overcome, the authors assume that the

aforementioned protein exchange mechanism is also partially responsible for the intracellular release

of the encapsulated therapeutic protein.[46]

An alternative non‐covalent approach for the incorporation of therapeutics in charged nanogels is

based on electrostatic interactions between the biological agent and the ionized polymer matrix. This

approach has proved valuable for e.g. the complexation of low molecular weight anionic compounds

like nucleoside analog 5’‐triphosphates (NTPs) in crosslinked poly(ethylene glycol)‐poly(ethylene

imine) (PEG‐g‐PEI) nanogels (Figure 4.5)[55] Electrostatic complexation can also be exploited for

encapsulation of polynucleotides in cationic nanogels if the pore size of the nanogels is large enough

to allow penetration of the polynucleotides into the interior of the gel particle. In this way, antisense

117


oligonucleotides[56‐58] and short interfering RNA (siRNA)[18,57] can be loaded in preformed cationic

hydrogel networks. On the other hand, Hu et al. reported on nanogels for the surface binding of

proteins, inactivated influenza virions (for the purpose of antigen delivery to dendritic cells) and

siRNA duplexes (for gene silencing activity). These (predominantly) anionic cargos are adsorbed onto

preformed core‐shell particles, based on electrostatic interactions with the positively charged outer

layer.[45]

FIGURE 4.5 | Schematic representation of PEG‐g‐PEI nanogels, electrostatically loaded with 5’‐ nucleoside triphosphate anticancer agents (5’‐NTPs). The nanogels consist of a PEI core for 5’‐NTP binding and a PEG envelope, containing targeting ligands (●). Reprinted and modified from Vinogradov et al. [55].

Exceptionally, bioactive compounds are incorporated into the nanogel interior through covalent

conjugation. One example is given by Standley and co‐workers who linked a methacrylate‐modified

CpG oligonucleotide ligand into acid‐degradable nanogels prepared through an inverse emulsion

radical polymerization technique.[49] In addition to the oligonucleotide strand, also ovalbumin (OVA)

as an antigen model was encapsulated. The incorporation of CpG oligonucleotides resulted in a more

potent production of interleukin‐12 and enhanced OVA‐specific CD8 T‐cell immunity in vivo due to

activation of Toll‐like receptor 9 (TLR‐9). However, covalent attachment is not feasible for every type

of drug or application as it may also alter the drug’s effectiveness.

Triggered drug release

Hydrogels tend to swell when they are brought into a polymer compatible medium due to solvent

penetration into free spaces between the macromolecular chains. The resulting volumetric

118


expansion of the hydrogel network is governed by the balance between the internal osmotic

pressure and its elastic deformability.[59] An attractive feature of hydrogels is that their swelling

behaviour can be influenced by a diversity of external triggers, such as changes in environmental pH,

ionic strength or temperature and the application of light or a magnetic or electric field.[2,5,60] A

plethora of stimuli‐responsive nanocarriers is described in the literature, e.g. polymer‐drug

conjugates, liposomes and micelles, which all have their strengths and weaknesses.[61] The latter two

examples are of particular interest in drug delivery, but they suffer from drawbacks such as a lack of

controlled loadability, limitations in material composition, limitations in the type of biological cargo

that can be loaded/released and/or physical instability.[7,9,35] For these reasons, many nanogel

platforms are often regarded as a better alternative. The responsiveness of ‘intelligent’ hydrogels is

largely dependent on the type of polymer used in making the gel.[62,63] Control over the swelling‐

deswelling transitions or other physical changes on a nanoscopic scale are especially of interest for

intracellular controlled drug delivery.

Among the class of ‘intelligent nanogels’, those responding to changes in temperature and pH are

most extensively investigated. Nanogels constructed from polymers bearing pendant basic or acidic

groups (ionic nanogels) are generally regarded as pH responsive.[62] When they are dispersed in an

aqueous medium of appropriate pH and ionic strength, ionization of pendant groups occurs, thereby

introducing fixed charges in the polymer network. By electrostatic repulsion the pores in the

nanogels enlarge, allowing excessive solvent influx, which eventually leads to their swelling. Cationic

nanogels swell at a pH below the pKa of the polymer network, while anionic nanogels are ionized at

higher pH. Small changes in the external pH may have a significant influence on the swelling

behaviour.[2] Na and Bae reported on self‐assembled pH‐sensitive nanogels, composed of a

sulphonamide‐pullulan acetate conjugate[64] and hydrophobized pullulan‐Na‐Boc‐L‐histidine[65] for

the tumor specific release of doxorubicin. Most solid tumours elicit a lower extracellular pH (pH ~6.8)

when compared with the pH in other tissues or the blood circulation (pH 7.4).[61,66,67] The particles

proposed by Na and Bae were stable at pH 7.4. However, at pH 6.8 the sulphonamide (SA) tagged

nanogels shrank and aggregated due to SA deionization while the histidine tagged nanogels swelled

as a result of imidazole protonation in the histidine moieties. In both cases drug release occurred

faster in response to these physical changes resulting into a more specific deposition of the drug in

the tumour microenvironment.

In the intracellular microenvironment, such pH responsive carriers may also prove beneficial. It is

known that nanocarriers are internalized mostly through an endocytotic mechanism upon interaction

with the cellular membrane. Endocytosis eventually confines the particles in an acidic environment

as the pH in the late endosomal and lysosomal lumen gradually lowers from neutral to ~pH 5.[68]

119


Indeed, pH sensitive nanogels that undergo structural changes upon endocytosis could enhance the

intracellular bioavailability of the encapsulated drug. Chitosan nanogels around 180 nm in diameter

were described by Zhang and colleagues for the encapsulation of the anticancer agent methotrexate

disodium (MTX).[69] The nanogels originated from the ionic crosslinking of N‐[2‐hydroxy‐3‐

(trimethylammonium)propyl] chitosan chloride (HTCC) with sodium tripolyphosphate (TPP)

counterions. MTX is negatively charged and could be incorporated in the cationic HTCC nanogels at

physiological pH through electrostatic binding. Following efficient cellular internalization, these

nanogels showed a 2.2 fold increase in hydrodynamic diameter at endolysosomal pH due to higher

protonation of primary and secondary amines in HTCC and stronger electrostatic repulsion. As a

result, a faster diffusion driven release of MTX was achieved.[69] Shen et al. prepared chitosan

nanogels ~70‐80 nm through the crosslinking of chitosan with ethylenediamine tetraacetic acid

(EDTA). These nanogels exhibit a reversible change in surface charge and surface structure depending

on the environmental pH. The authors attempted to exploit this behaviour for the encapsulation and

release of the water‐insoluble anticancer drug camptothecin.[70] Similarly, Nagasaki’s lab showed pH‐

triggered release of doxorubicin due to protonation and subsequent swelling of ionic PEG‐poly[2‐

(N,N‐(diethylamino)ethyl methacrylate) nanogels at endolysosomal pH.[71,72] In particular for

biomacromolecules like protein antigens, pDNA and siRNA, an efficient transfer from the endosomal

compartment into the cytosol is crucial since otherwise they become degraded by endolysosomal

proteases and nucleases. Employing pH‐sensitive nanogels could enhance this endosomal escape

significantly. Indeed, it is thought that when nanogels are constructed from polyelectrolytes with

endosomal buffering capacity, they can disrupt endosomes via an osmotic pressure buildup (proton

sponge hypothesis). In addition, the swelling of the particles, due to endosomal protonation, may

facilitate endosomal destabilization.[45,73,74]

The pharmaceutical relevance of thermosensitive nanogels may lie within the treatment of local

infections (with elevated environmental temperature), tumour hyperthermia or by artificially

increasing temperature in the affected tissue to tailor local drug delivery.[7,61] Accumulation of a drug

loaded nanoparticle at the disease site, followed by a local temperature increase can trigger drug

release due to thermally triggered volume transitions. Poly(N‐isopropylacrylamide) or PNIPAAm is

the most widely investigated thermoresponsive polymer for biomedical applications.[63] PNIPAAm

nanogels can exhibit significant volume change in response to thermal fluctuations.[75‐78] In aqueous

solution the polymer has a lower critical solution temperature (LCST) around 32°C.[44] Below the LCST,

the polymer network is hydrated by hydrogen bonding of water to the amide side groups. When the

temperature increases towards or above the LCST, the polymer‐polymer hydrophobic interactions

become more dominant. As the hydrogen bonds are weakened, phase separation occurs and the

120


polymer chains collapse. The resulting expulsion of water and nanogel aggregation can be

accompanied with the release of both hydrophilic and hydrophobic drugs in the surrounding

medium.[79] A slightly different therapeutic concept, but also based on thermally triggered volume

transition, was recently proposed by Lee et al.. The authors prepared complex degradable nanogels

that exhibit a size increase from nano‐ to microscale when lowering the environmental temperature

from 37°C to 15°C. These nanogels are subsequently employed as ‘nanobombs’ to induce necrotic

cell death after cold‐shock treatment of cells that had internalized the nanogels in deswollen state.[80]

Modification of NIPAAm with different comonomers or post‐polymerization modifications on

PNIPAAm nanogels may significantly alter the hydrogel characteristics and shift the LCST. This feature

enables the application of PNIPAAm nanogels over a wide temperature range.[76] As an example,

incorporation of ionic and hydrophilic comonomers, such as acrylic acid (AAc) or methacrylic acid

(MAAc), increases the LCST and makes the nanogels sensitive both to changes in temperature as well

as changes in pH.[74,81‐86] A plethora of reports, describing different modifications of NIPAAm, can be

found in the literature.[5,6,63,83,86,87] A particularly interesting example of nanogel engineering is the

development of amphoteric, glucose‐responsive PNIPAAm based nanogels for triggered insulin

release.[88] Grafting of phenylboronic acid (PBA) to cationic‐anionic nanogels introduces glucose‐

responsiveness in the nanogels. The swelling‐deswelling behaviour of the nanogels as a function of

physiological glucose concentrations can be fine‐tuned by varying the relative amount of PBA,

cationic and anionic groups in the PNIPAAm hydrogel particles.[88] Besides pH and temperature

sensitive nanogels, also nanogels which are sensitive simultaneously to multiple triggers have already

been prepared. For instance, Bhattacharya et al. engineered hybrid nanogels that contain

ferromagnetic nanoparticles and which are responsive to both temperature, pH and magnetic

fields.[89]

Depending on the elastic properties of the polymer network, the deformation of hydrogels can be

reversible. In this way, hydrogels can regain their original shape and structure when the stimulus,

triggering the swelling or shrinking, is removed. This ‘shape memory effect’ can also have its

implications on drug delivery as recently disclosed by You Han Bae’s group. They prepared nanogels

consisting of a pH‐sensitive hydrophobic core of poly(L‐histidine‐co‐phenylalanine), held together by

a double hydrophilic shell.[90] These nanogels are colloidally stable at neutral pH, but swell

substantially at endosomal pH (< pH 6.8) explaining the faster release rate of encapsulated

doxorubicin at lower pH values. After endocytosis, the swelling of the nanogels together with the

buffering effect of poly(L‐histidine) is considered to disrupt the endosomal membranes, thereby

facilitating the transfer of the nanogels and the released doxorubicin to the cytoplasm. There the

nanogels shrink again to their original size, imparting further release of doxorubicin. The doxorubicin

121


released intracellularly induces apoptosis that allows the nanogels, still containing sufficient amounts

of doxorubicin, to be taken up by neighbouring cells. In this way, doxorubicin loaded nanogels may

‘infect’ several cells and release the anticancer agent in a pH triggered pulsatile fashion (Figure

4.6).[90]

Different types of degradable nanogels are currently explored with the aim of enhancing the

intracellular delivery of encapsulated drugs. Moreover, degradability of the nanogel carrier is

essential in drug delivery to lower toxicity and avoid accumulation in the body upon repeated

administration. Labile bonds can be incorporated in the polymer backbone or in the crosslinks of the

hydrogel network. This class of nanogels releases its payload upon degradation in response to

specific intracellular stimuli such as pH[47], reducing agents or enzymatic activity. As a first example,

the Fréchet group reported on DNA and protein antigen encapsulation into acrylamide nanogels

copolymerized with an acid cleavable bisacrylamide acetal crosslinker.[42,48,49] These nanogels are

designed to degrade under the acidic conditions found in the phagolysosomes of Antigen Presenting

Cell’s (APCs). Upon lysosomal degradation of the nanogels, the encapsulated DNA can be released

and interact with TLR‐9 for enhanced cytokine production.[42,49]

Moreover, the accumulation of degradation fragments in the lysosomal lumen induces a rise in

osmotic pressure which is believed to destabilize the endosomal membrane and enables the transfer

of the encapsulated antigen to the cytoplasm for MHC class I processing.[48,49] Second, disulfide

crosslinked nanogels employ the intracellular reductive potential to deliver their therapeutic

payload.[91] It is known that disulfide linkages can be selectively cleaved inside the target cell by the

reductive environment in the endo/lysosomes and by cytoplasmic glutathione.[92] Lee et al.

succeeded in the fabrication of hyaluronic acid (HA) based nanogels around 200‐500 nm in size.[43]

For this purpose, thiol modified HA was crosslinked by ultrasonic treatment through the formation of

disulfide bonds in an inverse water‐in‐oil emulsification process. SiRNA was added to the HA solution

during the emulsification process which led to their physical entrapment inside the resulting

nanogels. Upon cellular internalisation, disulfide reduction disassembled the nanogels which resulted

in siRNA release. In a comparable approach, cationic PEI‐cl‐PEG/Pluronic nanogels[93] were made

degradable by the use of branched PEI polycations in which the 2 kDa PEI segments were linked via

disulfide bridges.[92] Very recently, Morimoto et al. reported on the self‐assembly of PNIPAAm‐g‐

pullulan chains carrying thiol end groups, resulting in nanogels responding to changes in

environmental temperature and redox potential. Below the LCST, only disulfide bridges secure

nanogel stability, while above the LCST the hydrophobic interactions between the PNIPAAm side

chains introduce additional physical crosslinks in the hydrogel network.[94]

122


FIGURE 4.6 | (Top) Protonation of histidine moieties in poly(L‐histidine‐co‐phenylalanine) nanogels at lower pH results in significant swelling of the nanogel core triggering doxorubicin (DOX) release, (Bottom) Pulsatile release pattern of doxorubicin from poly(L‐histidine‐co‐phenylalanine) nanogels as a result of swelling/deswelling transitions in response to a change in environmental pH. Reprinted and modified from Lee et al. [90].

Third, intracellular release can also be triggered by enzymatic degradation of the nanogels. Mostly

macro‐ and microscopic enzyme‐responsive hydrogels have been investigated so far.[95‐97] However,

Glangchai et al. recently demonstrated the lithographic preparation of enzymatically‐triggered

nanogels.[38] The authors designed sub 100 nm sized PEG diacrylate nanogels that are copolymerized

123


with an acrylated GFLGK peptide crosslinker. Enzymatic cleavage of the crosslinker by thiol proteases,

e.g. cathepsin B, releases the therapeutic payload which was shown for antibodies and pDNA. As this

class of enzymes is mainly located in the extracellular matrix of tumours and inside lysosomes, the

drug will be mainly released locally in these confined compartments.

Optimizing nanogel architecture for in vivo application

Ideally, drug delivery vehicles have to safely guide their therapeutic payload to the desired target

tissue. However, to reach the target tissue, several extracellular barriers need to be overcome.[1]

Shielding of the nanoparticle surface with inert, hydrophilic polymers and introducing targeting

ligands may significantly improve in vivo delivery and therapeutic efficacy (Figure 4.7). The

hydrophilic modification of nanoparticles can (a) prevent particle uptake by the mononuclear

phagocytic system, (b) decrease the recognition by the immune system and (c) enhance their

circulation time in the bloodstream. Indeed, the presence of a hydrophilic surface is known to reduce

non‐specific interactions with serum proteins and thereby also lowers the risk of phagocytosis by

immune cells as opsonisation is prevented.[98] Longer circulation times offer nanogels a better chance

of targeting the site of interest and may increase passive accumulation of drug loaded nanogels into

distant tumours due to the enhanced permeability and retention (EPR) effect. It has been described

that the leaky discontinuous vasculature often found in tumour tissue, together with the slower

efflux, can result in high local deposition of nanoparticles.[61,99,100] In addition, targeting of cell‐surface

receptors is a very attractive concept to achieve cell specific binding and subsequent internalization.

In this way, the fraction of the administered dose that accumulates at the target site can be

enhanced thereby also minimizing the risk of adverse effects. Many recent investigations have

evaluated both strategies for drug loaded nanogels.

With regard to the shielding of nanogels, most research is focused on PEG based nanogels or the

surface modification of nanogels with a hydrophilic PEG shell (also termed PEGylation). Several

groups constructed stimuli‐responsive core‐crosslinked polymeric micelles with a PEG corona.[101‐104]

However, some controversy exists on whether or not this type of nanocarriers can be regarded as

nanogels. Nagasaki and co‐workers copolymerized PEG, carrying a vinylbenzyl group, with 2‐(N,N‐

diethylamino)ethyl methacrylate to obtain cationic nanogels with a hydrophilic PEG shell.[71,72,105] Lee

et al. fabricated a double hydrophilic shell, consisting of PEG and bovine serum albumin (BSA), at the

surface of polypeptide nanogels.[90] Besides PEG, also other hydrophilic polymers have been

investigated for their ‘stealth’ properties, such as various polysaccharides.[106‐109]

124


FIGURE 4.7 | Biofunctionalization of stimuli‐responsive nanogels through the introduction of a hydrophilic shell and the coupling of targeting ligands may result in nanogel carriers that are of interest for clinical applications.

With regard to the active targeting of nanoparticles, the iron transporting plasma protein transferrin

is frequently used as a targeting ligand for tumour targeting since the expression of the transferrin

receptor is often upregulated in fast‐dividing tissues.[110] Chitosan[69], poly(NIPAAm) based[82,87] and

PEG‐cl‐PEI nanogels[93] were modified with transferrin to improve their cellular uptake and

intracellular drug delivery. The folate receptor is also overexpressed in many tumours[111], making

folic acid a prime candidate for active targeting of nanogels.[55,74,75,90,112] Other valuable approaches

include incorporating ligands for the asialoglycoprotein receptor, used in hepatocyte targeting[81,113],

or the conjugation of monoclonal antibodies for specific cell‐surface markers.[57,114] A recent

approach by the group of Andrew Lyon involves the coupling of a peptide shell to PNIPAAm core

nanogels for receptor mediated siRNA targeting towards carcinoma cells overexpressing the

erythropoietin producing hepatocellular A2 receptor (EphA2).[44]

CONCLUSIONS AND FUTURE PERSPECTIVES

Owing to our increasing knowledge on the biological barriers that nanomaterials encounter in vivo,

research in the field of drug delivery has gradually evolved to a top‐to‐bottom approach.[1] This

125


approach implies that scientists now aim to design nanoparticles with appropriate drug delivery

properties in relation to the envisioned biological target. Nanogels fit extremely well into this

strategy, since they offer the possibility for stimuli‐controlled release of therapeutic agents. Several

proof‐of‐concept studies have highlighted the potential of such ‘intelligent’ nanogels that are able to

sense and respond to environmental changes. In this way, drug release can be controlled in a spatial

and temporal fashion. Especially attractive so far has been the design of nanogels that retain their

therapeutic payload in the extracellular environment but release it once they are internalized by

target cells, in response to an intracellular trigger. However, like for all types of nanoparticles under

investigation for drug delivery, the in vivo biodistribution of nanogels is mainly governed by their size

and surface properties which should also be taken into careful consideration in the engineering of

nanogels for clinical applications.

Despite the progress made in this field of nanomaterials there still is a long way to go before drug

loaded nanogels will reach the clinic. More than a decade of intensive research provided the

technological basis for the development of ‘intelligent’ soft nanomaterials. Nonetheless, new

developments both in engineering stimuli‐responsive materials, bioconjugation methods as in

controlled colloid synthesis are key considerations to further improve the design of advanced

hydrogel nanomaterials. Moreover, it must be emphasized that there is an urgent need for in vivo

data from nanogels in preclinical development. To translate the nanogel concept into a viable

therapeutic approach, it will indeed become increasingly important to study toxicity, immunogenicity

and pharmacokinetics together with drug effects in validated in vivo models.

Amongst the different classes of polymeric materials in drug delivery, polysaccharides are very

frequently used to prepare nanoparticles and nanogels. As abundant natural biomaterials, they can

be regarded as safe, biocompatible and they usually have a low processing cost.[22] Chapter 3 already

mentioned the use of methacrylate functionalized dextran (dex‐HEMA) for the production of

hydrogels and microgels. In the next chapter, this material will be investigated for the preparation of

nanosized hydrogels that are to be deployed as carriers of small interfering RNA (siRNA).

ACKNOWLEDGMENTS

Koen Raemdonck is a doctoral fellow of the Institute for the Promotion of Innovation through

Science and Technology in Flanders (IWT‐Vlaanderen). The authors would like to acknowledge the EU

(FP6) for funding of MediTrans, an integrated project on nanomedicines.

126


REFERENCE LIST

1. Remaut K., Sanders N.N., De Geest B.G., Braeckmans K. et al., Nucleic acid delivery: Where material sciences and bio‐sciences meet, Materials Science & Engineering R‐Reports 2007, 58: 117‐161.

2. Gupta P., Vermani K., and Garg S., Hydrogels: from controlled release to pH‐responsive drug delivery, Drug Discovery Today 2002, 7: 569‐579.

3. Peppas N.A., Bures P., Leobandung W., and Ichikawa H., Hydrogels in pharmaceutical formulations, European Journal of Pharmaceutics and Biopharmaceutics 2000, 50: 27‐46.

4. Peppas N.A., Hilt J.Z., Khademhosseini A., and Langer R., Hydrogels in biology and medicine: From molecular principles to bionanotechnology, Advanced Materials 2006, 18: 1345‐1360.

5. Nayak S. and Lyon L.A., Soft nanotechnology with soft nanoparticles, Angewandte Chemie‐International Edition 2005, 44: 7686‐7708.

6. Oh J.K., Drumright R., Siegwart D.J., and Matyjaszewski K., The development of microgels/nanogels for drug delivery applications, Progress in Polymer Science 2008, 33: 448‐477.

7. Malmsten M., Soft drug delivery systems, Soft Matter 2006, 2: 760‐769. 8. Ulanski P. and Rosiak J.M., polymeric nano/microgels, encyclopedia of nanoscience and nanotechnology

2008, 8: 845‐871. 9. Yallapu M.M., Reddy M.K., and Labhasetwar V., Nanogels: chemistry to drug delivery, Biomedical

applications of nanotechnology 2008, 131‐171. 10. Hennink W.E. and van Nostrum C.F., Novel crosslinking methods to design hydrogels, Advanced Drug

Delivery Reviews 2002, 54: 13‐36. 11. Landfester K., Synthesis of colloidal particles in miniemulsions, Annual Review of Materials Research

2006, 36: 231‐279. 12. Pelton R., Temperature‐sensitive aqueous microgels, Advances in Colloid and Interface Science 2000, 85:

1‐33. 13. Akiyama E., Morimoto N., Kujawa P., Ozawa Y. et al., Self‐assembled nanogels of cholesteryl‐modified

polysaccharides: Effect of the polysaccharide structure on their association characteristics in the dilute and semidilute regimes, Biomacromolecules 2007, 8: 2366‐2373.

14. Kuroda K., Fujimoto K., Sunamoto J., and Akiyoshi K., Hierarchical self‐assembly of hydrophobically modified pullulan in water: Gelation by networks of nanoparticles, Langmuir 2002, 18: 3780‐3786.

15. Nagahama K., Mori Y., Ohya Y., and Ouchi T., Biodegradable nanogel formation of polylactide‐grafted dextran copolymer in dilute aqueous solution and enhancement of its stability by stereocomplexation, Biomacromolecules 2007, 8: 2135‐2141.

16. Yu S.Y., Hu J.H., Pan X.Y., Yao P. et al., Stable and pH‐sensitive nanogels prepared by self‐assembly of chitosan and ovalbumin, Langmuir 2006, 22: 2754‐2759.


18. Raemdonck K., Van Thienen T.G., Vandenbroucke R.E., Sanders N.N. et al., Dextran microgels for time‐controlled delivery of siRNA, Advanced Functional Materials 2008, 18: 993‐1001.

19. Huang X., Misra G.P., Vaish A., Flanagan J.M. et al., Novel Nanogels with Both Thermoresponsive and Hydrolytically Degradable Properties, Macromolecules 2008, 41: 8339‐8345.

20. Daoud‐Mahammed S., Ringard‐Lefebvre C., Razzouq N., Rosilio V. et al., Spontaneous association of hydrophobized dextran and poly‐[beta]‐cyclodextrin into nanoassemblies.: Formation and interaction with a hydrophobic drug, Journal of Colloid and Interface Science 2007, 307: 83‐93.

21. Broaders K.E., Cohen J.A., Beaudette T.T., Bachelder E.M. et al., Acetalated dextran is a chemically and biologically tunable material for particulate immunotherapy, Proc. Natl. Acad. Sci. U. S. A 2009, 106: 5497‐5502.

22. Liu Z.H., Jiao Y.P., Wang Y.F., Zhou C.R. et al., Polysaccharides‐based nanoparticles as drug delivery systems, Advanced Drug Delivery Reviews 2008, 60: 1650‐1662.

23. Hong J.S., Vreeland W.N., Paoli Lacerda S.H., Locascio L.E. et al., Liposome‐templated supramolecular assembly of responsive alginate nanogels, Langmuir 2008, 24: 4092‐4096.

24. Kazakov S., Kaholek M., Teraoka I., and Levon K., UV‐induced gelation on nanometer scale using liposome reactor, Macromolecules 2002, 35: 1911‐1920.

25. Van Thienen T.G., Lucas B., Flesch F.M., van Nostrum C.F. et al., On the synthesis and characterization of biodegradable dextran nanogels with tunable degradation properties, Macromolecules 2005, 38: 8503‐8511.

127


26. Van Thienen T.G., Raemdonck K., Demeester J., and De Smedt S.C., Protein release from biodegradable dextran nanogels, Langmuir 2007, 23: 9794‐9801.

27. Van Thienen T.G., Demeester J., and De Smedt S.C., Screening poly(ethyleneglycol) micro‐ and nanogels for drug delivery purposes, International Journal of Pharmaceutics 2008, 351: 174‐185.

28. An S.Y., Bui M.P., Nam Y.J., Han K.N. et al., Preparation of monodisperse and size‐controlled poly(ethylene glycol) hydrogel nanoparticles using liposome templates, J. Colloid Interface Sci. 2009, 331: 98‐103.

29. Braunecker W.A. and Matyjaszewski K., Controlled/living radical polymerization: Features, developments, and perspectives, Progress in Polymer Science 2007, 32: 93‐146.

30. Kim K.H., Kim J., and Jo W.H., Preparation of hydrogel nanoparticles by atom transfer radical polymerization of N‐isopropylacrylamide in aqueous media using PEG macro‐initiator, Polymer 2005, 46: 2836‐2840.

31. Oh J.K., Tang C.B., Gao H.F., Tsarevsky N.V. et al., Inverse miniemulsion ATRP: A new method for synthesis and functionalization of well‐defined water‐soluble/cross‐linked polymeric particles, Journal of the American Chemical Society 2006, 128: 5578‐5584.

32. Oh J.K., Siegwart D.J., Lee H.I., Sherwood G. et al., Biodegradable nanogels prepared by atom transfer radical polymerization as potential drug delivery carriers: Synthesis, biodegradation, in vitro release, and bioconjugation, Journal of the American Chemical Society 2007, 129: 5939‐5945.

33. Oh J.K., Siegwart D.J., and Matyjaszewski K., Synthesis and biodegradation of nanogels as delivery carriers for carbohydrate drugs, Biomacromolecules 2007, 8: 3326‐3331.

34. Euliss L.E., Dupont J.A., Gratton S., and DeSimone J., Imparting size, shape, and composition control of materials for nanomedicine, Chemical Society Reviews 2006, 35: 1095‐1104.

35. Napier M.E. and Desimone J.M., Nanoparticle drug delivery platform, Polymer Reviews 2007, 47: 321‐327.

36. Rolland J.P., Maynor B.W., Euliss L.E., Exner A.E. et al., Direct fabrication and harvesting of monodisperse, shape‐specific nanobiomaterials, Journal of the American Chemical Society 2005, 127: 10096‐10100.

37. Gratton S.E.A., Ropp P.A., Pohlhaus P.D., Luft J.C. et al., The effect of particle design on cellular internalization pathways, Proceedings of the National Academy of Sciences of the United States of America 2008, 105: 11613‐11618.

38. Glangchai L.C., Caldorera‐Moore M., Shi L., and Roy K., Nanoimprint lithography based fabrication of shape‐specific, enzymatically‐triggered smart nanoparticles, Journal of Controlled Release 2008, 125: 263‐272.

39. Brown J.M. and Giaccia A.J., The unique physiology of solid tumors: Opportunities (and problems) for cancer therapy, Cancer Research 1998, 58: 1408‐1416.

40. Gao D., Xu H., Philbert M.A., and Kopelman R., Ultrafine hydrogel nanoparticles: Synthetic approach and therapeutic application in living cells, Angewandte Chemie‐International Edition 2007, 46: 2224‐2227.

41. Missirlis D., Kawamura R., Tirelli N., and Hubbell J.A., Doxorubicin encapsulation and diffusional release from stable, polymeric, hydrogel nanoparticles, European Journal of Pharmaceutical Sciences 2006, 29: 120‐129.

42. Goh S.L., Murthy N., Xu M.C., and Frechet J.M.J., Cross‐linked microparticles as carriers for the delivery of plasmid DNA for vaccine development, Bioconjugate Chemistry 2004, 15: 467‐474.

43. Lee H., Mok H., Lee S., Oh Y.K. et al., Target‐specific intracellular delivery of siRNA using degradable hyaluronic acid nanogels, Journal of Controlled Release 2007, 119: 245‐252.

44. Blackburn W.H., Dickerson E.B., Smith M.H., McDonald J.F. et al., Peptide‐Functionalized Nanogels for Targeted siRNA Delivery, Bioconjug. Chem. 2009, 286: 563‐569.

45. Hu Y., Atukorale P.U., Lu J.J., Moon J.J. et al., Cytosolic delivery mediated via electrostatic surface binding of protein, virus, or siRNA cargos to pH‐responsive core‐shell gel particles, Biomacromolecules 2009, 10: 756‐765.

46. Ayame H., Morimoto N., and Akiyoshi K., Self‐assembled cationic nanogels for intracellular protein delivery, Bioconjugate Chemistry 2008, 19: 882‐890.

47. Shi L.J., Khondee S., Linz T.H., and Berkland C., Poly(N‐vinylformamide) nanogels capable of pH‐sensitive protein release, Macromolecules 2008, 41: 6546‐6554.

48. Kwon Y.J., Standley S.M., Goh S.L., and Fréchet J.M.J., Enhanced antigen presentation and immunostimulation of dendritic cells using acid‐degradable cationic nanoparticles, Journal of Controlled Release 2005, 105: 199‐212.

128


49. Standley S.M., Mende I., Goh S.L., Kwon Y.J. et al., Incorporation of CpG oligonucleotide ligand into protein‐loaded particle vaccines promotes antigen‐specific CD8 T‐cell immunity, Bioconjugate Chemistry 2007, 18: 77‐83.

50. Shimizu T., Kishida T., Hasegawa U., Ueda Y. et al., Nanogel DDS enables sustained release of IL‐12 for tumor immunotherapy, Biochemical and Biophysical Research Communications 2008, 367: 330‐335.

51. Akiyoshi K., Deguchi S., Moriguchi N., Yamaguchi S. et al., Self‐Aggregates of Hydrophobized Polysaccharides in Water ‐ Formation and Characteristics of Nanoparticles, Macromolecules 1993, 26: 3062‐3068.

52. Kageyama S., Kitano S., Hirayama M., Nagata Y. et al., Humoral immune responses in patients vaccinated with 1‐146 HER2 protein complexed with cholesteryl pullulan nanogel, Cancer Science 2008, 99: 601‐607.

53. Kitano S., Kageyama S., Nagata Y., Miyahara Y. et al., HER2‐specificT‐cell immune responses in patients vaccinated with truncated HER2 protein complexed with nanogels of cholesteryl pullulan, Clinical Cancer Research 2006, 12: 7397‐7405.

54. Buyens K., Lucas B., Raemdonck K., Braeckmans K. et al., A fast and sensitive method for measuring the integrity of siRNA‐carrier complexes in full human serum, Journal of Controlled Release 2008, 126: 67‐76.

55. Vinogradov S.V., Zeman A.D., Batrakova E.V., and Kabanov A.V., Polyplex Nanogel formulations for drug delivery of cytotoxic nucleoside analogs, Journal of Controlled Release 2005, 107: 143‐157.

56. McAllister K., Sazani P., Adam M., Cho M.J. et al., Polymeric nanogels produced via inverse microemulsion polymerization as potential gene and antisense delivery agents, Journal of the American Chemical Society 2002, 124: 15198‐15207.

57. Vinogradov S.V., Colloidal microgels in drug delivery applications, Current Pharmaceutical Design 2006, 12: 4703‐4712.

58. Vinogradov S., Batrakova E., and Kabanov A., Poly(ethylene glycol)‐polyethyleneimine NanoGel(TM) particles: novel drug delivery systems for antisense oligonucleotides, Colloids and Surfaces B: Biointerfaces 1999, 16: 291‐304.

59. Eichenbaum G.M., Kiser P.F., Simon S.A., and Needham D., pH and ion‐triggered volume response of anionic hydrogel microspheres, Macromolecules 1998, 31: 5084‐5093.

60. Ahn S.K., Kasi R.M., Kim S.C., Sharma N. et al., Stimuli‐responsive polymer gels, Soft Matter 2008, 4: 1151‐1157.

61. Ganta S., Devalapally H., Shahiwala A., and Amiji M., A review of stimuli‐responsive nanocarriers for drug and gene delivery, Journal of Controlled Release 2008, 126: 187‐204.

62. Dai S., Ravi P., and Tam K.C., pH‐Responsive polymers: synthesis, properties and applications, Soft Matter 2008, 4: 435‐449.

63. Rzaev Z.M.O., Dincer S., and Piskin E., Functional copolymers of N‐isopropylacrylamide for bioengineering applications, Progress in Polymer Science 2007, 32: 534‐595.

64. Na K. and Bae Y.H., Self‐assembled hydrogel nanoparticles responsive to tumor extracellular pH from pullulan derivative/sulfonamide conjugate: Characterization, aggregation, and adriamycin release in vitro, Pharmaceutical Research 2002, 19: 681‐688.

65. Na K., Lee E.S., and Bae Y.H., Self‐organized nanogels responding to tumor extracellular pH: pH‐dependent drug release and in vitro cytotoxicity against MCF‐7 cells, Bioconjugate Chemistry 2007, 18: 1568‐1574.

66. Gerweck L.E. and Seetharaman K., Cellular pH gradient in tumor versus normal tissue: Potential exploitation for the treatment of cancer, Cancer Research 1996, 56: 1194‐1198.

67. Stubbs M., McSheehy P.M.J., Griffiths J.R., and Bashford C.L., Causes and consequences of tumour acidity and implications for treatment, Molecular Medicine Today 2000, 6: 15‐19.

68. Van Dyke R.W., Acidification of lysosomes and endosomes, Subcell. Biochem. 1996, 27: 331‐360. 69. Zhang H., Mardyani S., Chan W.C.W., and Kumacheva E., Design of biocompatible chitosan microgels for

targeted pH‐mediated intracellular release of cancer therapeutics, Biomacromolecules 2006, 7: 1568‐1572.

70. Shen X.C., Zhang L.Y., Jiang X.Q., Hu Y. et al., Reversible surface switching of nanogel triggered by external stimuli, Angewandte Chemie‐International Edition 2007, 46: 7104‐7107.

71. Hayashi H., Iijima M., Kataoka K., and Nagasaki Y., pH‐sensitive nanogel possessing reactive PEG tethered chains on the surface, Macromolecules 2004, 37: 5389‐5396.

72. Oishi M., Hayashi H., Michihiro I.D., and Nagasaki Y., Endosomal release and intracellular delivery of anticancer drugs using pH‐sensitive PEGylated nanogels, Journal of Materials Chemistry 2007, 17: 3720‐3725.

129


73. Hu Y., Litwin T., Nagaraja A.R., Kwong B. et al., Cytosolic delivery of membrane‐impermeable molecules in dendritic cells using pH‐Responsive core‐shell nanoparticles, Nano Letters 2007, 7: 3056‐3064.

74. Soppimath K.S., Liu L.H., Seow W.Y., Liu S.Q. et al., Multifunctional core/shell nanoparticles self‐assembled from pH‐induced thermosensitive polymers for targeted intracellular anticancer drug delivery, Advanced Functional Materials 2007, 17: 355‐362.

75. Blackburn W.H. and Lyon L.A., Size‐controlled synthesis of monodisperse core/shell nanogels, Colloid and Polymer Science 2008, 286: 563‐569.

76. Debord J.D. and Lyon L.A., Synthesis and characterization of pH‐responsive copolymer microgels with tunable volume phase transition temperatures, Langmuir 2003, 19: 7662‐7664.

77. Gan D.J. and Lyon L.A., Tunable swelling kinetics in core‐shell hydrogel nanoparticles, Journal of the American Chemical Society 2001, 123: 7511‐7517.

78. Kuckling D., Vo C.D., Adler H.J.P., Volkel A. et al., Preparation and characterization of photo‐cross‐linked thermosensitive PNIPAAm nanogels, Macromolecules 2006, 39: 1585‐1591.

79. Shin Y.S., Chang J.H., Liu J., Williford R. et al., Hybrid nanogels for sustainable positive thermosensitive drug release, Journal of Controlled Release 2001, 73: 1‐6.

80. Lee Y., Park S.Y., Kim C., and Park T.G., Thermally triggered intracellular explosion of volume transition nanogels for necrotic cell death, J. Control Release 2009, 135: 89‐95.

81. Choi S.H., Yoon J.J., and Park T.G., Galactosylated poly(N‐isopropylacrylamide) hydrogel submicrometer particles for specific cellular uptake within hepatocytes, Journal of Colloid and Interface Science 2002, 251: 57‐63.

82. Das M., Mardyani S., Chan W.C.W., and Kumacheva E., Biofunctionalized pH‐responsive microgels for cancer cell targeting: Rational design, Advanced Materials 2006, 18: 80‐83.

83. Soppimath K.S., Tan D.C.W., and Yang Y.Y., pH‐triggered thermally responsive polymer core‐shell nanoparticles for drug delivery, Advanced Materials 2005, 17: 318.

84. Teng D.Y., Hou J.L., Zhang X.G., Wang X. et al., Glucosamine‐carrying temperature‐ and pH‐sensitive microgels: Preparation, characterization, and in vitro drug release studies, Journal of Colloid and Interface Science 2008, 322: 333‐341.

85. Tian P., Wu Q.L., and Lian K., Preparation of temperature‐ and pH‐sensitive, stimuli‐responsive poly(N‐isopropylacrylamide‐co‐methacrylic acid) nanoparticles, Journal of Applied Polymer Science 2008, 108: 2226‐2232.

86. Zhang L.Y., Guo R., Yang M., Jiang X.Q. et al., Thermo and pH dual‐responsive nanoparticles for anti‐cancer drug delivery, Advanced Materials 2007, 19: 2988‐+.

87. Quan C.Y., Sun Y.X., Cheng H., Cheng S.X. et al., Thermosensitive P(NIPAAm‐co‐PAAc‐co‐HEMA) nanogels conjugated with transferrin for tumor cell targeting delivery, Nanotechnology 2008, 19.

88. Hoare T. and Pelton R., Charge‐switching, amphoteric glucose‐responsive microgels with physiological swelling activity, Biomacromolecules 2008, 9: 733‐740.

89. Bhattacharya S., Eckert F., Boyko V., and Pich A., Temperature‐, pH‐, and magnetic‐field‐sensitive hybrid microgels, Small 2007, 3: 650‐657.

90. Lee E.S., Kim D., Youn Y.S., Oh K.T. et al., A virus‐mimetic nanogel vehicle, Angewandte Chemie‐International Edition 2008, 47: 2418‐2421.

91. Bae K.H., Mok H., and Park T.G., Synthesis, characterization, and intracellular delivery of reducible heparin nanogels for apoptotic cell death, Biomaterials 2008, 29: 3376‐3383.

92. Kohli E., Han H.Y., Zeman A.D., and Vinogradov S.V., Formulations of biodegradable Nanogel carriers with 5'‐triphosphates of nucleoside analogs that display a reduced cytotoxicity and enhanced drug activity, Journal of Controlled Release 2007, 121: 19‐27.

93. Vinogradov S.V., Kohli E., and Zeman A.D., Comparison of nanogel drug carriers and their formulations with nucleoside 5 '‐triphosphates, Pharmaceutical Research 2006, 23: 920‐930.

94. Morinloto N., Qiu X.P., Winnik F.M., and Akiyoshi K., Dual stimuli‐responsive nanogels by self‐assembly of polysaccharides lightly grafted with thiol‐terminated poly(N‐isopropylacrylamide) chains, Macromolecules 2008, 41: 5985‐5987.

95. Thornton P.D., Mart R.J., and Ulijn R.V., Enzyme‐responsive polymer hydrogel particles for controlled release, Advanced Materials 2007, 19: 1252..

96. Thornton P.D., Mart R.J., Webb S.J., and Ulijn R.V., Enzyme‐responsive hydrogel particles for the controlled release of proteins: designing peptide actuators to match payload, Soft Matter 2008, 4: 821‐827.

97. Ulijn R.V., Enzyme‐responsive materials: a new class of smart biomaterials, Journal of Materials Chemistry 2006, 16: 2217‐2225.

130


98. Owens D.E. and Peppas N.A., Opsonization, biodistribution, and pharmacokinetics of polymeric nanoparticles, International Journal of Pharmaceutics 2006, 307: 93‐102.

99. Cho K.J., Wang X., Nie S.M., Chen Z. et al., Therapeutic nanoparticles for drug delivery in cancer, Clinical Cancer Research 2008, 14: 1310‐1316.

100. Maeda H., The enhanced permeability and retention (EPR) effect in tumor vasculature: The key role of tumor‐selective macromolecular drug targeting, 2001, 189‐207.

101. Bontha S., Kabanov A.V., and Bronich T.K., Polymer micelles with cross‐linked ionic cores for delivery of anticancer drugs, Journal of Controlled Release 2006, 114: 163‐174.

102. Rijcken C.J., Snel C.J., Schiffelers R.M., van Nostrum C.F. et al., Hydrolysable core‐crosslinked thermosensitive polymeric micelles: Synthesis, characterisation and in vivo studies, Biomaterials 2007, 28: 5581‐5593.

103. Tamura A., Oishi M., and Nagasaki Y., Enhanced Cytoplasmic Delivery of siRNA Using a Stabilized Polyion Complex Based on PEGylated Nanogels with a Cross‐Linked Polyamine Structure, Biomacromolecules 2009, doi: 10.1021/bm900252d.

104. Yusa S., Sugahara M., Endo T., and Morishima Y., Preparation and characterization of a pH‐responsive nanogel based on a photo‐cross‐linked micelle formed from block copolymers with controlled structure, Langmuir 2009, 25: 5258‐5265.

105. Oishi M. and Nagasaki Y., Synthesis, characterization, and biomedical applications of core‐shell‐type stimuli‐responsive nanogels ‐ Nanogel composed of poly[2‐(N,N‐diethylamino)ethyl methacrylate] core and PEG tethered chains, Reactive & Functional Polymers 2007, 67: 1311‐1329.

106. Lemarchand C., Gref R., and Couvreur P., Polysaccharide‐decorated nanoparticles, European Journal of Pharmaceutics and Biopharmaceutics 2004, 58: 327‐341.

107. Lemarchand C., Gref R., Passirani C., Garcion E. et al., Influence of polysaccharide coating on the interactions of nanoparticles with biological systems, Biomaterials 2006, 27: 108‐118.

108. Li J., Yu S.Y., Yao P., and Jiang M., Lysozyme‐dextran core‐shell nanogels prepared via a green process, Langmuir 2008, 24: 3486‐3492.

109. Labarre D., Vauthier C., Chauvierre C., Petri B. et al., Interactions of blood proteins with poly(isobutylcyanoacrylate) nanoparticles decorated with a polysaccharidic brush, Biomaterials 2005, 26: 5075‐5084.

110. Ogris M. and Wagner E., Targeting tumors with non‐viral gene delivery systems, Drug Discovery Today 2002, 7: 479‐485.

111. Leamon C.P. and Low P.S., Folate‐mediated targeting: from diagnostics to drug and gene delivery, Drug Discovery Today 2001, 6: 44‐51.

112. Nayak S., Lee H., Chmielewski J., and Lyon L.A., Folate‐mediated cell targeting and cytotoxicity using thermoresponsive microgels, Journal of the American Chemical Society 2004, 126: 10258‐10259.

113. Na K., Park K.H., Kim S.W., and Bae Y.H., Self‐assembled hydrogel nanoparticles from curdlan derivatives: characterization, anti‐cancer drug release and interaction with a hepatoma cell line (HepG2), Journal of Controlled Release 2000, 69: 225‐236.

114. Kwon Y.J., James E., Shastri N., and Frechet J.M.J., In vivo targeting of dendritic cells for activation of cellular immunity using vaccine carriers based on pH‐responsive microparticles, Proceedings of the National Academy of Sciences of the United States of America 2005, 102: 18264‐18268.

131


132

biodegradable dextran nanogels for RNAi | CHAPTER 5

CHAPTER 5 | Biodegradable dextran nanogels for RNAi: focusing on endosomal escape and intracellular

siRNA delivery


Koen Raemdonck1, Broes Naeye1, Kevin Buyens1, Roosmarijn E. Vandenbroucke1, Anders Høgset2,

Joseph Demeester1 and Stefaan C. De Smedt1, Advanced Functional Materials 2009, 19: 1406‐1415.



2PCI Biotech, Hoffsveien 48, N‐0377, Oslo, Norway

133

CHAPTER 5 | biodegradable dextran nanogels for RNAi

134

ABSTRACT

The successful therapeutic application of small interfering RNA (siRNA) largely relies on the

development of safe and effective delivery systems that are able to guide the siRNA therapeutics to

the cytoplasm of the target cell. In this report, biodegradable cationic dextran nanogels are

engineered by inverse emulsion photopolymerization and their potential as siRNA carriers is

evaluated. The nanogels are able to entrap siRNA with a high loading capacity, based on electrostatic

interaction. Once inside the cells, the degradation of the nanogel core by ester hydrolysis again

releases the siRNA cargo. Confocal microscopy and flow cytometry analysis revealed that high

amounts of siRNA loaded nanogels can be internalized by Huh‐7 human hepatoma cells, without

significant cytotoxicity. Following their endocytic uptake, it is found that they are mainly trafficked

towards the endolysosomes. Since endosomal confinement of siRNA carrying nanoparticles is known

to be one of the most important intracellular barriers for efficient gene silencing, we investigated the

influence of two different strategies to enhance endosomal escape on the extent of gene silencing.

We found that both the application of photochemical internalization (PCI) and the use of an

influenza‐derived fusogenic peptide (diINF‐7), could significantly improve the silencing efficiency of

our siRNA loaded nanogels. Furthermore, it is shown that an efficient gene silencing requires the

degradation of the nanogels. As the degradation kinetics of the nanogels can easily be tailored, these

particles show potential for intracellular controlled release of small interfering RNA.



Chapter 5 provides:

• a protocol for the production of nanosized dextran hydrogel particles (nanogels) and their

subsequent loading with siRNA

• an evaluation of their cellular internalization and the subsequent intracellular siRNA release

• details regarding the improvement of intracellular siRNA delivery by application of validated

endosomolytic tools

Key terms:

Nanogel – siRNA – cationic – flow cytometry – PCI – fusogenic peptide ‐ luciferase

INTRODUCTION

RNAi is now generally regarded as a promising strategy to modulate aberrant expression of disease‐

causing genes in the treatment of among others, cancer, viral infections, inflammatory diseases, and

hypercholesterolemia.[1‐3] One approach to evoking RNAi is by the intracellular delivery of chemically

synthesized siRNAs. However, the relatively large size of siRNA duplexes (approximately 14 kDa; 1 Da

equals 1.66 x 10‐27 kg) and their negative charge (~40 ionized phosphate groups at physiological pH)

do not allow them to spontaneously cross cellular membranes. As a consequence, clinical

applications of siRNA therapeutics require the development of delivery strategies that can guide

intact siRNA into the cytoplasm of the target cell, where the RNAi machinery resides.[4,5] Both viral

and non‐viral carriers have been used for the intracellular delivery of RNAi molecules. Although viral

carriers are beyond doubt the most efficient delivery systems for nucleic acids, they still suffer from

135


important disadvantages, such as the risk of insertional mutagenesis [6], undesired immunogenicity [7]

and challenging and expensive large‐scale production. For these reasons, many research groups now

focus on non‐viral siRNA delivery. The use of non‐viral carriers is often based on electrostatic

complexation of the negatively charged siRNA with cationic lipids [8] (to form lipoplexes) or polymers

(to form polyplexes).[9‐11]

We previously reported cationic microgels (mean diameter ~2‐3 µm) that consist of dextran

hydroxyethyl methacrylate (dex‐HEMA, Figure 5.1a), which was copolymerized with cationic

methacrylate monomers. The functionalization of dextran with methacrylate moieties, as discussed

in Chapter 3, makes it feasible to incorporate different functionalities in the dextran hydrogel

network by copolymerization with varying methacrylate monomers. In this way, the physicochemical

properties of the hydrogel and its behaviour in a biological system may be adjusted according to the

final needs. HEMA is coupled to the dextran backbone through a carbonate ester that is subject to

hydrolysis under physiological conditions, which makes the cationic hydrogel network

biodegradable.[12,13] We showed in Chapter 3 that the micron‐sized dex‐HEMA gel particles allow a

slow release of encapsulated siRNA over several days.[14] In the present chapter, we aim to translate

this concept to the nanoscale by designing cationic dex‐HEMA nanogels in order to achieve time‐

controlled siRNA delivery in an intracellular environment. Cationic biodegradable dextran nanogels

were synthesized and their ability to complex and subsequently release the entrapped siRNA was

evaluated. Data on cellular internalization and intracellular trafficking were linked to the eventual

biological RNAi effect in a human hepatoma cell line. Since cationic nanogels are most likely

internalized through endocytosis, the effect of validated endosomolytic tools on the extent of gene

silencing was evaluated. Finally, to gain more insight into the exact mechanism of intracellular siRNA

delivery, we conducted control experiments with non‐degradable cationic dextran methacrylate

nanogels (dex‐MA, Figure 5.1a).


Preparation of cationic dextran nanogels

For the preparation of positively charged dextran nanogels, a mini‐emulsion photo‐polymerization

method was applied (Figure 5.1b). Briefly, 100 mg dextran hydroxyethyl methacrylate or dextran

methacrylate (dex‐HEMA or dex‐MA, Figure 5.1a) was dissolved in 20 mM HEPES buffer pH 7.4 and

supplemented with a known amount of the cationic methacrylate monomer [2‐(methacryloyloxy)‐

ethyl] trimethylammonium chloride (TMAEMA, Figure 5.1a, Sigma) to a final volume of 340 µL.

136


FIGURE 5.1 | (a) Chemical structure of dextran hydroxyethyl methacrylate (dex‐HEMA), [2‐(methacryloyloxy)‐ethyl]trimethylammonium chloride (TMAEMA) and dextran methacrylate (dex‐MA). Dex‐HEMA and dex‐MA were synthesized and characterized as previously reported [12]. (b) Dex‐HEMA, supplemented with a water‐soluble photoinitiator (irgacure 2959) and TMAEMA, is emulsified in mineral oil/ABIL EM 90 by ultrasonication to obtain a W/O emulsion. UV irradiation initiates the polymerization of dex‐HEMA and TMAEMA in the nanodroplets. The image on the right represents an atomic force microscopy (AFM) image of dex‐HEMA‐co‐TMAEMA nanogels (DS 21). The scale bar equals 1 µm.

Subsequently, 60 µL of the photoinitiator irgacure 2959 (1% w/v in HEPES buffer, Ciba Specialty

Chemicals), was added. This dextran phase was emulsified in 4 mL mineral oil (low viscosity, Sigma)

containing 10% v/v ABIL EM 90 surfactant (Degussa Goldschmidt) through ultrasonication (Branson

Tip Sonifier, 25 cycles, pulse on/off 1.0 s, and amplitude 10%). Immediately after ultrasonication the

formed nanodroplets were crosslinked by UV irradiation (900 s, Bluepoint 2.1 UV source, Hönle UV

technology). To remove the continuous phase, the emulsion was diluted 1:10 with a chilled mixture

of acetone/hexane (1:1 volume ratio). The precipitated nanogels were pelleted down by

centrifugation (3000 g, 5 min) and the supernatant was discarded. The nanogel pellet was washed

three times with the acetone/hexane mixture before being dispersed in 5 mL distilled water (4°C).

Traces of organic solvent were removed by vacuum evaporation at ambient temperature. Finally, the

nanogels were lyophilized and stored at ‐20°C. When needed for experiments, a weighed amount of

137


the lyophilized nanogels was dispersed in a known volume of buffer. Throughout this paper the

concentration of the nanogel suspensions is denoted as µg nanogel per mL buffer.

Atomic Force Microscopy (AFM)

AFM images were acquired on a PicoPlus instrument (Molecular Imaging) operating in tapping mode

(open air, ambient temperature). For sample preparation, a drop of nanogel dispersion was dried on

a silica slide overnight before imaging. Height images were flattened using WSxM Scanning Probe

Microscopy software (Nanotec electronica).

Dynamic Light Scattering (DLS)

Nanogel sizes, size distributions and surface charge were determined by dynamic light scattering

(DLS) and zeta‐potential measurements respectively, using a Zetasizer Nano ZS (Malvern,

Worcestershire, UK), equipped with Dispersion Technology Software (DTS). The samples were

equilibrated at 25°C prior to measurement and measurement settings were put on automatic. To

screen the degradation of the nanogels, nanogel suspensions (500 µg mL‐1) were prepared in

respectively 20 mM HEPES buffer (pH 7.4 or pH 6.0) and 20 mM sodium acetate buffer (pH 5.0) in a

low volume cuvette that was subsequently sealed with parafilm. The nanogel suspensions were kept

at 37°C and the intensity of the scattered light was measured every 5 minutes. Each measurement

time point consisted of 6 runs, with 5 s per run.

Small Interfering RNA (siRNA)

Two 21 mer siRNA duplexes from different providers, targeting the pGL3 firefly luciferase, were used

in our experiments: anti‐luc siRNA‐1 from Dharmacon (Lafayette, CO, USA) and pGL3 luciferase siRNA

from Eurogentec (Seraing, Belgium), respectively denoted as siLUCD and siLUCE. Alexa fluor 488

(AF488) labelled pGL3 luciferase siRNA and a non‐targeting negative control siRNA duplex

(siCONTROL) were also purchased from Eurogentec and siGLO® green fluorescent Transfection

Indicator was from Dharmacon (Lafayette, CO).

Preparation and characterization of siRNA loaded nanogels

In a first step, a nanogel suspension (1 mg mL‐1) was prepared by dispersing a weighed amount of

lyophilized particles in RNase free water or buffer, followed by short sonication (Branson sonifier,

30”‐10% amplitude). From this 1 mg mL‐1 stock solution, dilutions were prepared according to the

138


experimental needs. Nanogels were loaded with siRNA by mixing equal volumes of a diluted nanogel

dispersion and a siRNA solution at room temperature. The mixture was incubated for at least 15 min

at ambient temperature before initiating further experiments. The average size and zeta‐potential of

the nanogels before and after siRNA impregnation were measured by dynamic light scattering as

described above.

Fluorescence Fluctuation Spectroscopy (FFS)

A fluorescence fluctuation spectroscopy instrument was used as described previously.[15] FFS

monitors fluorescence intensity fluctuations in the excitation volume of a confocal microscope and

can be used to evaluate the complexation of fluorescently labelled molecules to cationic lipids or

polymers.[16‐19] The fluorescence fluctuations originate from the movement of (poorly concentrated)

fluorescent molecules in and out the fixed confocal excitation volume. In this study the 488 nm laser

line of a Krypton‐Argon ion laser (BioRad, Cheshire, UK) was used to excite the samples that contain

AF488‐siRNA. Typically, for FFS experiments, 50 µL of a nanogel dispersion was mixed with 50 µL of

an AF488‐siRNA solution (200 pmol mL‐1) in RNase free eppendorf tubes. After 20 min incubation, the

sample was transferred to the wells of a glass‐bottomed 96‐well plate (Greiner Bio‐one, certified

DNase/RNase free). Subsequently, the focal volume of the microscope was positioned 50 µm above

the bottom of the wells and the fluorescence fluctuations were recorded during a 30 s time interval.

All samples were prepared in triplicate and fluorescence measurements on every well were

performed in duplicate. The average fluorescence intensity of freely diffusing and complexed AF488‐

siRNA in the fluorescence fluctuation profile was determined as described previously.[16,20]

Cell lines and culture conditions

All cell experiments in this study were conducted on a human hepatoma cell line (Huh‐7). The wild

type cells are denoted as Huh‐7_wt. For luciferase gene silencing experiments, Huh‐7_eGFPLuc_rLuc

cells, stably expressing both the firefly and renilla luciferase (hereafter abbreviated as Huh‐7_LUC

cells), were generated by stably transfecting Huh‐7_eGFPLuc cells (gift from E. Wagner, Munchen)

with the pGL4.76_CMV plasmid. The pGL4.76_CMV plasmid was generated by ligating the PCR

amplified (forward primer AATAGTCGACTAGTTATTAATAGTAATCAA and reversed primer

AATAGGATCCGATCTGACG‐GTTCACTAAAC) and SalI/BamHI double digested CMV promotor into the

XhoI/BglII double digested pGL4.76 plasmid (Promega). The resulting pGL4.76_CMV plasmid was

linearized with the BamHI restriction enzyme and transfected using linear polyethylenimine (PEI) 22

kDa. Transfected cells were incubated in fresh medium for 48 h and then selected with 250 µg mL‐1

139


hygromycin. After two weeks, clones were isolated and expanded. Subsequently, the generated

clones were analyzed and a firefly and renilla luciferase positive clone was selected. Huh‐7_LUC cells

are grown in DMEM:F12, supplemented with 2 mM glutamine, 10% heat‐inactivated fetal bovine

serum (FBS) and 100 U mL‐1 P/S at 37°C in a humidified atmosphere containing 5% CO2.

Cytotoxicity assay

The influence of the nanogels on cell viability was analyzed using the CellTiter‐Glo® Assay (Promega).

Huh‐7_LUC cells (7 x 104 cells per cm2) were seeded in 24‐well plates and allowed to adhere. After 48

h, the cell culture medium was removed and 200 µL of a nanogel dispersion in OptiMEM was brought

onto the cells. Following 4 h of incubation (37°C, 5% CO2), nanogels that were not internalized were

removed from the cells and replaced with 200 µL of preheated culture medium. The 24‐well plates

were incubated for another 24 h at 37°C before cell viability was analyzed. Therefore, the cell plates

were incubated at room temperature during 30 min and 200 µL CellTiter‐Glo® reagent was added to

each well. After shaking the plate for 2 min (to lyse the cells) and 10 min incubation at room

temperature (to stabilize the luminescence signal), 100 µL of each well was transferred to a white‐

opaque 96‐well plate. Luminescence was measured on a GloMax™ 96 luminometer with 0.6 s

integration time. Cells treated with OptiMEM only and cells incubated with phenol (10 mg mL‐1)

served as negative and positive controls.

Confocal Laser Scanning Microscopy (CLSM)

Huh‐7_LUC cells were seeded in sterile glass‐bottomed culture dishes (MatTek Corporation, Ashland,

MA) at a density of 7 x 104 cells per cm2 and allowed to attach overnight. Nanogels loaded with

AF488‐siRNA were incubated with the cells in 200 µL OptiMEM during 4h (37°C, 5% CO2).

Subsequently, the cells were washed with phosphate buffered saline (PBS) and preheated culture

medium was added. The intracellular fluorescence distribution was visualized with a Nikon EZC1‐si

confocal scanning module installed on a motorized Nikon TE2000‐E inverted microscope (Nikon

Benelux, Brussels, Belgium). For lysosomal staining, cells were incubated for 20 min with 1:104

diluted Lysotracker Red® (Molecular Probes). Prior to image recording, the cells were fixed with a 4%

paraformaldehyde solution in PBS for 20 minutes at room temperature. After fixation, the cells were

washed twice with PBS and mounted with VectaShield® Mounting Medium (Vector Labs). Confocal

images were captured with a Nikon Plan Apochromat 60x oil immersion objective (numerical

aperture 1.4) using a 488 nm and 561 nm laser line for the excitation of AF488‐siRNA and

Lysotracker Red® respectively. A non‐confocal diascopic DIC (differential interference contrast) image

140


was collected simultaneously. An object based co‐localization algorithm was used to quantify the

number of nanogel particles that are located inside acidified endosomes. First, the AF488‐siRNA

loaded nanogels were identified in the green image by intensity thresholding. Only “objects”

consisting of at least 8 pixels were retained as nanogels. The thresholded image was then binarized in

order to generate a mask for superposition on the corresponding red image (containing the

Lysotracker Red® stained endosomes). This allows to calculate the red (endosomal) intensity R for

each gel particle. Next, the background intensity (mean value BG and standard deviation S) of the cell

in the image was determined within a rectangular area not containing any endosomes. Finally, a gel

particle is decided to be co‐localized with an acidified endosome if its red intensity R > BG+S.

Flow cytometry

Huh‐7_LUC cells were plated in 6‐well plates (7 x 104 cells per cm2) and treated two days later with

AF488‐siRNA loaded nanogels (200 pmol mL‐1 AF488‐siRNA), prepared as described above. After the

incubation period, the cells were washed twice with PBS, trypsinized (750 µL trypsin/EDTA 0.25% per

well) and diluted with 7 mL cell culture medium. Following centrifugation (7 min, 1200 rpm – 4°C),

the cell pellet was resuspended in 400 µL flow buffer (PBS supplemented with 1% BSA and 0.1%

sodium azide) and placed on ice until flow analysis. The cells were analyzed using a Beckman Coulter

Cytomics™ FC500 flow cytometer. Generally, a minimum of ~104 cells was analyzed in each

measurement. Dead cells were discriminated from viable cells by the addition of propidium iodide

(Sigma). The data were analyzed using Beckman Coulter CXP analysis software.

Luciferase silencing in human hepatoma cells

For gene silencing experiments, Huh‐7_LUC cells were seeded in 24‐well plates at a density of 7 x 104

cells per cm2 and allowed to attach during 48 h. Two hundred microliter of a siRNA loaded nanogel

dispersion (in OptiMEM) was added to the cells in each well (50 pmol of siRNA per well). After 4h

incubation (37°C), the nanogels were removed from the cells and replaced with 400 µL prewarmed

cell culture medium. Forty eight hours later, the cells were lysed with 85 µL passive lysis buffer

(Promega, Leiden, The Netherlands). Firefly and Renilla luciferase activities were determined with the

Promega Dual‐Luciferase® Reporter Assay according to the manufacturer’s instructions. For

luciferase analysis we used a GloMax™ 96 luminometer with two injectors. Briefly, 10 µL of cell

lysate was mixed with 50 µL firefly luciferase substrate and after a 2 sec delay, firefly luminescence

(relative light units, RLU) was integrated during 10 s. Subsequently, 50 µL of renilla luciferase

substrate in Stop & Glo Buffer (Promega), was added and again after 2 s renilla luminescence (RLU)

141


was measured during a 10 s interval. For each sample, the ratio of firefly luminescence to renilla

luminescence was calculated. The ratios obtained with siLUC nanogels were compared to the ratios

obtained with siCONTROL nanogels and expressed as remaining luciferase expression (%). As a

positive control, siRNA complexes based on Lipofectamine™RNAiMAX (Invitrogen) were used and

prepared according to the manufacturer’s instructions. The time‐dependent gene silencing

experiment was conducted as described above albeit that 5 x 104 cells per cm2 were plated and the

transfection was performed in 5‐fold. Every day one 24‐well plate was analyzed for luciferase

expression.

Fusogenic peptide diINF‐7

The influenza‐derived fusogenic peptide diINF‐7 [21,22] was a kind gift from S. Oliveira (Utrecht

University). For luciferase gene silencing, Huh‐7_LUC cells were plated in 24‐well plates (7 x 104 cells

per cm2). After 48 h, a dispersion of siRNA loaded nanogels was prepared as described above (with

15 min incubation) and subsequently mixed with the fusogenic peptide (15 µg mL‐1). The mixture was

left at room temperature for an additional 15 minutes before incubation with the cells (200 µL per

well). After 4h incubation (37°C), the nanogels were removed from the cells and replaced with 400 µL

prewarmed cell culture medium. Luciferase silencing was analyzed two days later as described

above.

Photochemical internalization (PCI)

The photosensitizer (PS) meso‐tetraphenylporphine disulfonate (TPPS2a) was kindly provided by Dr.

A. Høgset (PCI Biotech, Oslo, Norway). The PS was protected from light and stored at 4°C until use.

Huh‐7_LUC cells were seeded in 24‐well plates at a density of 7 x 104 cells per cm2. One day after

seeding, the cells were incubated overnight with 250 µL PS (0.4 µg mL‐1 in complete cell culture

medium) at 37°C and protected from light. The next day, the cells were washed with PBS and 200 µL

siRNA loaded nanogels was added to the cells. Following 4 h of incubation, the nanogel complexes

were replaced with 400 µL fresh cell culture medium and the cells were exposed to blue light (375

nm – 450 nm) emitted by LumiSource® (Osram 18W/67, 13 mW per cm2, PCI Biotech, Oslo, Norway).

Cells were incubated for another 48 h in the dark before luciferase activity was quantified as

described above.

Fluorescence microscopy before and after PCI, was evaluated by seeding 12.5 x 104 Huh‐7_wt cells in

Lab‐Tek™ II chambered coverglasses (Nalge Nunc International, Rochester, NY), two days before

transfection. One day pre‐transfection, PS was added to half of the wells (0.4 µg mL‐1). At the day of

142


transfection, the PS was removed and the cells were washed with PBS. Cationic nanogels were

loaded with fluorescent siGLO® in HEPES buffer as indicated earlier and incubated with the cells in

400 µl OptiMEM. The final concentration of nanogels and siGLO® was 25 µg mL‐1 and 200 pmol mL‐1

respectively. After 3 h incubation, the remaining complexes were removed and replaced with

complete cell culture medium (37°C). One hour of incubation later, the culture chambers were

illuminated with LumiSource® (45s). The cells were placed at 37°C for another 36 h, protected from

light. Afterwards, the cell‐bound siGLO® fluorescence was quenched by treatment with 0.4% trypan

blue (8 min) and after washing with PBS the cells were fixed with 4% paraformaldehyde (20 min, RT).

Following 3 washing steps with PBS, the cells were mounted with Vectashield® DAPI Mounting

Medium (Vector Labs, UK) and kept at 4°C (dark) until fluorescence imaging.


Preparation of cationic nanogels

Several methods for the engineering of nanogels are described in the literature.[23] We synthesized

dextran nanogels by using an inverse mini‐emulsion photopolymerization method, as schematically

depicted (Figure 5.1b). In this process, the internal aqueous dex‐HEMA phase, supplemented with

TMAEMA (Figure 5.1a) and photoinitiator, is dispersed in the continuous mineral oil phase with the

aid of an oil‐soluble surfactant (ABIL EM 90, HLB value ~5). Sonification provides the energy that is

necessary to form nanoscopic droplets.[23] Exposure to UV light initiates the radical polymerization

process within the dex‐HEMA droplets, yielding stable crosslinked dextran nanogels. Using this

procedure, we obtained dextran nanogels with a hydrodynamic diameter ~200‐300 nm with an

acceptable polydispersity index (Table 5.1).

We also visualized the nanogels by atomic force microscopy (Figure 5.1b, scale bar = 1 µm) showing

particles with a size in good agreement with the values obtained through dynamic light scattering.

Following polymerization, the particles were purified through centrifugation/washing steps with

organic solvent. Finally, the obtained aqueous dispersion of nanogels was lyophilized. We observed

that lyophilization had no significant influence on nanogel size and surface charge (data not shown).

Adding TMAEMA to the aqueous dex‐HEMA solution before emulsification enables copolymerization

of this methacrylate monomer with dex‐HEMA in the emulsion droplets, leading to nanogels with a

cationic surface charge, as revealed from zeta‐potential measurements (Table 5.1). Increasing the

amount of TMAEMA increased the surface charge of the nanogels, indicating that more TMAEMA

units get incorporated in the nanogel network.

143


TABLE 5.1 | Size, polydispersity and surface charge of dex‐HEMA‐co‐TMAEMA nanogels, with a degree of substitution (DS) of 3.2 and variable amounts of TMAEMA. The nanogels were dispersed in HEPES buffer (pH 7.4, 20 mM). Mean values with corresponding standard deviations on the measurement are shown (n=5).

dextran sample TMAEMA (µmol) dh (nm)a PDIb ζ‐potential (mV)

dex‐HEMA DS 3.2 0 259 ± 3 0.163 ± 0.023 ‐0.9 ± 0.3

dex‐HEMA DS 3.2 100 197 ± 1 0.193 ± 0.010 13.2 ± 0.6

dex‐HEMA DS 3.2 300 200 ± 2 0.185 ± 0.025 21.4 ± 0.5

dex‐HEMA DS 3.2 500 267 ± 8 0.274 ± 0.026 27.9 ± 0.9

dex‐MA DS 5.4 500 180 ± 5 0.142 ± 0.012 29.5 ± 0.3

a hydrodynamic diameter, b polydispersity index

In our further experiments we focused on dex‐HEMA‐co‐TMAEMA nanogels prepared as described in

the experimental section using 500 µmol TMAEMA.The cationic dextran microgels described in

chapter 3, contain 2‐(dimethylamino)ethyl methacrylate (DMAEMA) in their crosslinks instead of

TMAEMA. The only difference between both methacrylate monomers is the extra methyl group on

the nitrogen, which provides TMAEMA with a quaternary amine function (Figure 5.1b). In this way,

dex‐HEMA‐co‐TMAEMA nanogels carry a fixed positive charge independent of environmental pH,

whereas the charge density of dex‐HEMA‐co‐DMAEMA nanogels increases when the pH drops below

7.4, due to additional protonation of the tertiary amines (data not shown). It is indeed shown in

literature that the pKa of pDMAEMA is < 8, depending on the molecular weight of the polymer, so we

can expect that not all the DMAEMA units in the hydrogels will be protonated at physiological

pH.[24,25] Since the eventual charge density of the hydrogel network could also influence the

degradation rate of the dex‐HEMA carbonate esters, we chose to use TMAEMA for copolymerization,

thereby excluding any influence of pH on the hydrogel network characteristics.

Evaluation of the nanogel degradation kinetics as a function of pH

To assess the (bio)degradation of the nanogels, through hydrolysis of the hydrogel network, we

measured the decrease in turbidity of the nanogel dispersion as a function of time. The experiment

was performed at 37°C and at different pH values, mimicking the micro‐environment that the

nanogels may encounter upon cellular internalization through an endocytotic process. As expected,

dex‐HEMA‐co‐TMAEMA nanogels show a pH‐dependent degradation, with slower hydrolysis at acidic

pH when compared to physiological pH (Figure 5.2).

144


FIGURE 5.2 | Degradation kinetics at 37°C of dex‐HEMA‐co‐TMAEMA nanogels (DS 3.2, 500 µmol TMAEMA) as a function of pH as measured by dynamic light scattering. Data are normalized to value 1 to allow comparison. As a control, non‐degradable dex‐MA‐co‐TMAEMA nanogels (DS 3.4, 500 µmol TMAEMA) were analyzed under identical conditions.

This observation is in agreement with the pH‐dependent kinetics of dex‐HEMA hydrolysis as

published earlier.[12,26] As a control, we repeated the experiment at pH 7.4 with cationic dextran

methacrylate nanogels, with a comparable crosslinking density (dex‐MA‐co‐TMAEMA, DS 3.4, Table

5.1). Note that the lack of a carbonate ester between the methacrylate moiety and the dextran

backbone (Figure 5.1a) makes dex‐MA stable at physiological conditions.[26] As is shown in Figure 5.2,

the light scattered by a dex‐MA nanogel dispersion at pH 7.4 only slightly decreases over time

(~10%), possibly due to some non‐specific adsorption of the nanogels to the sides of the sample

cuvette.

In our earlier studies on cationic dex‐HEMA microgels, we showed that network degradation and

siRNA release are governed by hydrolysis of the carbonate ester linking the HEMA group to the

dextran backbone.[14] Microgels with a higher crosslink density degrade slower and therefore showed

a slower siRNA release profile. Dex‐MA microgels are undegradable and thus did not release any of

the electrostatically incorporated siRNA.[14] In vitro, we can expect the same behaviour for cationic

dex‐HEMA and dex‐MA nanogels.

145


Quantification of siRNA complexation to cationic dextran nanogels

The data in chapter 3 demonstrated that cationic dextran microgels could be loaded with siRNA after

microgel formation, based on electrostatic interactions.[14] This ‘post‐loading’ step has the advantage

that the siRNA molecules are not exposed to free radicals and UV light during the

photopolymerization process, thereby avoiding potential interference with the siRNA integrity.[14]

Similarly, cationic dex‐HEMA‐co‐TMAEMA nanogels could be post‐loaded with siRNA by dispersing

the nanogels in an siRNA solution.

We quantified the amount of siRNA that can be complexed by dex‐HEMA‐co‐TMAEMA nanogels by

fluorescence fluctuation spectroscopy (see experimental section), as described earlier by Buyens et

al.[16] In a solution of free (thus non‐complexed) AF488 labelled siRNA, the fluorescence signal in the

confocal excitation volume of the microscope fluctuates around an average value that is proportional

to the AF488‐siRNA concentration (Figure 5.3a). Upon complexation of the siRNA to cationic

nanogels, the fluorescence intensity of the ‘baseline’, which is representative for the concentration

of free siRNA, lowers and peaks of high fluorescence appear (Figure 5.3b). The latter originate from

the movement of highly fluorescent nanogel particles that complex many fluorescent siRNA

molecules, through the excitation volume. By determination of the free siRNA concentration,

through analysis of the average fluorescence intensity of the baseline, the percentage of siRNA in a

complexed state can be quantified.[16] Figure 5.3c shows the degree of siRNA complexation to

different concentrations of dex‐HEMA‐co‐TMAEMA nanogels, dispersed in respectively HEPES buffer

and OptiMEM. In every FFS experiment, the final siRNA concentration was 100 pmol mL‐1. We

estimated the maximal siRNA loading for the nanogels to be between 50 and 100 pmol siRNA per µg

nanogel in HEPES buffer which corresponds to >60 wt% loading. A closer look to Figure 5.2c shows

that in OptiMEM the maximal siRNA loading of the nanogels is slightly lower than in HEPES buffer,

probably due to partial screening of the cationic charges on the nanogels by the higher ionic strength

of OptiMEM.

Influence of siRNA complexation on nanogel size and surface charge

Besides information on the degree of siRNA complexation, it is equally important to evaluate

whether the loading of the nanogels with siRNA influences their size and surface charge. Indeed,

many studies have reported that these physicochemical parameters significantly influence the extent

and method of cellular internalization of nanoparticulate matter.[9,27,28] Negatively charged siRNA

duplexes will bind electrostatically to the charged ammonium groups incorporated in the nanogel

network.

146


FIGURE 5.3 | Analysis of AF488‐siRNA complexation by dex‐HEMA‐co‐TMAEMA nanogels (DS 3.2, 500 µmol TMAEMA) with fluorescence fluctuation spectroscopy (FFS). (a) Fluorescence fluctuation profiles obtained on freely diffusing AF488‐siRNA (left graph) and complexed AF488‐siRNA (right graph, 3 µg mL‐1 nanogels) in OptiMEM. The concentration of fluorescent siRNA equaled 100 pmol mL‐1 in both samples. (b) Quantification of AF488‐siRNA complexation by dex‐HEMA‐co‐TMAEMA nanogels in HEPES buffer (gray line) and OptiMEM (black line). Line graphs represent a sigmoidal fit (Origin 5.0). The concentration of fluorescent siRNA equaled 100 pmol mL‐1 in all the samples.

Figure 5.4 displays the change in size and zeta‐potential of dex‐HEMA‐co‐TMAEMA nanogels upon

the addition of siRNA. In this experiment the nanogel concentration was kept constant (20 µg mL‐1)

but increasing amounts of siRNA were added. The addition of siRNA to a nanogel dispersion causes a

gradual drop of the nanogels’ zeta‐potential as more and more siRNA molecules will reside at the

particle surface. At an siRNA concentration of 0.5 µM in Figure 5.4 (which corresponds to a loading of

25 pmol siRNA per µg nanogels) charge neutralization occured in HEPES buffer, resulting in the

147


formation of large aggregates. Further increasing the amount of siRNA reversed the charge of the

particles, which again stabilizes the dispersion. Since we observed the maximal loading to be > 50

pmol siRNA per µg nanogels (Figure 5.3) and as charge neutralization already occurred at 25 pmol

siRNA per µg nanogels (Figure 5.4), our data indicate that the binding of siRNA molecules to the

nanogels is not completely restricted to the surface of the nanogels. Either siRNA molecules have the

ability to penetrate deeper into the nanogel core, or some duplexes may only be partially adsorbed

onto the surface. Our previous data on cationic microgels seem to confirm the former hypothesis.[14]

FIGURE 5.4 | Size and zeta‐potential of dex‐HEMA‐co‐TMAEMA nanogels (20 µg mL‐1) after titration with siRNA. (■) normalized size in HEPES buffer 20 mM, pH 7.4 (□) size in OptiMEM, relative to values measured in HEPES.

(○) ζ‐potential measured in HEPES buffer. The nanogels were prepared with dex‐HEMA DS 3.2 and 500 µmol TMAEMA.

Size data obtained on nanogels in OptiMEM revealed that upon addition of siRNA to the cationic

nanogels, also aggregation can occur, but apparently to a much lesser extent than in HEPES buffer

(Figure 5.4). It is important to emphasize that Figure 5.4 also reveals that under appropriate mixing

conditions, siRNA loaded nanogels can be obtained which roughly have the same hydrodynamic size

as the nanogels before loading with siRNA. This is often not the case for polyplex nanoparticles of

which the size can be highly unpredictable and strongly dependent on the ratio of siRNA to cationic

polymer.

148


Cytotoxicity of cationic dextran nanogels

Before performing gene silencing experiments, we assessed the acute cytotoxicity of the dextran

nanogels on Huh‐7_LUC hepatoma cells, by quantifying the intracellular ATP levels 24 h after

transfection. The ATP concentration is regarded as a more sensitive marker for cell viability than the

commonly used MTT‐test.[29] In the concentration range tested, we found that dex‐HEMA‐co‐

TMAEMA nanogels only slightly reduce cell viability with a cytotoxicity of 12% and 27% when

respectively a 25 µg mL‐1 and 50 µg mL‐1 nanogel dispersion was applied on the cells (Figure 5.5).

Figure 5.5 | Cell viability of Huh‐7_LUC hepatoma cells, incubated during 4 h with dex‐HEMA‐co‐TMAEMA nanogels (DS 3.2, 500 µmol TMAEMA). The intracellular ATP‐levels were determined 24 h after the nanogels were removed from the cells. The metabolic activity of non‐treated cells was set at 100% and all data are shown as mean ± SD (n=4).

Note that the data in Figure 5.5 are based on ‘empty’ nanogels, i.e. not loaded with siRNA molecules.

Complexation with siRNA can influence surface charge and particle size (Figure 5.4), which are known

to govern cellular binding and uptake, and may therefore also influence the outcome of a cytotoxicity

assay.

Cellular binding and internalization of siRNA loaded nanogels by Huh‐7 hepatoma cells

To be effective in the silencing of a pathogenic gene, siRNA therapeutics need to be delivered in the

cell cytoplasm to enable interaction with the RNAi machinery. To study the binding of the nanogels

to the cellular membrane and the subsequent cellular uptake, we applied flow cytometry and

confocal laser scanning microscopy.

149


Figure 5.6a shows the histograms as obtained by flow cytometry on Huh‐7_LUC cells after they were

incubated during 4 h with respectively free and complexed AF488‐siRNA. Both lipofectamine based

siRNA lipoplexes (LPXs) and siRNA loaded dex‐HEMA‐co‐TMAEMA nanogels resulted in a highly

increased mean fluorescence of the Huh‐7_LUC cells (Table 5.2 and Figure 5.6a) when compared

with naked (i.e. non‐complexed) siRNA.

FIGURE 5.6 | (a) Flow cytometry histograms obtained on Huh‐7_LUC cells, treated with respectively free AF488‐siRNA and dex‐HEMA‐co‐TMAEMA nanogels (DS 3.2, 500 µmol TMAEMA) loaded with AF488‐siRNA. Lipofectamine™RNAiMAX:siRNA lipoplexes (LPXs) were also included as a control. The siRNA concentration was 200 pmol mL‐1 in all samples. (b) Non‐confocal DIC image (left) and CLSM image (right) of Huh‐7_LUC cells after 4 h incubation with 7.5 µg mL‐1 nanogels. The AF488‐siRNA concentration was fixed to 100 pmol mL‐1. The scale bar corresponds with 20 µm.

Even a low nanogel concentration of 5 µg mL‐1, already led to 95% positive cells. However, further

increasing the concentration of nanogels applied on the cells, also significantly further enhanced the

150


mean fluorescence values (Table 5.2). We found that quenching of the extracellular fluorescence

with 0.2% trypan blue did not significantly lower the mean fluorescence intensity, indicating that the

siRNA loaded nanogels were effectively internalized by the hepatoma cells (data not shown).

TABLE 5.2 | Flow cytometry data on Huh‐7_LUC cells, treated with respectively free AF488‐siRNA and AF488‐siRNA complexed to lipofectamine™RNAiMAX and dex‐HEMA‐co‐TMAEMA nanogels (DS 3.2, 500 µmol TMAEMA). The final siRNA concentration was 200 pmol mL‐1.

sample name AF488 positive cells (%) mean AF488 fluorescence (a.u.)

mock control (OptiMEM) 1 0.502

uncomplexed AF488‐siRNA 1.1 0.536

lipofectamine™RNAiMAX 79.9 11.3

nanogels (5 µg mL‐1) 95.0 35.6

nanogels (10 µg mL‐1) 97.4 54.2

nanogels (20 µg mL‐1) 99.0 104

Non‐phagocytic cells are generally thought to internalize cationic nanoparticles preferably through

endocytosis.[9] A recent report revealed that cationic groups on lipoplexes and polyplexes can

interact with negatively charged heparan sulphate proteoglycans on the cell surface, thereby

triggering a clathrin‐ and caveolin‐independent cellular uptake process.[30] This eventually confines

the particles in late endosomes, with a luminal pH of 5.0.[31] The latter can further fuse with

prelysosomal vesicles, containing hydrolyzing enzymes, eventually forming mature degrading

endolysosomes. On the other hand, macropinocytosis may also be involved in the cellular uptake of

these types of particles. The predominant type of cellular internalization and therefore subsequent

intracellular trafficking is highly cell type dependent and rigorous labour‐intensive experiments are

required to gain insight in this matter (D. Vercauteren and R.E. Vandenbroucke, unpublished data).

Nonetheless, it is generally assumed that the intracellular confinement of nucleic acid carriers in

membrane‐sealed vesicles is a main bottleneck in achieving efficient delivery into the cytosol.

To visualize the cellular entry of the siRNA loaded nanogels, we applied confocal laser scanning

microscopy (CLSM). Figure 5.6b shows a typical CLSM image as obtained after a 4 hours incubation of

Huh‐7_LUC cells with AF488‐siRNA loaded nanogels. One can observe a punctuated non‐

homogeneous fluorescence pattern, indicating the sequestration of the nanogels in endocytic

vesicles. Confocal z‐stack imaging revealed that after 4 hours of incubation the majority of the

151


nanogels are effectively internalized by the cells and almost no particles are remaining at the cell

surface (data not shown). To evaluate the intracellular fate of the nanogels we performed co‐

localization experiments with Lysotracker Red®, a fluorescent marker for acidified vesicles. The data

are shown in Figure 5.7. There is a clear co‐localization of the green (Figure 5.7b, AF488‐siRNA) and

red (Figure 5.7c, Lysotracker Red®) fluorescent signal, which confirms the endocytic uptake of the

siRNA loaded nanogels (yellow fluorescence in Figure 5.7d).

FIGURE 5.7 | Confocal images of the intracellular distribution of AF488‐siRNA loaded dex‐HEMA‐co‐TMAEMA nanogels in Huh‐7_LUC hepatoma cells. (A) Non‐confocal DIC image, (B) AF488‐siRNA nanogels, (C) Lysotracker

Red® labelled endolysosomes, (D) overlay image of green and red fluorescence. AF488‐siRNA nanogels (10 µg mL‐1, 100 pmol mL‐1 siRNA) were incubated during 2 h, followed by a 2 h chase with cell culture medium before fixation. The scale bar represents 5 µm.

We also performed a more detailed co‐localization analysis using appropriate image processing

software (see experimental section). We found that approximately 85% of the analyzed siRNA‐

nanogels can be detected to co‐localize with labelled endolysosomes (data not shown, >2500

particles were included in the analysis). As mentioned above, the endosomal localization is not

expected to lead to efficient gene silencing as the siRNA released during nanogel degradation will be

likely destroyed by endolysosomal nucleases. Since the RNAi machinery resides predominantly in the

152


cytosol, the siRNA or siRNA loaded nanogels will have to escape from the endosomal compartment

(see further).

Luciferase gene silencing in Huh‐7_LUC hepatoma cells

The RNAi gene silencing potential of siRNA loaded dex‐HEMA‐co‐TMAEMA nanogels was tested in

human hepatoma cells (Huh‐7) that stably express both firefly and renilla luciferase (Huh‐7_LUC

cells). Targeting a stably expressed gene can be regarded as the better model for oncogene silencing

in tumour cells in comparison with the targeting of a transient gene that is not integrated in the

genome.[32,33] As demonstrated in Figure 5.8 (gray bars), the extent of luciferase gene silencing

depends on the amount of nanogels that is incubated with the hepatoma cells. Increasing the

nanogel concentration more strongly suppresses gene expression, most probably due to a higher

cellular binding and subsequent internalization as can be deduced from the mean fluorescence

values in Table 5.2. The latter will lead to a better intracellular bioavailability of functional siRNA,

thus resulting in a stronger luciferase silencing.

FIGURE 5.8 | Luciferase gene silencing by siRNA loaded (siLUCE) dex‐HEMA‐co‐TMAEMA nanogels (DS 3.2, 500 µmol TMAEMA) without (black bars) and with (grey bars) the aid of PCI. Different concentrations of dex‐HEMA‐

co‐TMAEMA nanogels were applied (4 h incubation) and Lipofectamine™RNAiMAX was used as a positive control. The amount of siRNA per well was fixed to 50 pmol in all samples. The luciferase expression of cells, treated with siCONTROL loaded nanogels either with or without PCI, was set at 100%.

So, although a clear endosomal accumulation was observed in Figure 5.7, it seems that still a fraction

of the internalized siRNA is able to reach the cytosol. However, even at a nanogel concentration of 50

µg mL‐1 we observed a rather moderate gene silencing with siLUCE (~55%) which is still significantly

153


lower than that achieved with the positive control (lipofectamine based siRNA complexes) which

show ~75% silencing. Moreover, at a nanogel concentration of 50 µg mL‐1, cytotoxicity concerns arise

(Figure 5.5) and therefore it is imperative that appropriate measures are taken to achieve sufficient

gene silencing without interfering too much with cell viability.

Since the data on cellular uptake, presented in Figure 5.7, suggest that the nanogels are trafficked

toward the endolysosomes, it is likely that this intracellular fate poses a major barrier for efficient

RNAi. For this reason, we investigated the influence of photochemical internalization (PCI), a

technique known to enhance endosomal escape, on the luciferase gene silencing efficiency achieved

with dex‐HEMA‐co‐TMAEMA nanogels. PCI involves the use of amphiphilic photosensitizers (PS) that

accumulate in the membranes of endocytic vesicles.[34,35] In this study, meso‐tetraphenylporphine

carrying two sulfonate groups on adjacent phenyl rings (TPPS2a) was used. Upon illumination with a

specific light source (blue light in our experiments), excitation of the PS compound induces the

formation of reactive oxygen species (ROS), primarily singlet oxygen. This highly reactive

intermediate can damage cellular components, but this effect is mainly confined to the local

production site owing to its short range of action and short life‐time.[34] This localized effect will

therefore selectively disrupt the endosomal membranes, releasing the entrapped macromolecules or

nanoparticles into the cytosol (Figure 5.9). Since its discovery in 1999 [36], PCI has been successfully

applied to stimulate the cytosolic delivery of several types of macromolecules (peptides, proteins,

nucleic acids), incorporated in non‐viral carrier systems.[34,35,37‐39]

PCI experiments were conducted under conditions that induce moderate cytotoxicity on adherent

cells, i.e. 0.4 µg mL‐1 PS and 50s illumination time (data not shown and reference [37]). The effect of

PCI on gene silencing is shown in Figure 5.8 (white bars). Upon application of PCI a 60% and 78%

knockdown of luciferase expression was achieved, with 10 µg mL‐1 and 25 µg mL‐1 nanogels

respectively, in comparison with a 10% and 40% knockdown when the cells were not treated with

PCI. This large difference in gene silencing is a strong indication that the endosomal accumulation of

the siRNA loaded nanogels is a significant barrier for RNAi gene silencing. PCI had almost no influence

on the gene silencing by Lipofectamine™ RNAiMAX:siRNA complexes, suggesting that this carrier is

able to overcome the endosomal barrier by itself. Comparable results for Lipofectamine™2000 were

described by Bøe et al.[40]

To visually confirm the improvement in intracellular siRNA delivery after PCI treatment, we

conducted fluorescence microscopy experiments in which our cationic dextran nanogels were loaded

with siGLO® green transfection indicator.

154


FIGURE 5.9 | Schematic overview of photochemical internalization (PCI). In a typical PCI experiment, the amphiphilic photosensitizer TPPS2a accumulates in the endosomal membranes of the target cells (I). Excitation with blue light induces the local production of mainly singlet oxygen, that disrupts the endocytic vesicles (II). This rupture of the endosomal barrier allows internalized nanosized matter to be released into the cytoplasm (III).

This green fluorescent (FAM‐labelled) siRNA duplex is chemically modified to accumulate

predominantly in the cell nucleus following successful delivery in the cell cytoplasm. Figure 5.10 (top

row) shows fluorescence micrographs of Huh‐7_wt cells that have internalized a large amount of

siGLO‐nanogels before PCI treatment. However, the punctuated fluorescence pattern suggest

vesicular confinement, as already pointed out above, and as a result none of the cells show

fluorescent nuclei. In contrast, applying PCI on this cell population leads to a detectable fluorescence

in the cytoplasm and nuclei of a fraction of cells which are signs of successful endosomal escape

(Figure 5.10, bottom row, indicated by the white arrows). Contrast enhancement further accentuates

the difference in cytoplasmic and nuclear siGLO fluorescence between the control cells and the cells

treated with PCI. Clearly, PCI promotes the delivery of siRNA to the cytoplasm of the cell and hence

this observation further supports the enhancement of the RNAi effect we observed in Figure 5.8.

To additionally validate the conclusions drawn from the PCI data, we also tested a different approach

to enhance the endosomal escape, i.e. by adding a fusogenic peptide to the siRNA loaded dex‐HEMA‐

co‐TMAEMA nanogel dispersion. We used a dimeric influenza‐derived peptide (diINF‐7) [22], which has

previously been shown to enhance the oncogene silencing by non‐viral siRNA complexes.[21] Other

investigators also successfully applied the INF‐7 peptide or its dimeric form to enhance intracellular

gene delivery and gene expression by improving the endosomal escape of polyplexes or liposomal

gene carriers.[41‐43]

155


FIGURE 5.10 | Intracellular distribution of siGLO® green transfection indicator in dex‐HEMA‐co‐TMAEMA transfected Huh‐7_wt cells, before and after the application of PCI. The images on the right are obtained through contrast enhancement by histogram stretching in Nikon EZC1 freeviewer software.

INF‐7 resembles the NH2‐terminal domain of the HA‐2 haemagglutinin subunit and becomes

fusogenic at endosomal pH (~pH 5).[22] The peptide is attached to the nanogels through simple

electrostatic attraction. Addition of the peptide to a dispersion of siRNA loaded nanogels (25 µg mL‐1

and 40 µg mL‐1) markedly improved the silencing efficiency from 56% to 77% and from 69% to 81%

respectively, as shown in Figure 5.11. Note that in this experiment, a different siRNA duplex (siLUCD)

was used than in Figure 5.8, which tends to give better silencing compared to siLUCE. Taken together

with the results from the PCI experiment, one can conclude that the endosomal barrier is a

significant barrier for gene silencing by siRNA loaded dextran nanogels.

We sought to delineate the intracellular siRNA release mechanism in more detail by performing

control experiments with non‐degradable dex‐MA‐co‐TMAEMA nanogels. A comparative experiment

between degradable and non‐degradable nanogels under identical conditions revealed that

significant gene silencing can only be achieved when siRNA is loaded into nanogels with a

hydrolysable dextran backbone (Figure 5.12a).

We hypothesize that the accumulation of nanogel degradation products in the endosomal lumen is

able to disrupt the vesicular membrane, possibly through an osmotic effect, thereby releasing active

siRNA into the cytosol. However, the exact mechanism explaining how the siRNA eventually reaches

the cytoplasm is still unclear and remains a subject of future investigation. Figure 5.12b shows that

by applying PCI on the nanogels in the same concentration range, both degradable and non‐

degradable dextran nanogels are able to silence luciferase expression for up to 90%.

156


FIGURE 5.11 | Influence of the fusogenic peptide diINF‐7 on luciferase gene silencing by dex‐HEMA‐co‐TMAEMA nanogels (DS 3.2, 500 µmol TMAEMA). Every well contained 50 pmol siCONTROL (black bars) or siLUCD (gray bars). The addition of diINF‐7 to the mixture significantly improves the extent of luciferase gene silencing (white bars).

This result excludes insufficient cellular internalization of siRNA mediated by dex‐MA nanogels as

potential explanation for the measured difference in luciferase silencing in Figure 5.12a. This

observation could eventually be explained on the assumption that intact dex‐MA nanogels can reach

the cell cytoplasm after endosomal burst where charged proteins can interact with and displace

siRNA from the nanogel surface to make it available for the RNAi machinery. In vitro tests with

dynamic light scattering showed that applying PCI does not influence neither the size nor the surface

charge of the nanogels and this observation makes us believe that PCI mainly affects the integrity of

the endosomal vesicular membranes and leaves the nanogel particles in the endosomal lumen intact

(data not shown).

In a final experiment, we assayed the luciferase gene silencing as a function of time for a chosen

concentration of dex‐HEMA‐co‐TMAEMA nanogels (30 µg mL‐1 and 40 µg mL‐1) and a fixed amount of

siRNA. The results of this experiment are shown in Figure 5.13. We already observed a strong gene

silencing after 24 h of incubation, albeit that the maximal gene silencing for this nanogel sample

occurred two days after transfection. This maximal silencing is maintained for ~48 h after which the

gene silencing effect gradually decreases. Five days following transfection we still obtain 50%

downregulation of the luciferase expression, relative to siCONTROL loaded nanogels. The RNAi gene

silencing kinetics for different dextran nanogel samples is studied in more detail in the following

chapter.

157


FIGURE 5.12 | Luciferase gene silencing with biodegradable cationic dex‐HEMA and non‐degradable dex‐MA nanogels (DS 3.2, 500 µmol TMAEMA), (a) without PCI and (b) with PCI. The amount of siCONTROL (black bars) and siLUCD (gray bars) per well equaled 50 pmol.

CONCLUSIONS

Inhibiting the overexpression of highly active genes via siRNA delivery is an attractive approach for

the treatment of a wide variety of diseases. However, the application of siRNA in the clinic is still

limited since there is a lack of efficient and non‐toxic delivery systems. In this study, we report for the

first time on the preparation of cationic dextran based nanogels that can silence the luciferase gene

in hepatoma cells without relevant cytotoxicity. Following cellular internalization, the nanogels are

trafficked toward the endolysosomes. Our attempts to increase the effectiveness of the siRNA loaded

158


nanogels by enhancing endosomal escape, were successful. Photochemical internalization (PCI) and

an influenza‐derived fusogenic peptide (diINF‐7) could both significantly improve luciferase gene

silencing. This is the first evidence that biodegradable cationic nanogels are suitable for siRNA

delivery. More importantly, we showed that in the absence of endosomolytic tools, nanogel

degradation is a prerequisite to obtain a meaningful RNAi effect. This observation can pave the way

towards nanogel carriers with controllable degradation kinetics to achieve a prolonged and time‐

controlled RNAi gene silencing.

FIGURE 5.13 | Luciferase gene silencing with degradable dex‐HEMA‐co‐TMAEMA nanogels (DS 3.2, 500 µmol TMAEMA) as a function of time. In correspondence with previous experiments, the amount of siRNA was fixed to 50 pmol per well.

ACKNOWLEDGMENTS



(FP6) for funding of MediTrans, an integrated project on nanomedicines. A. Høgset is supported by

the Norwegian Research Council. Prof. Dr. K. Braeckmans is thanked for his help with the image

processing. The assistance of Magalie Ysebaert with the practical experiments is much appreciated.

The Huh‐7_eGFPluc cells were a kind gift from Prof. Dr. E. Wagner (LMU University, Munich). Prof.

Dr. D. Deforce and Dr. Franki are acknowledged for the use of the flow cytometer.

159


REFERENCE LIST

1. Aagaard L. and Rossi J.J., RNAi therapeutics: Principles, prospects and challenges, Advanced Drug Delivery Reviews 2007, 59: 75‐86.

2. Kim D.H. and Rossi J.J., Strategies for silencing human disease using RNA interference, Nature Reviews Genetics 2007, 8: 173‐184.

3. Grimm D. and Kay M.A., Therapeutic application of RNAi: is mRNA targeting finally ready for prime time?, J. Clin. Invest 2007, 117: 3633‐3641.

4. Novobrantseva T.I., Akinc A., Borodovsky A., and de Fougerolles A., Delivering silence: Advancements in developing siRNA therapeutics, Current Opinion in Drug Discovery & Development 2008, 11: 217‐224.

5. de Fougerolles A., Vornlocher H.P., Maraganore J., and Lieberman J., Interfering with disease: a progress report on siRNA‐based therapeutics, Nature Reviews Drug Discovery 2007, 6: 443‐453.

6. Bohne J. and Cathomen T., Genotoxicity in gene therapy: An account of vector integration and designer nucleases, Current Opinion in Molecular Therapeutics 2008, 10: 214‐223.

7. Zaiss A.K. and Muruve D.A., Immunity to adeno‐associated virus vectors in animals and humans: a continued challenge, Gene Therapy 2008, 15: 808‐816.

8. Akinc A., Zumbuehl A., Goldberg M., Leshchiner E.S. et al., A combinatorial library of lipid‐like materials for delivery of RNAi therapeutics, Nature Biotechnology 2008, 26: 561‐569.


10. Juliano R., Alam M.R., Dixit V., and Kang H., Mechanisms and strategies for effective delivery of antisense and siRNA oligonucleotides, Nucleic Acids Research 2008, 36: 4158‐4171.

11. Raemdonck K., Vandenbroucke R.E., Demeester J., Sanders N.N. et al., Maintaining the silence: reflections on long‐term RNAi, Drug Discov. Today 2008, 13: 917‐931.

12. van Dijk‐Wolthuis W.N.E., Tsang S.K.Y., Kettenes‐vandenBosch J.J., and Hennink W.E., A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer, Polymer 1997, 38: 6235‐6242.


14. Raemdonck K., Van Thienen T.G., Vandenbroucke R.E., Sanders N.N. et al., Dextran microgels for time‐controlled delivery of siRNA, Advanced Functional Materials 2008, 18: 993‐1001.

15. Raemdonck K., Remaut K., Lucas B., Sanders N.N. et al., In situ analysis of single‐stranded and duplex siRNA integrity in living cells, Biochemistry 2006, 45: 10614‐10623.


17. Remaut K., Lucas B., Raemdonck K., Braeckmans K. et al., Can we better understand the intracellular behavior of DNA nanoparticles by fluorescence correlation spectroscopy?, Journal of Controlled Release 2007, 121: 49‐63.

18. Lucas B., Remaut K., Braeckmans K., Haustraete J. et al., Studying pegylated DNA complexes by dual color fluorescence fluctuation spectroscopy, Macromolecules 2004, 37: 3832‐3840.

19. Lucas B., Van Rompaey E., De Smedt S.C., Demeester J. et al., Dual‐color fluorescence fluctuation spectroscopy to study the complexation between poly‐l‐lysine and oligonucleotides, Macromolecules 2002, 35: 8152‐8160.

20. Van Craenenbroeck E., Matthys G., Beirlant J., and Engelborghs Y., A statistical analysis of fluorescence correlation data, Journal of Fluorescence 1999, 9: 325‐331.

21. Oliveira S., van Rooy I., Kranenburg O., Storm G. et al., Fusogenic peptides enhance endosomal escape improving siRNA‐induced silencing of oncogenes, International Journal of Pharmaceutics 2007, 331: 211‐214.

22. Mastrobattista E., Koning G.A., van Bloois L., Filipe A.C.S. et al., Functional characterization of an endosome‐disruptive peptide and its application in cytosolic delivery of immunoliposome‐entrapped proteins, Journal of Biological Chemistry 2002, 277: 27135‐27143.

23. Oh J.K., Drumright R., Siegwart D.J., and Matyjaszewski K., The development of microgels/nanogels for drug delivery applications, Progress in Polymer Science 2008, 33: 448‐477.

24. van de Wetering P., Zuidam N.J., van Steenbergen M.J., van der Houwen O.A.G.J. et al., A mechanistic study of the hydrolytic stability of poly(2‐(dimethylamino)ethyl methacrylate), Macromolecules 1998, 31: 8063‐8068.

160


25. Van Tomme S.R., van Nostrum C.F., Dijkstra M., De Smedt S.C. et al., Effect of particle size and charge on the network properties of microsphere‐based hydrogels, European Journal of Pharmaceutics and Biopharmaceutics 2008, 70: 522‐530.

26. van Dijk‐Wolthuis W.N., van Steenbergen M.J., Underberg W.J., and Hennink W.E., Degradation kinetics of methacrylated dextrans in aqueous solution, J. Pharm. Sci. 1997, 86: 413‐417.

27. Rejman J., Oberle V., Zuhorn I.S., and Hoekstra D., Size‐dependent internalization of particles via the pathways of clathrin‐and caveolae‐mediated endocytosis, Biochemical Journal 2004, 377: 159‐169.

28. Gratton S.E.A., Ropp P.A., Pohlhaus P.D., Luft J.C. et al., The effect of particle design on cellular internalization pathways, Proceedings of the National Academy of Sciences of the United States of America 2008, 105: 11613‐11618.

29. Vandenbroucke R., De Geest B., Demeester J., De Smedt S. et al., Prolonged gene silencing in hepatoma cells and primary hepatocytes after siRNA delivery with biodegradable poly(alpha‐amino ester[s]), Human Gene Therapy 2007, 18: 1056‐1056.

30. Payne C.K., Jones S.A., Chen C., and Zhuang X.W., Internalization and trafficking of cell surface proteoglycans and proteoglycan‐binding ligands, Traffic 2007, 8: 389‐401.

31. Van Dyke R.W., Acidification of lysosomes and endosomes, Subcell. Biochem. 1996, 27: 331‐360. 32. Vandenbroucke R.E., De Geest B.G., Bonne S., Vinken M. et al., Prolonged gene silencing in hepatoma cells

and primary hepatocytes after small interfering RNA delivery with biodegradable poly(beta‐amino esters), J. Gene Med. 2008, 10: 783‐794.

33. Grayson A.C.R., Ma J., and Putnam D., Kinetic and efficacy analysis of RNA interference in stably and transiently expressing cell lines, Molecular Pharmaceutics 2006, 3: 601‐613.

34. Berg K., Folini M., Prasmickaite L., Selbo P.K. et al., Photochemical internalization: A new tool for drug delivery, Current Pharmaceutical Biotechnology 2007, 8: 362‐372.

35. Fretz M.M., Hogset A., Koning G.A., Jiskoot W. et al., Cytosolic delivery of liposomally targeted proteins induced by photochemical internalization, Pharm. Res. 2007, 24: 2040‐2047.

36. Berg K., Selbo P.K., Prasmickaite L., Tjelle T.E. et al., Photochemical internalization: A novel technology for delivery of macromolecules into cytosol, Cancer Research 1999, 59: 1180‐1183.

37. Oliveira S., Fretz M.M., Hogset A., Storm G. et al., Photochemical internalization enhances silencing of epidermal growth factor receptor through improved endosomal escape of siRNA, Biochimica et Biophysica Acta‐Biomembranes 2007, 1768: 1211‐1217.

38. Kloeckner J., Prasmickaite L., Hogset A., Berg K. et al., Photochemically enhanced gene delivery of EGF receptor‐targeted DNA polyplexes, Journal of Drug Targeting 2004, 12: 205‐213.

39. Oliveira S., Hogset A., Storm G., and Schiffelers R.M., Delivery of siRNA to the target cell cytoplasm: photochemical internalization facilitates endosomal escape and improves silencing efficiency, in vitro and in vivo, Curr. Pharm. Des 2008, 14: 3686‐3697.

40. Boe S., Longva A.S., and Hovig E., Photochemically induced gene silencing using small interfering RNA molecules in combination with lipid carriers, Oligonucleotides 2007, 17: 166‐173.

41. El‐Sayed A., Masuda T., Khalil I., Akita H. et al., Enhanced gene expression by a novel stearylated INF7 peptide derivative through fusion independent endosomal escape, J. Control Release 2009, doi: 10.1016/j.jconrel.2009.05.18.

42. Moore N.M., Sheppard C.L., Barbour T.R., and Sakiyama‐Elbert S.E., The effect of endosomal escape peptides on in vitro gene delivery of polyethylene glycol‐based vehicles, J. Gene Med. 2008, 10: 1134‐1149.

43. Jiang X., Lok M.C., and Hennink W.E., Degradable‐brushed pHEMA‐pDMAEMA synthesized via ATRP and click chemistry for gene delivery, Bioconjug. Chem. 2007, 18: 2077‐2084.

161


162

evaluation of gene silencing kinetics with siRNA loaded dextran nanogels | CHAPTER 6

CHAPTER 6 | Evaluation of gene silencing kinetics with

siRNA loaded dextran nanogels

Manuscript of this chapter is in preparation:

Koen Raemdonck1, A. Høgset2, Joseph Demeester1 and Stefaan C. De Smedt1



2PCI Biotech, Hoffsveien 48, N‐0377, Oslo, Norway

163

CHAPTER 6 | evaluation of gene silencing kinetics with siRNA loaded dextran nanogels

164



Chapter 6 provides:

• preparation and characterization of cationic dextran nanogels with varying gel network

density

• dose response suppression of EGFP with cationic nanogels

• the screening of RNAi knockdown kinetics with the aforementioned nanogels, in hepatoma

cells that stably express (1) firefly luciferase and (2) EGFP

• influence of photochemical internalization on the RNAi kinetics outcome

Key terms:

RNAi kinetics – nanogels – siRNA – EGFP – luciferase – degradation – PCI

INTRODUCTION

Since the discovery of RNA interference (RNAi) at the end of the 20th century,[1] intensive research on

RNAi therapeutics and their potential applications has led to an astonishing progress in just over a

single decade. The research investments resulted in a fast transition of promising lead siRNA

therapeutics from in vitro tests to clinical trials.[2‐4] Advances are made, both in the area of siRNA

design[5‐9] e.g. aiming for improved intrinsic siRNA activity as in the field of siRNA delivery. New

formulation strategies emerge, designed to overcome intra‐ and extracellular hurdles and to improve

the fraction of the administered siRNA dose that eventually reaches the target cell cytoplasm.[3,10,11]

165


In the majority of in vitro and in vivo assays to screen different siRNA delivery strategies, the maximal

gene suppression for a given siRNA dose is regarded as the main parameter to determine what

strategies should be taken into consideration. As already mentioned in previous chapters, not only

the maximal gene silencing effect at a fixed time‐point, but also and in our opinion more importantly,

the longevity of this effect must be routinely evaluated. While the magnitude of target gene (or

protein) suppression, relative to a control sample, may be a good indicator for the success of an

siRNA treatment, looking at the time‐window that the siRNA therapeutic intervention is able to keep

the intracellular protein titers below a certain threshold, is perhaps a better parameter to decide on

the efficacy of the treatment, even in vitro, and may also be used to compare different treatment

regimens and siRNA delivery strategies.[12] The duration of the silencing effect will eventually strongly

determine the dosage frequency in clinical therapy. Especially when long‐term RNAi activity is

necessary to obtain a desired downstream phenotype,[13] it should be common practice to check

upon the long‐term target gene suppression.

In the present chapter, dex‐HEMA‐co‐TMAEMA nanogels were prepared with dex‐HEMA of varying

methacrylate substitution and therefore also different nanogel crosslink densities.[14] The nanogel

batches were carefully characterised with regard to size, surface charge and siRNA interaction. In the

light of the comments made above, their gene silencing effect was analyzed as a function of time in

both luciferase and EGFP expressing human hepatoma cells. Finally, the influence of photochemical

internalization (PCI, see also chapter 5) on the gene silencing duration was assessed.


Preparation of dextran nanogels and loading with siRNA

Cationic dextran nanogels (dex‐HEMA‐co‐TMAEMA) were prepared using the UV induced emulsion

photopolymerization as described in chapter 5, with minor adjustments. Dex‐HEMA (65 mg) was

dissolved in a solution containing 40 µL irgacure 1% (w/v), 80 µL TMAEMA (75% w/v) and 80 µL

HEPES buffer (pH 7.4, 20 mM). The obtained dextran solution was emulsified in 5 mL pre‐cooled

mineral oil, supplemented with 5% (v/v) of ABIL EM 90 as emulsifier. Following sonification (Branson

sonifier, 30 s ‐ 20%), the emulsion was exposed to UV light during 15 min under cooling (~4°C). The

prepared nanogels were obtained in pure form by precipitation with acetone:hexane (1:1) and

centrifugation. After washing, the precipitate was redispersed in 5 mL distilled water and traces of

organic solvent were removed by vacuum evaporation. To assure long‐term stability, the nanogels

were lyophilized and stored exsiccated. To load the nanogels with siRNA, a stock dispersion of

166


nanogels (~2 mg mL‐1) was prepared in ice‐cooled HEPES buffer and sonicated to loosen the gel

particles (20 s, 10%). From this stock dispersion, dilutions are prepared according to the experimental

needs. Subsequently, equal volumes of nanogels and siRNA are mixed at room temperature and left

to stand at RT for ≥ 15 min to allow complexation.

Dynamic light scattering and zeta‐potential

The size and surface charge of the different nanogel dispersions, listed in Table 6.1, was analyzed

with DLS and zeta‐potential measurements respectively, as described earlier in this thesis. For more

details on the analysis of nanogel degradation, using the DLS instrument, we would like to refer to

the experimental section of chapter 5.

siRNA duplexes

Twenty‐one nt siRNAs targeting the pGL3 firefly luciferase gene (siLUC), the enhanced green

fluorescent protein (siEGFP) and a universal negative control duplex (siCONTROL) were purchased

from Eurogentec and stored in 20 µM aliquots in DEPC treated water (‐20°C). For flow cytometry and

fluorescence microscopy experiments, a green fluorescent siRNA duplex was ordered from

Dharmacon (siGLO® green transfection indicator).

Cell lines and culture conditions

All cell experiments were performed on Huh‐7 hepatoma cells of which the cell culture conditions

can be found in previous chapters. Huh‐7_LUC cells are cells that stably express the firefly and renilla

luciferase gene (see also chapter 5). Huh‐7_EGFP cells stably produce the enhanced green

fluorescent protein (see also chapter 3).

Luciferase gene silencing kinetics

Two days before transfection, Huh‐7_LUC cells were seeded in 96‐well plates (1 x 103 cells per well).

For transfection, first 90 µL of pre‐heated OptiMEM (37°C) was pipetted on the cells and

subsequently 10 µL of a siRNA loaded nanogel dispersion (in HEPES buffer) was added to the cells in

each well (20 pmol of siRNA). After 4 h incubation (37°C), the nanogels were removed from the cells

and replaced with 150 µL pre‐heated cell culture medium. This transfection protocol was performed

in duplicate (plate 1 and plate 2 respectively) for every sample tested. Twenty‐four hours later, the

167


cells on plate 1 were lysed with 20 µL passive lysis buffer (Promega, Leiden, The Netherlands). Firefly

and Renilla luciferase activities were determined with the Promega Dual‐Luciferase® Reporter Assay

as explained in chapter 5. The cells on plate 2 were trypsinized with 100 µL trypsin/EDTA (0.25%) and

replated in 5 separate 96‐well plates. Four of these plates are analyzed for luciferase expression on

day 2 till day 5 following transfection. On day 5, the trypsinization step was repeated on the

remaining 96‐well plate in order to avoid overgrowth of the cells and to maintain the cells in a state

of active cell division. Luciferase expression was additionally analyzed at day 6 and day 7.

Silencing EGFP with DS 2.5 cationic dextran nanogels

Two days before transfection, 7.5 x 104 Huh‐7_EGFP cells were seeded in 12‐well plates. At the day of

transfection, the culture medium was replaced with 900 µL pre‐heated OptiMEM. DS 2.5 siEGFP

nanogel complexes were prepared in HEPES buffer (~20 min incubation at room temperature) and

added to the wells to obtain a final volume of 1 mL. The nanogel and siEGFP concentration on the

cells eventually equalled 60 µg mL‐1 and 75 pmol mL‐1 respectively. As controls, empty nanogels or

nanogels loaded with siCONTROL and siLUC were included in the transfection experiment. After 5 h

incubation at 37°C, the transfection medium was replaced with 2 mL of complete cell culture

medium. Five days post‐transfection, the cells were washed with PBS and collected for flow

cytometry by trypsinization (500 µL trypsin/EDTA 0.25% per well, followed by dilution in 5 mL cell

culture medium). After centrifugation (1200 rpm, 7 min) the cell pellet was resuspended in 300 µL

flow buffer and kept on ice until flow analysis. The cells were analyzed using a Beckman Coulter

Cytomics™ FC500 flow cytometer. Dead cells were discriminated from viable cells by the addition of

propidium iodide (Invitrogen). The data were analyzed using Beckman Coulter CXP analysis software.

Silencing EGFP with DS 5.2 cationic dextran nanogels – fluorescence microscopy

1.5 x 104 Huh‐7_EGFP cells were grown on Lab‐Tek™ II chambered coverglasses (Nalge Nunc

International, Rochester, NY) two days pre‐transfection. As described above, the siRNA‐nanogel

complexes were prepared in HEPES buffer as a 10x stock dispersion. The transfection was performed

by incubating the cells during 4 h with 50 µg mL‐1 DS 5.2 nanogels, loaded with 40 pmol siEGFP in a

final volume of 400 µL OptiMEM and after this incubation period, the remaining complexes were

removed and 400 µL of fresh cell culture medium was added per well. The same controls as

indicated in the above paragraph were used. After 96 h, the cells were washed with PBS and fixed in

4% paraformaldehyde solution (20 min incubation at RT). Subsequently, after three additional

washing steps with PBS, the coverglasses were mounted with Vectashield® DAPI mounting medium

168


(Vector Labs). EGFP expression was visualized with fluorescence microscopy using a Nikon EZC1‐si

confocal scanning module installed on a motorized Nikon TE2000‐E inverted microscope (Nikon

Benelux, Brussels, Belgium ‐ see previous chapters) with open pinhole setting.

Dose response EGFP silencing with DS 5.2 cationic nanogels

Huh‐7_EGFP cells were seeded in 24‐well plates (3 x 104 cells per well). Two days after seeding, the

cells were transfected with 50 µg mL‐1 nanogels (500 µL in OptiMEM) during 4 h after which the

medium was replaced with 500 µL of fresh complete cell culture medium. We verified for the DS 5.2

nanogels that 4 h incubation with the cells was the optimal incubation time. Longer incubation did

not improve the RNAi activity but rather induced more cytotoxic effects (data not shown). The

amount of siEGFP per well was varied between 0.025 pmol and 125 pmol. Nanogels that complex

siLUC are used as a control. At day 3 post‐transfection, the cells were prepared for flow cytometry by

trypsinization (250 µL trypsin/EDTA 0.25% per well, followed by dilution in 1 mL cell culture medium).

After centrifugation (1200 rpm, 7 min) the cell pellet was resuspended in 200 µL flow buffer and kept

on ice until flow analysis as explained above.

EGFP gene silencing kinetics

Huh‐7_EGFP cells were seeded in 24‐well plates at a density of 5 x 104 cells per well. Twenty‐four

hours later, the cells were incubated during 4 h with DS 2.5, DS 5.2 and DS 8.9 nanogels (50 µg mL‐1),

loaded with siEGFP or siLUC (50 pmol per well) in a final volume of 500 µL (OptiMEM).

Lipofectamine™RNAiMAX complexes were prepared in OptiMEM and included in the transfection

experiment (1.56 µL RNAiMAX stock per well). For all samples, the transfection was performed in

duplicate. Subsequently, the transfection medium was replaced with 500 µL cell culture medium.

EGFP expression was quantified at different time‐points post‐transfection. At every time‐point, the

cells in the duplicate transfection were again replated in two new 24‐well plates (1/3 diluted) to be

analysed at the next time‐point.

Influence of photochemical internalization on EGFP gene silencing kinetics

The effect of PCI on the silencing duration was evaluated in two distinct protocols where PCI was

conducted at the day of transfection (PCI t0) or two days post‐transfection (PCI t2). More details on

the protocol time‐scale can be found in Figure 6.8. For the transfection, 5 x 104 cells are seeded in 24‐

well plates and left to grow for 48 h. Transfection of the cells (in 500 µL OptiMEM) was performed

169


with DS 5.2 cationic nanogels at varying concentration. The amount of siRNA (siEGFP and siLUC) per

well was 50 pmol. Reminiscent of previous transfections, an incubation period of 4 h with the cells

was respected.


Preparation and characterization of cationic dextran nanogels with different DS values

The main goal of this chapter is to evaluate the magnitude and duration of the RNAi effect of cationic

dextran nanogels with different crosslink density. In order to unambiguously draw conclusions from

the experiments performed, it is equally important to rigorously characterize the nanogel batches

used. Size and zeta‐potential values of the different batches can be found in Table 6.1.

TABLE 6.1 | Overview data on size and surface charge of the dex‐HEMA‐co‐TMAEMA nanogel dispersions used in this chapter. Analysis was performed in HEPES buffer (pH 7.4, 20 mM) with a final nanogel concentration of 0.2 mg mL‐1. Values are the average ± SD of 5 measurements on the same sample.

dextran sample dh (nm)a PDIb ζ‐potential (mV)

dex‐HEMA DS 2.5 254 ± 5 0.081 ± 0.055 24.1 ± 0.7

dex‐HEMA DS 5.2 212 ± 1 0.106 ± 0.029 24.8 ± 0.4

dex‐HEMA DS 8.9 207 ± 2 0.137 ± 0.034 17.1 ± 0.7

dex‐MA DS 3.4 219 ± 5 0.105 ± 0.012 30.2 ± 0.8 a hydrodynamic diameter, b polydispersity index

This table shows that all the nanogels have a hydrodynamic diameter that slightly exceeds 200 nm

and a zeta‐potential ~20‐30 mV, indicating successful copolymerization of dex‐HEMA with TMAEMA.

The chemical structures of these compounds can be found in Figure 5.1a. The surface charge

measured for the DS 8.9 sample deviates slightly to lower values in comparison with the other

nanogels.

To verify the non‐degradable nature of the dex‐MA DS 3.4 nanogels, DLS degradation data (see also

chapter 5) were compared between DS 3.4 (dex‐MA) and DS 5.2 (dex‐HEMA). Hydrogel degradation

is typically accompanied by a volume increase due to increasing water influx as a function of crosslink

cleavage. This swelling behaviour is clearly observed for the hydrolysable dex‐HEMA DS 5.2 sample

(Figure 6.1a), but is absent for the dex‐MA DS 3.4 sample. During swelling, a ~2.5 fold increase in size

170


is measured, corresponding with a ~15.5 fold increase in nanogel volume. As was expected based on

the observations in Figure 5.3, the degradation of the nanogels correlates with a loss of scattering

intensity although there is a clear increase in nanogel size. Apparently, the increase in transparency

during dex‐HEMA hydrolysis has a larger impact on the resultant particle scattering than the swelling.

Also this downwards trend in light scattering intensity was not observed for the non‐degradable

nanogel dispersion (Figure 6.1b).

FIGURE 6.1 | Evaluation of degradation behaviour of cationic dex‐HEMA DS 5.2 nanogels (white circles) and non‐degradable cationic dex‐MA DS 3.4 nanogels (black circles). (a) Changes in the size of both nanogels during incubation. Measurements are performed on nanogels at a final concentration of 0.5 mg mL‐1, in HEPES buffer pH 7.4, 20 mM at a temperature of 37°C. (b) Scattering intensity as a function of incubation time.

A prerequisite to successfully deliver siRNA across the plasma membrane and into the cytosol, is a

good complexation of the negatively charged siRNA to the cationic dex‐HEMA‐co‐TMAEMA nanogels.

Reminiscent of the data in Figure 5.3, we titrated the nanogel samples listed in Table 6.1 with siRNA

to evaluate if there are significant differences in their siRNA complexation behaviour. Apparently,

these differences do exist when different nanogels batches are prepared (Figure 6.2). Nanogels with

DS 5.2 (dex‐HEMA) and DS 3.4 (dex‐MA) display a decrease in zeta‐potential (onset of surface charge

neutralization and charge reversal) at substantially higher siRNA concentrations compared with DS

2.5 and DS 8.9 (both dex‐HEMA). Since the nanogels have approximately the same dimensions, this

observation suggests that the former nanogels are able to load more siRNA molecules per µg

nanogel. Possible explanations are that (a) due to variability in the production of the nanogels, the

same weighed amount of nanogel lyophilisate may contain a different absolute number of nanogel

particles, (b) due to a smaller network pore size (DS 8.9) or substantial dex‐HEMA hydrolysis during

nanogel production and handling (DS 2.5), the ability to complex siRNA decreases significantly.

Apparently, one has to be careful when linking zeta‐potential data to siRNA complexation capacity,

171


since other parameters need to be considered that may also influence the nanogels’ complexation

behaviour.

FIGURE 6.2 | Titration of nanogel dispersions with increasing amounts of siRNA. From left to right: DS 8.9 (up triangles) – DS 2.5 (down triangles) – DS 5.2 (circles) – DS 3.4 (squares). Line graphs represent a Boltzmann sigmoidal fit (Origin 5.0). DLS measurements are performed at 25°C in HEPES buffer 20 mM, pH 7.4.

RNAi gene silencing kinetics in Huh‐7_LUC cells

One of the drawbacks of non‐viral carriers for synthetic siRNA delivery, especially when long‐term

treatment is advised, is the transient nature of the gene silencing effect.[15] It has been stated in the

literature that the intracellular dilution of siRNA below a critical threshold, due to cell division, is

responsible for the fast recovery of gene expression in fast dividing cells.[12] On the other hand, major

concerns are raised with regard to adverse effects of siRNA therapeutics, such as unwanted immune

activation, [16] off‐target silencing [17] and saturation of the RNAi machinery.[18] As these effects are

concentration‐dependent it is obvious that delivery of siRNA requires rigorous control over the

intracellular siRNA concentrations.[15,19] As a consequence, non‐viral carriers that are able to control

the amount of siRNA that is released in the cytoplasm as a function of time could be of interest to

obtain a prolonged gene silencing and/or to minimize side‐effects. In addition, the ability to control

the duration of RNAi gene silencing could have many in vitro applications since a single transfection

experiment could suffice for reliable phenotypic analysis at later time‐points post‐transfection.

Evaluation of the RNAi kinetics is not yet ‘standard operating procedure’, but in our opinion should

be routinely tested for every new type of siRNA carrier as a function of cell type and protein target.

172


Dependent on the degree of protein suppression that is needed to observe the desired phenotype,

these kinetic data could provide the basis for the outline of a clinical treatment schedule.

In this chapter we aimed to investigate the kinetics of RNAi gene silencing for our various nanogel

dispersions, to identify possible variation in the RNAi outcome induced by differences in DS value and

aiming to gather more information to further optimize our concept of siRNA delivery through

biodegradable nanogels.

For every nanogel sample, first the expression of both firefly and renilla luciferase was followed for

the course of one week. To maintain the cells in a state of active cell division and prevent cellular

overgrowth, we replated the cells at day 1 and day 5 post‐transfection (Figure 6.3). As a positive

control, the RNAi gene silencing kinetics for lipofectamine™RNAiMAX was evaluated using exactly the

same protocol. Since the cellular uptake of siRNA varies as a function of the nanogel concentration

that is incubated with the cells, we screened different concentrations for every nanogel type, ranging

from 10 µg mL‐1 to 70 or 80 µg mL‐ 1.

Figure 6.3 shows that none of the siRNA carriers investigated was able to maintain sufficient gene

silencing up to 7 days. Although the DS 2.5, DS 5.2 and Lipofectamine™RNAiMAX achieve a maximal

gene silencing of > 75% two days post‐transfection, this level of gene suppression is only maintained

for an additional 24 h at most. Five days post‐transfection, on average gene expression has recovered

to ~50% compared to control (reminiscent of the luciferase silencing kinetics presented in Figure

5.13), and seven days post‐transfection, in most cases little gene suppression remains. Another

interesting observation is the lack of substantial gene silencing (< 50% knockdown) when performing

experiments with slowly degrading nanogels (DS 8.9). DS 8.9 nanogels reach their maximal effect at a

concentration of 60 µg mL‐1 after which no further improvement is observed. Also non‐degradable

nanogels (DS 3.4, dex‐MA) were not able to silence the luciferase gene to an appreciable extent

(Figure 6.4).

173


FIGURE 6.3 | Overview experiments on time‐dependent gene silencing of dex‐HEMA‐co‐TMAEMA nanogels as a function of DS with (a) DS 2.5, (b) DS 5.2, (c) DS 8.9, (d) lipofectamine™RNAiMAX. For comparison, data on nanogels (30 µg mL‐1 till 70 µg mL‐1) and lipofectamine™RNAiMAX (all 3 samples) were averaged and summarized in (e). Transfection experiments were performed in a 96‐well plate format with a fixed amount of siRNA (20 pmol) per well. Data points represent average values ± SD (n=4).

174


FIGURE 6.4 | Luciferase expression quantified 48 h after incubation (4 h) of Huh‐7_LUC cells with 50 µg mL‐1 nanogels as indicated. The final amount of siRNA per well (with 1 x 104 seeded cells in 96‐well plates) equaled 20 pmol. Results were normalized to nanogels loaded with siCONTROL (100% luciferase expression).

From these data we conclude that the degree of substitution (DS) of the nanogel dispersion

apparently has an important effect on the extent of gene silencing that can be obtained, but that it

has little to no influence on the silencing duration. It seems that an low initial nanogel density is

necessary to reach a satisfactory gene silencing outcome (DS 2.5 and DS 5.2), but that the difference

in crosslink density of the nanogels does not affect the recovery to steady state protein expression

levels. A plausible explanation to support these observations may be given by the pH‐dependency of

the dex‐HEMA degradation (Figure 5.3 and literature reports)[20] and the lack of a buffering effect for

the incorporated quaternary TMAEMA groups. It was shown in the previous chapter that upon

cellular entry, the siRNA loaded nanogels are trafficked towards acidified organelles (presumable

endolysosomes) where the pH value can drop down to ~5. Early data in the literature state that

hydrolysis of the dex‐HEMA carbonate ester linkage is minimal at this acidic pH and as a consequence

a substantial fraction of the internalized nanogels will remain undegraded.[20] Upon endocytosis, the

nanogels only have a short time‐window in which they are able to degrade, which is during their

presence in early endosomes, but transfer from early endosomes to late endosomes and lysosomes

likely inhibits further hydrolysis of the dex‐HEMA carbonate ester. Since we demonstrated in chapter

5, that nanogel degradation could improve siRNA release to the cell cytoplasm (also see Figure 6.4),

we could assume that a too slow degradation and/or a fast transition of the nanogels to acidified

vesicles may significantly impede their gene silencing potential.

Nevertheless, additional experimental evidence will be required to further substantiate this

hypothesis. Moreover, Table 6.1 and Figure 6.2 attribute a significantly lower zeta‐potential and

175


siRNA complexation capacity to the DS 8.9 nanogels, which leads to decreased cellular internalization

of siRNA in this concentration range in comparison with the other nanogels (flow cytometry

preliminary observation, data not shown). Also these observations will obviously have their

implications on the measured gene silencing outcome and therefore severely hamper a direct and

conclusive comparison of the maximal gene silencing effect between the different nanogel batches.

RNAi gene silencing kinetics in Huh‐7_EGFP cells

Besides investigating the RNAi outcome in Huh‐7_LUC cells, several experiments were also repeated

in Huh‐7_EGFP cells. Figure 6.5 proves that our siRNA loaded dextran nanogels (DS 2.5 and DS 5.2 are

investigated) are also able to induce a significant reduction of EGFP expression.

FIGURE 6.5 | (a) EGFP silencing with dex‐HEMA DS 2.5 nanogels (60 µg mL‐1), without siRNA addition (empty nanogels) or loaded with control siRNA (siCONTROL, siLUC) and active siRNA (siEGFP). The amount of siRNA added per well was 75 pmol per 7.5 x 104 seeded cells in 12‐well format and the EGFP expression was quantified 5 days post‐transfection. (b) The amount of siRNA was titrated with 50 µg mL‐1 DS 5.2 nanogels and the effect on the EGFP suppression is shown. Data are expressed as percentage EGFP expression relative to DS 5.2 nanogels loaded with siLUC. Expression of EGFP was quantified 72 h post‐transfection and transfection was performed in 24‐well plates with 3 x 104 seeded cells per well. (c) Visual control of EGFP silencing with DS 5.2 nanogels as evaluated with fluorescence microscopy.

176


Figure 6.5a and 6.5c confirm the sequence‐specificity of the downregulation effect, since none of the

controls used in these experiments, i.e. empty nanogels and nanogels loaded with either a universal

control siRNA (siCONTROL) or an irrelevant but validated siRNA (siLUC), showed deviant EGFP

expression. A dose response experiment with DS 5.2 nanogels demonstrated a clear influence of the

amount of siRNA used on the knockdown efficiency with an eventual IC50 value ~0.5 pmol siRNA

under the experimental conditions.

Following our kinetic data on Huh‐7_LUC cells, we also conducted a time‐dependent gene silencing

experiment with EGFP as RNAi target. A first observation is that the magnitude of EGFP silencing

differs significantly between the nanogels, in correspondence with what we observed earlier. This

can be caused both by differences in nanogel carrier degradation (as we stated in the previous

paragraph based on our data obtained in Huh‐7_LUC cells) or differences in siRNA complexation and

cellular uptake.

FIGURE 6.6 | Time‐dependent EGFP suppression with different dex‐HEMA‐co‐TMAEMA nanogels and lipofectamine™RNAiMAX. The concentration of nanogels was fixed to 50 µg mL‐1 and the amount of siRNA per well equaled 50 pmol (24‐well plate).

With regard to the RNAi kinetics, we also generally observe the same trend as in our luciferase

knockdown experiments. Irrespective of the amount of siRNA that a certain nanogel carrier may

deliver into the cell cytoplasm, the recovery of EGFP expression to steady state level follows

approximately the same course for DS 2.5, DS 8.9 and DS 5.2 dex‐HEMA‐co‐TMAEMA nanogels.

Therefore we again assume that the RNAi kinetics is independent of the nanogel characteristics and

that the recovery phase is most likely governed by cell division. From these data eventually the same

177


hypothesis arises, namely that during a short time‐window following cellular internalization, a small

fraction of the complexed siRNA may be delivered in the cell cytoplasm, but that a genuine

intracellular time‐controlled delivery based upon the characteristics of the dex‐HEMA nanogels is

hampered by the endolysosomal barrier. Other reports in the literature also state that only a minor

fraction of the internalized siRNA is eventually functionally active in the cytosol.[21‐24] Since most

cationic nanoparticles are effectively endocytosed by the target cells, exploiting possibilities to

overcome their resulting vesicular confinement will dramatically increase the biological availability of

siRNA and the RNAi activity.

RNAi gene silencing kinetics in Huh‐7_EGFP cells: influence of PCI

To achieve an intracellular degradation controlled release of siRNA with our cationic dextran

nanogels in order to improve the biological availability of siRNA in the target cell, it is imperative to

induce a transition of siRNA loaded nanogels from the endosomal compartment towards the

cytoplasm of the cell. Chapter 5 showed that photochemical internalization (PCI) was able to

enhance the RNAi effect of both degradable dex‐HEMA as non‐degradable dex‐MA nanogels (Figure

5.9 and 5.10).[25] For this reason, we again turned to PCI to test its effect on the nanogel samples

used in this chapter in order to implement a PCI driven time‐dependent gene silencing. In agreement

with the results in the previous chapter, we observed a significant improvement in EGFP gene

silencing for all nanogel samples used in this chapter, also for the non‐degradable dex‐MA DS 3.4

nanogels (data not shown). Improvement in cellular siRNA delivery with non‐degradable dex‐MA

nanogels is best explained by the fact that charged anionic macromolecules that the particles

encounter in the cell cytoplasm (e.g. proteins, mRNA, negatively charged phospholipids) are able to

displace siRNA molecules from the surface of the nanogels (see also PCI data in chapter 5). Currently,

we cannot be sure whether the observed improvement in the siRNA effect is a result of quantitative

siRNA displacement from the nanogel surface or that a fraction of the internalized siRNA remains

complexed after endosomal escape. In the latter case, this siRNA fraction may still be released in a

degradation controlled fashion from our dex‐HEMA‐co‐TMAEMA nanogels. Unfortunately, up to now

we have insufficient data at our disposal to unambiguously address this question.

As a proof of concept of how photochemical internalization may be applied not only to enhance the

intracellular release of siRNA but also to induce a more sustained gene silencing effect, we followed

an experimental gene silencing protocol as schematically depicted in Figure 6.7. In these

experiments, photochemical internalization was not only induced during transfection as is normally

178


done (Figure 6.7a – PCI t0), but also the influence of a two days delay in the photochemical treatment

was investigated (Figure 6.7b – PCI t2).

PCI t0 on dex‐HEMA‐co‐TMAEMA nanogels (DS 5.2, 25 µg mL‐1) already improved the EGFP expression

on day 3 post‐transfection (i.e. also 3 days following PCI treatment), an effect that shows even more

obvious at the next analysis point (day 6) (Figure 6.8a). Nevertheless, the improvement in the effect

is from this point on gradually lost as can be seen at day 9 and day 12 post‐transfection. As expected,

promoting endosomal escape by PCI, eventually enlarges the fraction of functional siRNA molecules

in the cell cytoplasm, resulting in an improvement of target gene suppression, lasting for at least 6

days under the experimental conditions. Surprisingly, when applied in a transfection with 50 µg mL‐1

as nanogel concentration, the PCI t0 protocol was not able to enhance the eventual gene silencing

outcome (data not shown).

FIGURE 6.7 | Schematic overview of the experimental protocols used with the application of PCI. (a) PCI t0: one day after seeding of the Huh‐7_EGFP cells, photosensitizer (PS) was added and incubation proceeded overnight. At day 0, PS was removed from the cells and the transfection was performed in PS‐free OptiMEM (4 h) after which the cells were illuminated. (b) PCI t2: 48 h after seeding the cells, they were transfected with siRNA loaded nanogels. One day post‐transfection, PS was added and left to incubate with the cells overnight. At day 2 post‐transfection, also a 4 h PS‐free incubation period was respected before illumination. In both protocols, EGFP expression was analyzed day 3 – day 6 – day 9 and day 12 post‐transfection.

The results on the delayed PCI treatment (PCI t2) are displayed in Figure 6.8b‐d. At day 3 post‐

transfection, being 24 h after photochemical treatment, already a significant difference is observed

between the treated and non‐treated cells for the lower nanogel concentrations (10 µg mL‐1, Figure

6.8b and 25 µg mL‐1, Figure 6.8c), but not when 50 µg mL‐1 nanogels were applied on the cells. More

interestingly, for the remainder of the investigated time course, this difference in gene silencing is at

179


least maintained (Figure 6.8b) or even substantially enhanced (Figure 6.8c and 6.8d) with up to 25%

additional gene suppression. These results clearly indicate that 48 hours following transfection, PCI is

still able to release some of the internalized siRNA thereby lengthening the EGFP knockdown, albeit

only a moderate effect when higher nanogel concentrations are used (Figure 6.8d). On the other

hand, PCI did not have a beneficial effect at all on the RNAi kinetics for lipofectamine™RNAiMAX

(data not shown).

FIGURE 6.8 | EGFP expression as a function of time following transfection with DS 5.2 cationic nanogels in the absence (black squares) or presence (white circles) of TPPS2a PS according to the indicated PCI protocol. (a) 25 µg mL‐1 nanogels, PCI t0, (b) 10 µg mL‐1 nanogels, PCI t2, (c) 25 µg mL‐1 nanogels, PCI t2, (d) 50 µg mL‐1 nanogels, PCI t2. Data points are mean values ± SD (n=3).

One has to keep in mind that two days post‐transfection, the internalized siRNA‐nanogels are divided

over a larger cell population as a result of cell proliferation. Moreover, since we applied the same

amount of PS for both the PCI t0 and PCI t2 protocol, this also implies that the number of PS

molecules per cell will be lower, consequently not reaching the full potential that PCI may offer. It

can be assumed that when applying the same method on cells with a slower division rate or even

primary cells, more dramatic PCI effects could emerge. The ‘delayed’ PCI protocol is far from being

optimized and better results could be expected with optimized photochemical dose and investigation

180


of the ideal illumination time‐point. In conclusion, this PCI strategy indicates that delayed application

of PS to transfected cells is still able to liberate a fraction of the siRNA‐nanogels trapped in endocytic

vesicles, which is a potential strategy to prolong the therapeutic effect in the case when endosomal

escape is the effect‐limiting step. Furthermore, the proposed data suggest that the prolonged

residence time in intracellular vesicles (e.g. endolysosomes) is not detrimental at least for some of

the internalized siRNA duplexes. Still this is one of the major concerns when applying a delayed PCI

treatment.[26] PCI on nanogels complexing nuclease‐stabilized siRNA could therefore also be an

interesting alternative.

CONCLUSIONS

In the present chapter, we screened the RNAi gene silencing kinetics on Huh‐7_LUC and Huh‐7_EGFP

cells with dex‐HEMA‐co‐TMAEMA nanogels of varying crosslink density. We first verified that our

cationic nanogels could also induce a sequence‐specific knockdown of the EGFP expression, which

was not yet studied in this thesis. Although the maximal gene silencing effect differed for the nanogel

dispersions investigated in both screening experiments, no marked deviations could be noted in the

recovery to steady state expression levels of the target gene. Photochemical internalization (PCI) was

able to substantially enhance the extent and duration of EGFP knockdown, even when applied 48

hours post‐transfection. This opens up new possibilities to use PCI as a tool to prolong the siRNA

silencing effect when identifying the optimal time‐point for PCI application to maximize the effect of

a single administration.

ACKNOWLEDGMENTS



(FP6) for funding of MediTrans, an integrated project on nanomedicines. A. Høgset is supported by

the Norwegian Research Council. Saartje Steppe is acknowledged for her help with the practical

experiments.

181


182

REFERENCE LIST

1. Fire A., Xu S., Montgomery M.K., Kostas S.A. et al., Potent and specific genetic interference by double‐stranded RNA in Caenorhabditis elegans, Nature 1998, 391: 806‐811.

2. Whitehead K.A., Langer R., and Anderson D.G., Knocking down barriers: advances in siRNA delivery, Nature Reviews Drug Discovery 2009, 8: 129‐138.

3. de Fougerolles A.R., Delivery vehicles for small interfering RNA in vivo, Hum. Gene Ther. 2008, 19: 125‐132. 4. Davis M.E., The First Targeted Delivery of siRNA in Humans via a Self‐Assembling, Cyclodextrin Polymer‐

Based Nanoparticle: From Concept to Clinic, Mol. Pharm. 2009, 6: 659‐668. 5. Behlke M.A., Chemical modification of siRNAs for in vivo use, Oligonucleotides. 2008, 18: 305‐319. 6. Hefner E., Clark K., Whitman C., Behlke M.A. et al., Increased potency and longevity of gene silencing using

validated Dicer substrates, J. Biomol. Tech. 2008, 19: 231‐237. 7. Amarzguioui M. and Rossi J.J., Principles of Dicer substrate (D‐siRNA) design and function, Methods Mol.

Biol. 2008, 442: 3‐10. 8. Chu C.Y. and Rana T.M., Potent RNAi by short RNA triggers, RNA 2008, 14: 1714‐1719. 9. Sano M., Sierant M., Miyagishi M., Nakanishi M. et al., Effect of asymmetric terminal structures of short

RNA duplexes on the RNA interference activity and strand selection, Nucleic Acids Res. 2008, 36: 5812‐5821.

10. Gao K. and Huang L., Nonviral Methods for siRNA Delivery, Mol. Pharm. 2009, 6: 651‐658. 11. Novobrantseva T.I., Akinc A., Borodovsky A., and de F.A., Delivering silence: advancements in developing

siRNA therapeutics, Curr. Opin. Drug Discov. Devel. 2008, 11: 217‐224. 12. Bartlett D.W. and Davis M.E., Insights into the kinetics of siRNA‐mediated gene silencing from live‐cell and

live‐animal bioluminescent imaging, Nucleic Acids Research 2006, 34: 322‐333. 13. Mantei A., Rutz S., Janke M., Kirchhoff D. et al., siRNA stabilization prolongs gene knockdown in primary T

lymphocytes, Eur. J. Immunol. 2008, 38: 2616‐2625. 14. Raemdonck K., Van Thienen T.G., Vandenbroucke R.E., Sanders N.N. et al., Dextran microgels for time‐

controlled delivery of siRNA, Advanced Functional Materials 2008, 18: 993‐1001. 15. Raemdonck K., Vandenbroucke R.E., Demeester J., Sanders N.N. et al., Maintaining the silence: reflections

on long‐term RNAi, Drug Discovery Today 2008, 13: 917‐931. 16. Judge A. and Maclachlan I., Overcoming the innate immune response to small interfering RNA, Human

Gene Therapy 2008, 19: 111‐124. 17. Svoboda P., Off‐targeting and other non‐specific effects of RNAi experiments in mammalian cells, Current

Opinion in Molecular Therapeutics 2007, 9: 248‐257. 18. Grimm D., Streetz K.L., Jopling C.L., Storm T.A. et al., Fatality in mice due to oversaturation of cellular

microRNA/short hairpin RNA pathways, Nature 2006, 441: 537‐541. 19. Dykxhoorn D.M. and Lieberman J., Knocking down disease with siRNAs, Cell 2006, 126: 231‐235. 20. Van Dijk‐Wolthuis W.N., van Steenbergen M.J., Underberg W.J., and Hennink W.E., Degradation kinetics of

methacrylated dextrans in aqueous solution, J. Pharm. Sci. 1997, 86: 413‐417. 21. Detzer A., Overhoff M., Wunsche W., Rompf M. et al., Increased RNAi is related to intracellular release of

siRNA via a covalently attached signal peptide, RNA. 2009, 15: 627‐636. 22. Mescalchin A., Detzer A., Wecke M., Overhoff M. et al., Cellular uptake and intracellular release are major

obstacles to the therapeutic application of siRNA: novel options by phosphorothioate‐stimulated delivery, Expert. Opin. Biol. Ther. 2007, 7: 1531‐1538.

23. Veldhoen S., Laufer S.D., Trampe A., and Restle T., Cellular delivery of small interfering RNA by a non‐covalently attached cell‐penetrating peptide: quantitative analysis of uptake and biological effect, Nucleic Acids Res. 2006, 34: 6561‐6573.

24. Lu J.J., Langer R., and Chen J., A Novel Mechanism Is Involved in Cationic Lipid‐Mediated Functional siRNA Delivery, Mol. Pharm. 2009, 6: 763‐771.

25. Berg K., Folini M., Prasmickaite L., Selbo P.K. et al., Photochemical internalization: a new tool for drug delivery, Curr. Pharm. Biotechnol. 2007, 8: 362‐372.

26. Prasmickaite L., Hogset A., Selbo P.K., Engesaeter B.O. et al., Photochemical disruption of endocytic vesicles before delivery of drugs: a new strategy for cancer therapy, Br. J. Cancer 2002, 86: 652‐657.

liposomal template‐assisted synthesis of dextran based nanogels | CHAPTER 7

CHAPTER 7 | Liposomal

template‐assisted synthesis of dextran based nanogels

Manuscript of this chapter is in preparation:

Koen Raemdonck1, Joseph Demeester1, Stefaan C. De Smedt1 (2009)



183

CHAPTER 7 | liposomal template‐assisted synthesis of dextran based nanogels

184



Chapter 7 provides:

• details on the creation of hybrid lipid‐enveloped (cationic) nanogels (LENs) using liposomes

as nanoreactor

• pilot experiments to evaluate if these nanoparticles are suited for siRNA encapsulation

• A brief overview of other literature reports describing lipid‐coated nanoparticles and a

discussion on the additional value of these nanocomposites for siRNA delivery

Key terms:

Liposomes – nanogels – siRNA – charge neutral – phospholipid gel – encapsulation

INTRODUCTION

The successful intracellular delivery of nucleic acid therapeutics, such as siRNA and pDNA, depends

on the efficient formulation into pharmaceutical nanocarriers that can guide their therapeutic

payload to the target cells. However, as already discussed in chapters 1 and 4, extracellular and

intracellular barriers may prevent efficient accumulation of the drug loaded particles at the site of

action.[1‐3] We provided evidence in chapter 5 and chapter 6 that the presented cationic dextran

nanogels are able to carry high amounts of siRNA therapeutics and transport them across the cellular

membrane. The designated chapters, however, also revealed that the endosomal sequestration of

the siRNA loaded nanogels, significantly limited the full RNAi potential. To achieve our goal of

intracellular controlled release of siRNA, a transition of intact siRNA loaded nanogels from the

endosomal lumen into the cytoplasm needs to occur.

185


Besides polymeric nanoparticles, liposomes have also been widely investigated as pharmaceutical

nanocarriers.[1,4‐6] Liposomes are defined as self‐assembled spherical vesicles consisting of (a) lipid

bilayer(s) that enclose an aqueous core. Like nanogels, liposomes can be tailored to respond to

environmental changes e.g. pH, redox potential and temperature,[6‐10] but they usually lack the ability

to provide a time‐controlled cytosolic release of macromolecular therapeutics. In this chapter, we

aim to combine the advantages of our biodegradable dextran nanogels for cytosolic controlled siRNA

release with the multifunctionality of liposomes into a hybrid nanogel‐lipid core‐shell composite.

Several reports in the literature describe the design of lipid coated nanoparticles.[11‐16] As an example,

Harashima’s group published several papers on multifunctional envelope‐type nanodevices (MENDs),

that consist of a condensed core containing the complexed nucleic acids (siRNA or pDNA) surrounded

by a multifunctional lipid envelope.[17,18] The MEND particles are usually obtained by an optimized

lipid hydration method, where a preformed lipid film is hydrated with an aqueous solution containing

the nucleic acid‐polycation core complexes.[19,20] Alternatively, encapsulation in a lipid bilayer may

also be achieved through fusion of anionic small unilamellar vesicles (SUVs) on the surface of

positively charged nanoparticles.[21] The authors state that during these processes the nucleic acid

core and the lipid bilayers exist as separate structures. It was shown that the packaging of a pDNA or

siRNA complex core into a pH‐responsive fusogenic lipid envelope may lead to efficient endosomal

escape of the nucleic acid complexes due to fusion triggered decoating upon interaction with the

endosomal membrane.[21] Alternatively, Jana et al. prepared polyvinylpyrrolidone nanoparticles that

were subsequently encapsulated in fusogenic Sendai virosomes. The authors indicate that the viral

envelope may trigger fusion with the plasma membrane of (hepatic) cells, thereby injecting the

nanoparticles directly into the cytosol.[22] A comparable approach is presented by Kunisawa et al.

who incorporated poly(vinyl)amine nanoparticles into Sendai viral envelopes.[23]

A potential drawback of the described preparation methods is that they generally consist of several

consecutive steps that need separate optimization (nanoparticle preparation, hydration of lipid coat,

sonication or extrusion,...) which makes the preparation process laborious and difficult to control. It

seems not so straightforward to incorporate the necessary functionalities in a hybrid nanosized

structure by a ‘one‐step’ production process.

However, in the literature one can find some examples of the single‐step preparation of lipid coated

particles, by using liposomes as nanoscaled reactors for particle formation.[24‐33] Hereby, the aqueous

core of the liposomal template is employed as a reaction vessel for the synthesis of nanostructures,

that are thereby automatically equipped with a lipid shell. In this chapter we aimed to evaluate if

hybrid lipid‐enveloped dextran nanogels (LENs) could be prepared for siRNA delivery, by using a

186


liposomal template‐assisted synthesis and we further discuss the potential advantages this delivery

system may provide.


Phospholipids

1,2‐dioleoyl‐sn‐glycero‐3‐phosphocholine (DOPC, Tm=‐20°C, 20 mg mL‐1 in chloroform); 1,2‐dioleoyl‐

sn‐glycero‐3‐phosphoethanolamine (DOPE, Tm=‐16°C, 10 mg mL‐1 in chloroform); 1,2‐dioleoyl‐sn‐

glycero‐3‐phosphoethanolamine‐N‐(lissamine rhodamine‐B sulfonyl) ammonium salt (Rho‐B‐DOPE,

λex/λem = 557 nm/571 nm) and 1,2‐dioleoyl‐sn‐glycero‐3‐phosphoethanolamine‐N‐(7‐nitro‐2‐1,3‐

benzoxadiazol‐4‐yl) (ammonium salt) (NBD‐DOPE, λex/λem = 460 nm/534 nm) were purchased from

Avanti Polar Lipids (Alabama, USA). Lyophilized DOPC and DOPE were obtained from Sigma Aldrich.

Cholesteryl Bodipy FLC12 (cholesteryl 4,4‐difluoro‐5,7‐dimethyl‐4‐bora‐3a,4a‐diaza‐s‐indacene‐3‐

dodecanoate) was from Molecular probes (Eugene Oregon, USA).

siRNA oligonucleotides

Twenty‐one nucleotide siRNA duplexes, targeted against pGL3 luciferase, were purchased from

Eurogentec (Seraing, Belgium). SiRNA batches were diluted in RNase free water and placed at ‐85°C.

For fluorescence experiments, a pGL3 siRNA duplex 5’‐labeled (sense strand) with Alexa Fluor 488

(AF488‐siRNA) was used.

Preparation of lipid‐enveloped dextran nanogels through extrusion

Dextran nanogels were prepared using liposomes as nanosized reactor. Appropriate amounts of

DOPC or DOPC and DOPE (molar ratio 1:1), dissolved in chloroform, were mixed in a glass vial. For

fluorescently labelled liposomes, rho‐B DOPE, NBD‐DOPE or cholesteryl Bodipy were added in 0.1‐0.2

mol% of total phospholipid. To obtain a lipid film, the chloroform was evaporated under a nitrogen

flow, while rotating the vial gently. The lipid film is then placed under a vigorous stream of nitrogen

for at least 1h to remove traces of chloroform and stored at ‐20°C. A dex‐HEMA solution (20% (w/w)

in 20 mM HEPES buffer pH 7.4, containing irgacure 2959 as photoinitiator (Ciba Specialty Chemicals),

is used to hydrate the lipids by vortexing at room temperature (2 min). To obtain lipid coated cationic

dextran nanogels, dimethyl aminoethyl methacrylate (DMAEMA, also see chapter 3) was added to

the dextran solution. The resulting dispersion of large multilamellar vesicles (LMVs) was allowed to

187


stabilize for 5 min and subsequently extruded through a double stacking of 0.4 or 0.2 µm

polycarbonate filters using a hand‐held extruder (Avanti polar lipids). To prevent the polymerization

of the dex‐HEMA which was not encapsulated in the liposome interior, the liposome dispersion was

diluted 1:10 in 20 mM HEPES buffer so that the dex‐HEMA concentration of the solution surrounding

the liposomes became too low to form a hydrogel. Subsequently the sample was UV irradiated at

room temperature for 450 s under N2 atmosphere (365 nm from a Bluepoint 2.1 UV source, Honle

UV Technology) thereby selectively crosslinking the dex‐HEMA inside the lipid bilayer which results in

the formation of lipid‐enveloped dextran nanogels. To obtain naked dextran nanogels, the lipid layer

was solubilized by addition of TX100 (10% w/v) at a molar ratio ≥ 10 relative to total phospholipid

content. In one experiment, the lipid coat was removed by subsequent washing steps with diluted

isopropanol. Briefly, a dispersion of LENs (containing ~1.3 mM total phospholipid) is filtrated on a

Vivaspin 6 ultrafiltration device (300.000 MWCO, polyethersulfone, Vivascience – Sartorius

instruments) by centrifugation (2000 g, 20 min). The remaining volume (~100 µL) is diluted with 1 mL

of distilled water followed by the addition of 700 µL isopropanol (35% final concentration) and again

subjected to a washing step (1000 g, 20 min). This washing step is repeated a second time. After

every centrifugation‐dilution step with distilled water, an aliquot is taken for zeta‐potential

measurement and phosphate analysis according to Rouser et al. after destruction of the lipids using

perchloric acid.[31,34]

Preparation of lipid coated dextran nanogels through ultrasonication

Lyophilized lipids (total amount 150 mg), i.e. DOPC or DOPC:DOPE (1:1), are weighed in a non‐stick

RNase free 2 mL eppendorf (Ambion, Austin, TX) and 500 µL of chloroform is added to dissolve the

lipids. Subsequently, the chloroform is removed under nitrogen flow followed by additional vacuum

evaporation (Büchi Rotavapor R‐200). The dried lipid constituents were hydrated with a dex‐MA

solution (25 wt% in HEPES buffer) with a CapMix™ (3*30s) and left to hydrate during 30 min (4°C). To

further size the formed lipid vesicles, the dispersion was sonicated using a probe sonicator (Branson

digital sonifier) in subsequent steps of maximum 25 s. After every cycle, the resulting vesicular

phospholipid gel (VPG) was briefly placed on ice before starting the next sonication step. To evaluate

the size of the liposomes in the VPG, ~1 mg of VPG was dispersed in 2 mL HEPES buffer (CapMix™

during 15 s) and analyzed with DLS at 25°C.

188


siRNA loading during preparation of lipid‐enveloped dextran nanogels

The loading of dex‐HEMA nanogels with siRNA was obtained through hydration of the DOPC:DOPE

lipid film with a siRNA containing dex‐HEMA solution (passive loading). To estimate the siRNA uptake

inside the liposomes formed by extrusion (final lipid concentration ~30 mM) in a preliminary

experiment, the non‐encapsulated siRNA was separated from the liposomes by gel permeation

chromatography (GPC) using a Sephadex CL‐4B column (Sigma), equilibrated with Tris/EDTA (10

mM/1 mM) buffer at pH 7.4. The elution of the liposomes was detected by turbidity measurements

with a Pharmacia LKB Biochrom UV/VIS spectrophotometer (λ = 633 nm). The siRNA concentration in

each elution fraction was quantified (after the addition of TX100) using the Quant‐it™ Ribogreen®

RNA Reagent (Molecular Probes, Eugene, Oregon, USA). To quantify the encapsulation efficiency of

siRNA in liposomes formed by ultrasonication, DOPC:DOPE lipids were hydrated with a 25% dex‐MA

solution in HEPES buffer, supplemented with siRNA (final lipid concentration 500 mg mL‐1, final siRNA

concentration 40 µM). The encapsulation efficiency was quantified by measuring the fluorescence of

Quant‐it™ Ribogreen® RNA Reagent before and after TX100 addition (0.25%) to the lipid dispersion.

In the absence of detergent, the signal comes from accessible (i.e. unencapsulated) siRNA only, while

the signal after detergent addition corresponds with total siRNA.[35] Calibration curves of siRNA were

first constructed with and without TX100 (R2= 0.998). Ribogreen fluorescence (λex= 500 nm, λem=525

nm) was measured using a Wallac Victor2 fluorescence plate reader (Perkin Elmer, Waltham, MA).

Dynamic Light Scattering (DLS) and zeta‐potential

The particle size of the nanogels was measured by dynamic light scattering using a Malvern Autosizer

4700, consisting of a He‐Ne laser (633 nm, 25 mW output power). Polystyrene nanospheres (220 ± 6

nm, Duke Scientific Corporation) were used as calibration standard. The mean hydrodynamic

diameter of the particles (dh) was obtained from the intensity fluctuations of the scattered light using

the software package provided by Malvern based on the Stokes‐Einstein equation: D = kbT.(3πηdh)‐1

where D is the diffusion coefficient, kb is de boltzmann constant, T is the temperature of the water

bath (298 K) and η is the solvent viscosity. Scattering was measured at a fixed angle of 90°. HEPES

buffer, used to dilute the nanoparticle dispersions, was filtered through 0.22 µm membrane before

use. Some of the size measurements were repeated and/or performed on the Malvern Zetasizer

Nano ZS, as described in previous chapters.

Zeta potential measurements in this chapter were performed on a Zetasizer 2000, Malvern,

Worcestershire, UK.

189


Scanning electron microscopy

Scanning electron microscopy on naked nanogels was carried out using a Zeiss DSM 40 instrument

operating at an accelerating voltage of 5 kV. For sample preparation, a drop of the dispersion of

naked nanogels (obtained through washing of the lipid coated nanogels with isopropanol) was dried

under low vacuum on a glass slide and subsequently sputtered with gold/palladium.

Fluorescence Fluctuation Spectroscopy (FFS)

A dual‐colour fluorescence fluctuation spectroscopy instrument (dual‐colour FFS) was used as

described previously.[36] FFS monitors fluorescence intensity fluctuations in the excitation volume of a

confocal microscope. In dual‐colour FFS measurements, two spectrally different fluorophores

(respectively AF488 and lissamine‐rhodamine B in this study) are used. The fluorescence of the two

fluorophores is registered by two separate detectors monitoring the same excitation volume. In this

study the 488 nm and 561 nm laser lines of a Krypton‐Argon ion laser (BioRad, Cheshire, UK) were

used to excite the samples and the fluorescence fluctuations were recorded during a 50 s time

interval. For a profound theoretical background on FFS we would like to refer to the indicated

references.[37‐39]

FFS experiments were carried out on dex‐HEMA‐co‐DMAEMA containing DOPC:DOPE liposomes

respectively before and after UV irradiation. Therefore 50 µL of the liposomal dispersions (50 nM

phospholipids) was applied in the wells of a glass‐bottomed 96‐well plate (Greiner Bio‐one, certified

DNase/RNase free). The focal volume was placed 50 µm above the bottom of the wells. The lipid

layer was fluorescently labelled by addition of 0.1 mol% of rhodamine B‐DOPE to the lipid mixture.

TX100 was added to solubilize the lipid layer (molar ratio TX100/phospholipids = 10). After

solubilization, AF488‐siRNA (final concentration = 50 nM) was mixed with the samples in the wells

and fluorescence fluctuations were again recorded.

Cellular uptake of lipid coated cationic nanogels in Huh‐7 cells

The wild‐type human hepatoma cell line Huh‐7_wt was cultured (37°C, 5% CO2) in DMEM:F12,

supplemented with 2 mM glutamine, 10 % heat‐inactivated FBS and 100 U mL‐1 P/S. Huh‐7 cells were

seeded in sterile glass‐bottomed culture dishes (MatTek Corporation, Ashland, MA) at a density of 3 x

104 cells per cm2 and allowed to attach overnight. DOPC:DOPE (1:1) coated cationic nanogels (20 %

w/w dex‐HEMA DS 5.7, molar ratio HEMA/DMAEMA = 0.25, 25 nmol lipids) containing 0.2 mol% of

cholesteryl Bodipy were incubated with the cells (37°C, 5% CO2) in OptiMEM during 4 h.

Subsequently, cells were washed with PBS and preheated culture medium was added. After another

190


20 h of incubation, the intracellular distribution of the lipid coated nanogels was visualized with a

Nikon EZC1‐si confocal scanning module installed on a motorized Nikon TE2000‐E inverted

microscope (Nikon Benelux, Brussels, Belgium). For lysosomal staining, cells were incubated for 15

min with 1:104 diluted Lysotracker Red® (Molecular Probes) prior to image recording. Nuclei were

stained through 10 min incubation with a 1:103 dilution of DraQ5™ (Alexis Biochemicals,

Switzerland). Extracellular fluorescence was quenched by short incubation with 0.4% trypan blue

(Sigma). Images were captured with a Nikon Plan Apochromat 60x oil immersion objective (numerical

aperture 1.4) using a 488 nm, 561 nm and 639 nm laser line for the excitation of cholesteryl Bodipy,

Lysotracker Red® and DraQ5™ respectively.

Quantification of cellular uptake of DOPC:DOPE LENs was analyzed using flow cytometry. Briefly, 6 x

104 cells were seeded in 24‐well plates and 48 h after seeding, they were incubated for 4 h with

increasing concentrations of fluorescently labelled LENs (DOPC:DOPE containing 0.2 mol% NBD‐

DOPE) in complete cell culture medium (0.5 mL final volume). Following incubation, the cells were

washed with PBS, quenched with 0.4% trypan blue (8 min, 37°C) and prepared for flow cytometry as

already explained in previous chapters.


Formation of biodegradable nanogels inside a liposome reactor

In this manuscript we aimed to design dextran nanogels using liposomes as nanoscaled reactor for

the selective photopolymerization of dextran hydroxyethyl methacrylate (dex‐HEMA), as described

earlier by Van Thienen et al..[32] Figure 7.1 schematically depicts the preparation procedure. In a first

step, a lipid film is hydrated with a buffered dex‐HEMA solution, that is supplemented with irgacure

2959 as a photoinitiator. In this way, polydisperse multilamellar liposomal vesicles are formed with

dex‐HEMA encapsulated in their aqueous core. To obtain a more homogeneous dispersion of

(predominantly) unilamellar vesicles, the sample is subsequently extruded through a polycarbonate

filter with a defined pore size. A final step is the UV induced photopolymerization of dex‐HEMA inside

the liposomal template, to obtain dextran nanogels surrounded by a lipid bilayer. In order to prevent

the polymerization of dex‐HEMA in between the liposomal vesicles, the liposome dispersion is

diluted at least ten‐fold. In this way, the dex‐HEMA crosslinking will be restricted to the liposome

inner compartment as the concentration of unencapsulated dex‐HEMA will be below the critical

threshold that is needed for hydrogel formation. Alternatively, the external dex‐HEMA may also be

removed by size exclusion chromatography or dialysis, which are however more time‐consuming and

191


labour intensive methods. Schillemans et al. reported on the use of ascorbic acid to inhibit the free

radical polymerization process in between the liposomal vesicles to ensure selective crosslinking in

the liposome interior.[31]

FIGURE 7.1 | Schematic overview of the preparation protocol to obtain dextran nanogels surrounded by a lipid coat. Hydration of a thin lipid film with a dex‐HEMA solution leads to a dispersion of large and polydisperse multilamellar vesicles (LMVs). Extrusion of the LMV dispersion through a polycarbonate membrane is used to ‘size’ the obtained multilamellar vesicles into monodisperse large unilamellar vesicles or LUVs. These liposomes carry dex‐HEMA chains in their aqueous core that can be polymerized during UV irradiation to form a dex‐HEMA gel. Redrawn from reference [24].

Figure 7.2a shows dynamic light scattering (DLS) data on a dispersion of DOPC:DOPE (1:1) liposomes

filled with dex‐HEMA respectively before and after UV treatment. In both cases vesicles of ~200 nm

were detected. To solubilise the lipid bilayer, TX100 was added to the dispersions. The removal of the

lipid coat was evidenced by the lowering of the scattering intensity (the removal of the lipid bilayer

lowers the difference in refractive index between the nanoparticles and the aqueous dispersion

buffer)[32] and the appearance of detergent‐phospholipid mixed micelles (~15‐20 nm in size; see

Figure 7.2c). Upon addition of TX100 to the liposome dispersion which was not exposed to UV light

only micelles remained (monomodal distribution, data not shown). In contrast, both micelles and

uncoated nanogels were observed after addition of TX100 to the UV irradiated liposome dispersions

(Figure 7.2c, bimodal distribution). Figure 7.2d displays a scanning electron micrograph of ‘naked’

nanogels, obtained after removal of the lipid coat. The DLS and electron microscopy data in Figure

7.2 clearly indicate that UV irradiation of the dex‐HEMA filled liposomes results in nanogel particles

formed within the liposomal reactor.

We also monitored the size of these lipid coated dex‐HEMA nanogels as a function of time at 37°C by

DLS. In agreement with the observations by Van Thienen et al.,[32] even after > 2 weeks of incubation,

still vesicles of about 200 nm in size were detected (Figure 7.2a ‐ UV degraded, white bar).

192


FIGURE 7.2 | (a) The white bars represent the hydrodynamic size of DOPC:DOPE (1:1) liposomes, hydrated with dex‐HEMA DS 5.7, respectively before and after UV irradiation (for nanogel formation) and after the degradation of the nanogels. The black bars represent the DLS data on the same samples after lipid solubilisation with TX100, (b) DLS profile of the lipid coated sample after UV irradiation, (c) DLS profile of the lipid coated sample after UV irradiation/TX100 treatment, (d) scanning electron micrograph of naked dex‐HEMA nanogels (DS 21). The scale bar represents 500 nm.

However, upon removal of the lipid bilayer with detergent, only mixed micelles and particle

remnants remained detectable by DLS (Figure 7.2a – UV degraded, black bar), indicating that the dex‐

HEMA nanogels inside the liposomes are indeed degraded. Although the hydrolysis of the dex‐HEMA

network is usually accompanied by a volume increase and leads to a mixture of dextran and

polyHEMA, which could increase the osmotic pressure inside the lipid vesicles,[32,40] the internal

pressure build‐up seems to be insufficient to rupture the lipid bilayer of the vesicles.

Preparation of cationic dextran nanogels using liposomal template

The introduction of positively charged groups inside the dex‐HEMA network could be advantageous

for the electrostatic entrapment of oppositely charged therapeutics, both for low molecular weight

compounds as for proteins and nucleic acids, as already described in chapter 4. Additionally, the

193


copolymerization of cationic monomers with dex‐HEMA can alter the degradation rate of the dex‐

HEMA hydrogel network and may provide an extra means to tailor the disintegration of the dextran

nanogels and eventually the release of the encapsulated material. Third, the experiments described

below provide additional evidence of the formation of crosslinked nanogels in the liposome reactor.

Lipid coated positively charged nanogels were prepared by hydrating a DOPC:DOPE (1:1) lipid film

(supplemented with 0.1 mol% of Rho‐B DOPE – red fluorescence) with a DMAEMA containing dex‐

HEMA/irgacure2959 solution. After dilution of the liposome suspension, the sample was UV

irradiated as described in the experimental section. In this way, we anticipate the selective

copolymerization of DMAEMA with dex‐HEMA inside the lipid coat. We applied zeta‐potential

measurements (Figure 7.3) and dual‐colour fluorescence fluctuation spectroscopy (dual‐colour FFS,

Figure 7.4) to verify the presence of cationic nanogels inside the neutral lipid coat.

Evaluation of the zeta‐potential of DOPC:DOPE (1:1) coated dex‐HEMA‐co‐DMAEMA nanogels reveals

nanoparticles with a near‐neutral surface charge which indicates that no polymerization of DMAEMA

and dex‐HEMA has occurred on the external surface of the lipid vesicles. Subsequently, we removed

the lipid coat by sequential washing steps with isopropanol. The removal of the lipid coat, as

evidenced by quantification of the lipid phosphate groups, coincides with the appearance of a

positive zeta‐potential that is most likely due to the detection of naked dex‐HEMA‐co‐DMAEMA

nanogels (Figure 7.3).

FIGURE 7.3 | Removal of the DOPC:DOPE (1:1) lipid coat through isopropanol washing from LENs with added DMAEMA, reveal the emergence of positively charged particles. Gray bars represent the results of the phosphate analysis for phospholipid quantification while line graph depicts the zeta‐potential measurements. Graph inset shows the calibration curve for phosphate quantification (R2=0.9998).

194


FIGURE 7.4 | (a) FFS results on DOPC:DOPE (1:1) coated dex‐HEMA‐co‐DMAEMA nanogels, before UV irradiation (left graph) and after exposure to UV light (right graph). The lipid envelope is fluorescently labelled through the addition of 0.1 mol% rho‐B‐DOPE (red fluorescence), the siRNA duplexes are labelled with AF488 and therefore are green fluorescent. (b) Schematic representation of the particles/molecules that are likely measured during the fluorescence fluctuation spectroscopy experiment.

Figure 7.4 shows the fluorescence fluctuations obtained on red fluorescently labelled liposomes filled

with dex‐HEMA/DMAEMA, respectively before (Figure 7.4a – left graph) and after UV polymerization

(Figure 7.4a – right graph). Upon excitation with 561 nm laser light, highly intense red fluorescence

peaks appeared in both samples, originating from the migration of bright fluorescent nanoparticles

(in this case liposomes) through the excitation volume of the confocal microscope.[38,41] The lipid

bilayer is strongly fluorescent due to the incorporation of multiple rhodamine B‐DOPE lipid

molecules. When TX100 was added to the samples, fluorescence peaks were no longer detected

which can be explained by the solubilization of the lipid bilayer resulting in mixed micelles. Mixed

micelles are relatively small when compared to the liposomal vesicles and most likely contain only a

small amount of rhodamine‐B DOPE molecules. Therefore, diffusion of these mixed micelles through

the excitation volume does not give rise to high fluorescence peaks. In a next step, green fluorescent

siRNA was added to the TX100 treated samples. In Figure 7.4a (left graph), “normal” fluorescence

fluctuations (i.e. without high fluorescence peaks) were measured with an average of ~360 kHz.

Interestingly, when the sample was exposed to UV light, highly intense green fluorescence peaks

appeared whith a baseline fluorescence signal 45 kHz (Figure 7.4a, right graph). These observations

195


provide additional evidence for the formation of cationic dex‐HEMA‐co‐DMAEMA nanogels inside the

neutral lipid coat. Indeed, when the liposomes are not exposed to UV light, no polymerization of the

dex‐HEMA/DMAEMA can occur. After removal of the lipid coating with TX100, a solution of

unpolymerized dex‐HEMA/DMAEMA together with neutral mixed micelles remains, that cannot

complex the added fluorescent siRNA (Figure 7.4b). As a result “normal” fluorescence fluctuations

are detected, corresponding to freely diffusing AF488‐siRNA, and high fluorescence peaks are absent.

On the other hand, UV irradiation results in the formation of DMAEMA containing dextran nanogels

within the lipid bilayer. After addition of TX100, naked cationic nanogels emerge that are able to

complex the added fluorescent siRNA (Figure 7.4b). The diffusion of nanogels, loaded with AF488‐

siRNA, explains (a) the emergence of green fluorescence peaks and (b) the lowering of the baseline

fluorescence signal.

Cellular internalization of neutral lipid‐enveloped dextran nanogels

The cellular internalisation of nanoparticles often depends on the interaction of positively charged

groups with the oppositely charged heparan sulphate proteoglycans spanning the plasma membrane

of eukaryotic cells.[42] Since the lipid nanoreactor that is envisioned in this chapter, in accordance

with our experimental needs, does not contain charged components, we aimed to verify if this

characteristic significantly impedes cellular uptake of our LENs. Flow cytometry on DOPC:DOPE (1:1)

coated dex‐HEMA DS 5.2 nanogels (fluorescently tagged by 0.2 mol% NBD‐DOPE), still indicated

efficient cellular binding and internalization in Huh‐7_wt cells from a ratio of ~10 nmol total lipid per

6 x 104 seeded cells when a 4 h incubation time with the cells was respected (Figure 7.5a). The LENs

were characterized by DLS and zeta‐potential (198.6 ± 1.7 nm; PDI 0.093 ± 0.022; ‐6.1 ± 0.4 mV) .

During this incubation period, even at the highest amount of lipid tested (1 µmol), no signs of

cytotoxicity or morphological aberrations were observed. The uptake of fluorescently tagged LENs

was also visualized with confocal microscopy (Figure 7.5b).

In this section we chose to use DOPC:DOPE (1:1) neutral liposomes for several reasons: (1) neutral

liposomes will not interfere with the incorporation of charged methacrylate monomers in the dex‐

HEMA nanogels, (2) neutral liposomes generally have low cytotoxicity, especially when compared

with cationic lipid formulations,[4] and (3) DOPC, carrying two mono‐unsaturated acyl chains, forms

stable vesicles at ambient and physiological temperature and the ability to fuse (e.g. with endosomal

vesicular membranes) can be enhanced by incorporation of DOPE, which enhances the

hydrophobicity of the lipid bilayer.[43,44]

196


FIGURE 7.5 | (a) 4 h incubation of Huh‐7_wt cells with DOPC:DOPE (1:1) lipid coated dex‐HEMA nanogels (DS 5.2), labelled with 0.2 mol% of NBD‐DOPE, in complete cell culture medium (supplemented with 10% FBS) leads to efficient cellular uptake > 10 nmol total lipids, (b) cellular uptake of cholesteryl bodipy labelled (0.2 mol%) DOPC:DOPE (1:1) LENs (DS 5.7 + DMAEMA) after 4 hours incubation, (c) labelling of acidified vesicles with Lysotracker Red®, (d) merge of green and red fluorescence, (e) transmission image. Scale bar = 10 µm.

Efficient cellular internalization may not be a surprise according to the data on DOPE containing

liposomes provided by Simoes et al. since the authors discuss that the presence of the lowly

hydrated DOPE headgroup may increase the affinity of the liposomes for the plasma membrane. In

the same paper however, it is also stated that the cellular internalization and intracellular drug

delivery of negatively charged liposomes (such as the pH‐sensitive DOPE:CHEMS formulation) is

generally more efficient than with neutral liposomes,[44] an observation that corroborates with what

is found in other studies.[45,46] It could therefore also be interesting to include negatively charged

fusogenic LENs in our future studies.

The literature contains several reports on neutral DOPC‐siRNA liposomal nanoparticles as anticancer

delivery agents that already have been deployed in the treatment of lymphoma, melanoma, ovarian

and colorectal carcinoma.[47‐53] In various animal models a fraction of the DOPC‐siRNA nanoparticles

was detected inside the tumour cells following i.p. and i.v. administration, indicating cellular uptake

of the neutral liposomes by the target cells. Moreover, these reports also demonstrate effective

target gene suppression with the described neutral liposomal formulation, which indicates that

active siRNA is delivered to the cell cytoplasm. Unfortunately, the latter manuscripts do not provide

additional information on the cellular internalization pathway and the mechanism of endosomal

escape. Nonetheless, as neutral lipid formulations are generally well tolerated, these data are

promising in the context of achieving safe in vivo delivery of siRNA.[54]

197


Loading of lipid coated dextran nanogels with siRNA

One of the critical parameters that needs to be determined in the development of a drug delivery

system is the drug encapsulation efficiency (EE, i.e. the fraction of drug encapsulated by the delivery

carrier in relation to the total amount of drug added). Direct entrapment of siRNA in (neutral) lipid

coated dex‐HEMA nanogels was first tested through hydration of a DOPC:DOPE lipid film

(supplemented with 0.1 mol% of Rho‐B DOPE) with a dex‐HEMA solution containing AF488 labelled

siRNA. Fluorescence microscopy on the resulting LMVs revealed clear colocalization of green and red

fluorescence, suggesting that indeed a fraction of the siRNA was encapsulated in the aqueous core of

the lipid vesicles (data not shown).

FIGURE 7.6 | Gel permeation chromatography performed on siRNA loaded DOPC:DOPE (1:1) liposomes. Every elution fraction was analyzed by turbidimetry at 633 nm to detect elution of the liposomes (line plot). Also the siRNA concentration in every fraction was measured with ribogreen assay reagent (Molecular Probes) after TX100 addition to solubilise the lipids (gray bars).

However, the efficiency of this direct inclusion protocol was estimated by gel permeation

chromatography (Figure 7.6) and it seems that only a small fraction of the added siRNA (<5%) is

entrapped in the lipid nanoreactors (30 mM final lipid concentration was used in this experiment).

The low encapsulation efficiency of siRNA could be somewhat expected as the encapsulated aqueous

space in the liposome interior (mainly governed by the size of the liposomes, their lamellarity and the

lipid concentration used) will almost be negligible relative to the external (non‐encapsulated)

volume.[55] Similarly, low encapsulation efficiencies were also observed upon passive loading of e.g.

calcein,[44] proteins [56] and antisense oligonucleotides in neutral liposomes. In contrast, the literature

198


also describes protocols for the preparation of neutral and/or negatively charged liposomal

formulations to achieve higher loading efficiency for pDNA [4,43], DNA oligonucleotides [57] and

siRNA.[50] The DOPC lipid nanoparticles from A.K. Sood’s group that were described earlier in this

chapter, have an estimated ~65% loading efficiency (based on centrifugational ultrafiltration to

separate encapsulated from non‐encapsulated material), but unfortunately the physicochemical

characteristics of the lipid formulation remain uncertain.[50] A common feature of the latter reports is

the use of (water‐miscible) organic solvents during the process of liposome formation, which can

cause a precipitation of the added dextran during LENs formation. For this reason the proposed

methods are not the first choice when aiming to optimize siRNA loading inside our lipid coated

dextran nanogels.

The theoretical encapsulation efficiency of hydrophilic compounds is directly proportional to the

total lipid concentration.[55] Increasing the amount of lipids used during liposome preparation

augments the total internal liposome volume and thus also the potential entrapment of hydrophilic

(macro)molecules. At highly concentrated lipid mixtures, exceeding 300 mM of total phospholipid,

viscous vesicle dispersions are formed defined as a vesicular phospholipid gel or VPG. In this capacity,

the liposomal vesicles are so closely packed that the aqueous space in between the vesicles is

minimal and intervesicle interactions give rise to a semi‐solid consistency. Subsequent dilution of a

VPG eventually leads to a conventional liposomal dispersion. Due to the high viscosity of the VPG

sample, conventional liposomal preparation methods (e.g. the thin film hydration method follow by

extrusion) are not the best choice. Usually, high pressure homogenization (HPH) is considered to be

the gold standard for VPG production.[58] Recently, Massing and co‐workers introduced a different

homogenization technique called differential asymmetric centrifugation or DAC, that can be applied

to homogenize concentrated phospholipid mixtures [59] and this technique led to high encapsulation

of siRNA up to 70%.[60]

In analogy with these reports, we aimed to prepare lipid coated dextran nanogels at higher lipid

concentrations than used earlier, in order to achieve higher entrapment efficiency of our siRNA

duplexes. In a first experiment, it needed to be evaluated if such high lipid concentrations could be

hydrated and transformed into vesicles of monodisperse size. Lyophilized DOPC was hydrated with a

dex‐MA 20 wt% solution to reach a final lipid concentration of 500 mg mL‐1 (equal to 636 mM). To

homogenize the lipid dispersion at this high concentration, we tested probe sonication (for more

details see experimental section) and the results are summarized in Figure 7.7. Merely mixing the

lipids with a dex‐HEMA solution produces LMVs (average size > 700 nm) with high polydispersity.

Clearly, probe sonication provides enough shear force to ‘size’ the liposomal vesicles in the obtained

vesicular phospholipid gel down to ~110 nm with low polydispersity. The probe sonication has the

199


advantage that smaller dex‐HEMA LENs can be prepared compared with the extrusion method, since

the dex‐HEMA solutions (20‐25 wt%) appear too viscous to be able to easily pass through 100 nm

polycarbonate filters.

FIGURE 7.7 | Hydrodynamic diameter of dex‐MA DS 3.4 filled DOPC liposomes after sonication of a 500 mg mL‐1 DOPC vesicular gel for the indicated time. To prepare a liposomal dispersion from the VPG after every time point, an aliquot of the VPG lipid paste (~1 mg) was diluted with 2 mL HEPES buffer (20 mM, pH 7.4) and homogenized by using a CapMix™ for 15 s. DLS measurements (n=4) were performed after equilibration of the samples at 25°C.

The siRNA encapsulation efficiency in DOPC:DOPE (1:1) dispersions, hydrated with a HEPES buffered

dex‐HEMA solution (pH 7.4, 20 mM) containing 40 µM of siRNA, was ~30% in our hands (liposome

size of 138.6 ± 0.4 nm; PDI 0.108 ± 0.003; 500 mg mL‐1 total lipid). The liposomal encapsulation

efficiency needs to be optimized thoroughly as a function of lipid concentration, lipid composition,

size and lamellarity of the liposomal vesicles, etc..[4,57] Parameter design towards EE optimization is

not the primary focus of this chapter and will need to be conducted in future experiments with these

lipid‐dextran nanocomposites.

What can lipid coated nanogels mean for siRNA delivery?

The proposed lipid coated nanogels offer many advantages, both with regard to their preparation as

in the context of siRNA delivery. As discussed in the introduction, coating nanoparticles with a lipid

shell can be achieved via different mechanisms. Using liposomal vesicles as nanoreactors however, it

is possible to devise core‐shell LENs in a single‐step process with excellent control over particle size

and without having to rely on electrostatic interactions to create a lipid corona. Additionally, from a

200


combinatorial point of view, merging a liposomal and nanogel carrier into one single nanocomposite,

combines the controlled release properties of the nanogel matrix with the favourable membrane

properties of phospholipid vesicles.[61]

The multifunctionality of the lipid coat is of significant importance in overcoming the most prominent

barriers in siRNA delivery. The choice of lipids is imperative in tailoring the surface chemistry of the

LENs, that will eventually influence to a large extent their behaviour in the extra‐ and intracellular

environment. The mechanisms with which our lipid coated nanogels could improve siRNA delivery

are therefore in agreement with what literature reports on how liposomes can be functionalized for

improved in vivo performance.[1,5,8,62] First, the preparation of stealth LENs by the incorporation of

PEGylated lipids could be used to prevent (1) aggregation by interaction with serum proteins and (2)

massive non‐specific phagocytosis, mainly by Kupffer cells in the liver and/or macrophages in the

spleen. As a result, the increased residence time in the blood circulation should favour accumulation

in leaky tumours through the Enhanced Permeation and Retention (EPR) effect or could augment the

uptake in other tissues and organs. Nonetheless, this is mainly a prerequisite for particles carrying

charged lipid bilayers and is of less importance for neutral liposomes or the LENs we envision. Once

the LENs gain access to the target cells, in a next step they should cross the cell membrane and

deliver intact siRNA into the cytoplasm for RNAi activation. PEGylation is usually accompanied by

coupling of a targeting moiety to the PEG distal end to improve cell surface receptor binding and

subsequent internalization. Following cell uptake, lipid‐based nanoparticles generally end up being

trapped in (acidified) endosomal vesicles. The preparation of LENs with pH‐sensitive lipid bilayers

that become fusogenic at endosomal pH (pH 5‐6.5) could facilitate the intracytoplasmic injection of

the encapsulated siRNA loaded nanogels.[5,8,44] From these examples it should be clear that the

predominant motivation to equip nanogels with a lipid‐shell is to overcome the major extracellular

barriers and allow disposition of biodegradable siRNA loaded nanogels in the target cell cytoplasm. It

would be interesting to assess the effect of lipid composition on siRNA loading, intracellular siRNA

delivery and the eventual RNAi outcome (both maximal effect as gene silencing duration) in future

experiments.

201


REFERENCE LIST


2. Raemdonck K., Demeester J., and De Smedt S., Advanced nanogel engineering for drug delivery, Soft Matter 2009, 5: 707‐715.

3. Raemdonck K., Vandenbroucke R.E., Demeester J., Sanders N.N. et al., Maintaining the silence: reflections on long‐term RNAi, Drug Discovery Today 2008, 13: 917‐931.

4. Li W. and Szoka F.C., Jr., Lipid‐based nanoparticles for nucleic acid delivery, Pharm. Res. 2007, 24: 438‐449. 5. Torchilin V.P., Recent advances with liposomes as pharmaceutical carriers, Nature Reviews Drug Discovery

2005, 4: 145‐160. 6. Ganta S., Devalapally H., Shahiwala A., and Amiji M., A review of stimuli‐responsive nanocarriers for drug

and gene delivery, Journal of Controlled Release 2008, 126: 187‐204. 7. Ong W., Yang Y.M., Cruciano A.C., and McCarley R.L., Redox‐Triggered Contents Release from Liposomes,

Journal of the American Chemical Society 2008, 130: 14739‐14744. 8. Simoes S., Moreira J.N., Fonseca C., Duzgunes N. et al., On the formulation of pH‐sensitive liposomes with

long circulation times, Adv. Drug Deliv. Rev. 2004, 56: 947‐965. 9. Kullberg M., Mann K., and Owens J.L., A two‐component drug delivery system using Her‐2‐targeting

thermosensitive liposomes, J. Drug Target 2009, 17: 98‐107. 10. Auguste D.T., Furman K., Wong A., Fuller J. et al., Triggered release of siRNA from poly(ethylene glycol)‐

protected, pH‐dependent liposomes, J. Control Release 2008, 130: 266‐274. 11. Sengupta S., Eavarone D., Capila I., Zhao G.L. et al., Temporal targeting of tumour cells and neovasculature

with a nanoscale delivery system, Nature 2005, 436: 568‐572. 12. Messerschmidt S.K., Musyanovych A., Altvater M., Scheurich P. et al., Targeted lipid‐coated nanoparticles:

Delivery of tumor necrosis factor‐functionalized particles to tumor cells, J. Control Release 2009, doi: 10.1016/j.jconrel.2009.03.010.

13. Li S.D. and Huang L., Surface‐modified LPD nanoparticles for tumor targeting, Oligonucleotide Therapeutics 2006, 1082: 1‐8.

14. Li S.D., Chen Y.C., Hackett M.J., and Huang L., Tumor‐targeted delivery of siRNA by self‐assembled nanoparticles, Molecular Therapy 2008, 16: 163‐169.

15. De Miguel I, Imbertie L., Rieumajou V., Major M. et al., Proofs of the structure of lipid coated nanoparticles (SMBV) used as drug carriers, Pharm. Res. 2000, 17: 817‐824.

16. Mastrobattista E., Kapel R.H., Eggenhuisen M.H., Roholl P.J. et al., Lipid‐coated polyplexes for targeted gene delivery to ovarian carcinoma cells, Cancer Gene Ther. 2001, 8: 405‐413.

17. Hatakeyama H., Akita H., Kogure K., and Harashima H., A Novel Nonviral Gene Delivery System: Multifunctional Envelope‐Type Nano Device, Adv. Biochem. Eng Biotechnol. 2009, doi: 10.1007/10_2008_40.

18. Sakurai Y., Hatakeyama H., Akita H., Oishi M. et al., Efficient short interference RNA delivery to tumor cells using a combination of octaarginine, GALA and tumor‐specific, cleavable polyethylene glycol system, Biol. Pharm. Bull. 2009, 32: 928‐932.

19. Kogure K., Akita H., and Harashima H., Multifunctional envelope‐type nano device for non‐viral gene delivery: Concept and application of Programmed Packaging, Journal of Controlled Release 2007, 122: 246‐251.

20. El‐Sayed A., Khalil I.A., Kogure K., Futaki S. et al., Octaarginine‐ and octalysine‐modified nanoparticles have different modes of endosomal escape, J. Biol. Chem. 2008, 283: 23450‐23461.

21. Akita H., Kudo A., Minoura A., Yamaguti M. et al., Multi‐layered nanoparticles for penetrating the endosome and nuclear membrane via a step‐wise membrane fusion process, Biomaterials 2009, 30: 2940‐2949.

22. Jana S.S., Bharali D.J., Mani P., Maitra A. et al., Targeted cytosolic delivery of hydrogel nanoparticles into HepG2 cells through engineered Sendai viral envelopes, FEBS Letters 2002, 515: 184‐188.

23. Kunisawa J., Masuda T., Katayama K., Yoshikawa T. et al., Fusogenic liposome delivers encapsulated nanoparticles for cytosolic controlled gene release, Journal of Controlled Release 2005, 105: 344‐353.

24. An S.Y., Bui M.P., Nam Y.J., Han K.N. et al., Preparation of monodisperse and size‐controlled poly(ethylene glycol) hydrogel nanoparticles using liposome templates, J. Colloid Interface Sci. 2009, 331: 98‐103.

25. Hong J.S., Vreeland W.N., Lacerda S.H., Locascio L.E. et al., Liposome‐templated supramolecular assembly of responsive alginate nanogels, Langmuir 2008, 24: 4092‐4096.

202


26. Kazakov S., Kaholek M., Teraoka I., and Levon K., UV‐induced gelation on nanometer scale using liposome reactor, Macromolecules 2002, 35: 1911‐1920.

27. Kazakov S., Kaholek M., Kudasheva D., Teraoka I. et al., Poly(N‐isopropylacrylamide‐co‐1‐vinylimidazole) hydrogel nanoparticles prepared and hydrophobically modified in liposome reactors: Atomic force microscopy and dynamic light scattering study, Langmuir 2003, 19: 8086‐8093.

28. Kazakov S. and Levon K., Liposome‐nanogel structures for future pharmaceutical applications, Curr. Pharm. Des 2006, 12: 4713‐4728.

29. Patton J.N. and Palmer A.F., Engineering temperature‐sensitive hydrogel nanoparticles entrapping hemoglobin as a novel type of oxygen carrier, Biomacromolecules 2005, 6: 2204‐2212.

30. Patton J.N. and Palmer A.F., Photopolymerization of bovine hemoglobin entrapped nanoscale hydrogel particles within liposomal reactors for use as an artificial blood substitute, Biomacromolecules 2005, 6: 414‐424.

31. Schillemans J.P., Flesch F.M., Hennink W.E., and van Nostrum C.F., Synthesis of bilayer‐coated nanogels by selective cross‐linking of monomers inside liposomes, Macromolecules 2006, 39: 5885‐5890.


33. Van Thienen T.G., Demeester J., and De Smedt S.C., Screening poly(ethyleneglycol) micro‐ and nanogels for drug delivery purposes, International Journal of Pharmaceutics 2008, 351: 174‐185.

34. Rouser G., Fkeischer S., and Yamamoto A., Two dimensional then layer chromatographic separation of polar lipids and determination of phospholipids by phosphorus analysis of spots, Lipids 1970, 5: 494‐496.

35. Akinc A., Goldberg M., Qin J., Dorkin J.R. et al., Development of Lipidoid‐siRNA Formulations for Systemic Delivery to the Liver, Molecular Therapy 2009, 17: 872‐879.

36. Raemdonck K., Remaut K., Lucas B., Sanders N.N. et al., In situ analysis of single‐stranded and duplex siRNA integrity in living cells, Biochemistry 2006, 45: 10614‐10623.


38. Lucas B., Remaut K., Braeckmans K., Haustraete J. et al., Studying pegylated DNA complexes by dual color fluorescence fluctuation spectroscopy, Macromolecules 2004, 37: 3832‐3840.


40. Stubbe B.G., Horkay F., Amsden B., Hennink W.E. et al., Tailoring the swelling pressure of degrading dextran hydroxyethyl methacrylate hydrogels, Biomacromolecules 2003, 4: 691‐695.

41. Lucas B., Van Rompaey E., De Smedt S.C., Demeester J. et al., Dual‐color fluorescence fluctuation spectroscopy to study the complexation between poly‐l‐lysine and oligonucleotides, Macromolecules 2002, 35: 8152‐8160.

42. Payne C.K., Jones S.A., Chen C., and Zhuang X., Internalization and trafficking of cell surface proteoglycans and proteoglycan‐binding ligands, Traffic 2007, 8: 389‐401.

43. Bailey A.L. and Sullivan S.M., Efficient encapsulation of DNA plasmids in small neutral liposomes induced by ethanol and calcium, Biochim. Biophys. Acta 2000, 1468: 239‐252.

44. Simoes S., Slepushkin V., Duzgunes N., and Pedroso de Lima M.C., On the mechanisms of internalization and intracellular delivery mediated by pH‐sensitive liposomes, Biochim. Biophys. Acta 2001, 1515: 23‐37.

45. Skalko‐Basnet N., Tohda M., and Watanabe H., Uptake of liposomally entrapped fluorescent antisense oligonucleotides in NG108‐15 cells: conventional versus pH‐sensitive, Biol. Pharm. Bull. 2002, 25: 1583‐1587.

46. Malaekeh‐Nikouei B., Jaafari M.R., Tabassi S.A., and Samiei A., The enhancement of immunosuppressive effects of cyclosporine A on human T‐cells using fusogenic liposomes, Colloids Surf. B Biointerfaces. 2008, 67: 238‐244.

47. Gray M.J., Dallas N.A., Van B.G., Xia L. et al., Therapeutic targeting of Id2 reduces growth of human colorectal carcinoma in the murine liver, Oncogene 2008, 27: 7192‐7200.

48. Halder J., Kamat A.A., Landen C.N., Jr., Han L.Y. et al., Focal adhesion kinase targeting using in vivo short interfering RNA delivery in neutral liposomes for ovarian carcinoma therapy, Clin. Cancer Res. 2006, 12: 4916‐4924.

49. Landen C.N., Merritt W.M., Mangala L.S., Sanguino A.M. et al., Intraperitoneal delivery of liposomal siRNA for therapy of advanced ovarian cancer, Cancer Biol. Ther. 2006, 5: 1708‐1713.

203


204

50. Landen C.N., Jr., Chavez‐Reyes A., Bucana C., Schmandt R. et al., Therapeutic EphA2 gene targeting in vivo using neutral liposomal small interfering RNA delivery, Cancer Res. 2005, 65: 6910‐6918.

51. Lin Y.G., Immaneni A., Merritt W.M., Mangala L.S. et al., Targeting aurora kinase with MK‐0457 inhibits ovarian cancer growth, Clin. Cancer Res. 2008, 14: 5437‐5446.

52. Merritt W.M., Lin Y.G., Spannuth W.A., Fletcher M.S. et al., Effect of interleukin‐8 gene silencing with liposome‐encapsulated small interfering RNA on ovarian cancer cell growth, J. Natl. Cancer Inst. 2008, 100: 359‐372.

53. Villares G.J., Zigler M., Wang H., Melnikova V.O. et al., Targeting melanoma growth and metastasis with systemic delivery of liposome‐incorporated protease‐activated receptor‐1 small interfering RNA, Cancer Res. 2008, 68: 9078‐9086.

54. de Fougerolles A.R., Delivery vehicles for small interfering RNA in vivo, Hum. Gene Ther. 2008, 19: 125‐132. 55. Brandl M., Drechsler M., Bachmann D., Tardi C. et al., Preparation and characterization of semi‐solid

phospholipid dispersions and dilutions thereof, International Journal of Pharmaceutics 1998, 170: 187‐199. 56. Fretz M.M., Hogset A., Koning G.A., Jiskoot W. et al., Cytosolic delivery of liposomally targeted proteins

induced by photochemical internalization, Pharm. Res. 2007, 24: 2040‐2047. 57. Edwards K.A. and Baeumner A.J., DNA‐oligonucleotide encapsulating liposomes as a secondary signal

amplification means, Anal. Chem. 2007, 79: 1806‐1815. 58. Brandl M., Vesicular phospholipid gels: A technology platform, Journal of Liposome Research 2007, 17: 15‐

26. 59. Massing U., Cicko S., and Ziroli V., Dual asymmetric centrifugation (DAC) ‐ A new technique for liposome

preparation, Journal of Controlled Release 2008, 125: 16‐24. 60. Hirsch M., Ziroli V., Helm M., and Massing U., Preparation of small amounts of sterile siRNA‐liposomes with

high entrapping efficiency by dual asymmetric centrifugation (DAC), Journal of Controlled Release 2009, 135: 80‐88.

61. Soussan E., Cassel S., Blanzat M., and Rico‐Lattes I., Drug delivery by soft matter: matrix and vesicular carriers, Angew. Chem. Int. Ed Engl. 2009, 48: 274‐288.

62. Tseng Y.C., Mozumdar S., and Huang L., Lipid‐based systemic delivery of siRNA, Adv. Drug Deliv. Rev. 2009, doi: 10.1016/j.addr.2009.03.003.

summary & general conclusions | CHAPTER 8

205

CHAPTER 8 | Summary & general conclusions

Koen Raemdonck, J. Demeester and Stefaan C. De Smedt

Laboratory of General Biochemistry and Physical Pharmacy, Department of Pharmaceutics, Faculty of Pharmaceutical


CHAPTER 8 | summary & general conclusions

206


8 SUMMARY

RNA interference (RNAi) certainly is a hot topic among the scientific community, judging by the broad

financial investments and the tremendous research output to date.[1] This statement is further

illustrated through a PubMed search on the term ‘RNA interference’ (performed on June 15th 2009),

which yielded ~16.500 entries. RNAi is a naturally conserved gene silencing mechanism functioning in

eukaryotic cells and is activated by small interfering RNAs (siRNAs) that trigger the degradation of

mRNA in a sequence‐specific manner (CHAPTER 1). Besides its use as a laboratory tool in functional

genomics and drug target discovery, the therapeutic potential of RNAi by blocking the production of

disease‐causing proteins has also long been recognized. RNAi therapeutics have the advantage to act

(post)transcriptionally, while small molecular drugs are generally limited to interference on the

protein level. Thus, siRNAs may be designed to downregulate almost every gene transcript and hopes

are high for the future to make RNAi a therapeutic reality. In just over 10 years since the discovery of

RNAi, we already witnessed the transition of small interfering RNAs into the clinic and an overview of

siRNA compounds in active and already completed clinical trials is presented in Table 8.1.[1‐4]

The siRNAs that are in the most advanced stages of clinical evaluation are delivered locally at the site

of disease as a conventional saline formulation. Scientists however agree that to broaden the

therapeutic potential of RNAi, the principal challenge that remains is the hurdle of efficient siRNA

delivery in the human body. As summarized in chapter 1, many extra‐ and intracellular barriers are

encountered before siRNAs can be safely delivered to their site of action, which is the cytoplasm of

the target cells. Advanced RNAi delivery systems, either viral or non‐viral, are desperately needed

and constitute the main focus of the research on RNAi therapeutics. Since the RNAi machinery is

located in the cytoplasm, siRNA therapeutics need to be guided across the cellular membrane for

biological activity. However, owing to their high negative charge and relatively large size (~14.000

Da), naked siRNAs are not readily internalized by cells. Although the viral delivery systems are

generally the most effective to date, their clinical use is limited mainly due to safety issues. For this

reason, non‐viral delivery approaches have emerged as the safe(r) alternative. The experimental

207


chapters presented in this thesis mainly address non‐viral delivery of functional siRNAs across the

cellular barriers.

TABLE 8.1 | Completed and active clinical trials to date using siRNA drugs. More information can be found on the websites of the sponsors and/or www.clinicaltrials.gov

siRNA drug sponsor clinical target administration/formulation status

Bevasiranib

(Cand 5) OPKO Health Wet AMD Intravitreal injection/saline Phase III *

AGN‐211745

(sirna‐027) Allergan Wet AMD Intravitreal injection/saline Phase II

PF‐4523655

(RTP801i‐14)

Quark Pharmaceuticals/Pfizer

Wet AMD/DME Intravitreal injection/saline Phase II

ALN‐RSV01 Alnylam/Cubist/

Kyowa Hakko Kirin RSV infection Inhalation Phase II

QPI‐1002

(AKIi‐5/I5NP)

Quark Pharmaceuticals

AKI/

DGF in kidney transplant

Intravenous injection/saline Phase I/II(a)

TD101 IPCC/TransDerm PC Intradermal injection/saline Phase Ib

CALAA‐01 Calando

Pharmaceuticals Solid tumors

Intravenous injection/stabilized polymer

nanoparticle Phase I

ALN‐VSP (dual targeting)

Alnylam Liver

cancers/solid tumor

Intravenous injection/lipid nanoparticle

Phase I

* clinical trial terminated before completion, press release OPKO Health, March 6th 2009; IPCC, International pachyonychia congenita consortium; Wet‐AMD, wet age‐related macular degeneration; DME, diabetic macular edema; RSV, respiratory syncytial virus; AKI, acute kidney injury; DGF, delayed graft function; PC, pachyonychia congenita

In addition to siRNA delivery, chapter 1 discusses some important parameters that may influence the

magnitude and duration of the resulting RNAi effect. Gene silencing with siRNAs is fully reversible

and insight into the kinetics of gene suppression will be imperative to the design of siRNA‐based

treatment strategies. In this context, we briefly introduced transcriptional gene silencing,

208

http://www.clinicaltrials.gov/


mathematical modelling and combinatorial RNAi as new tools for the further optimization of genetic

interference.

In the exponentially growing field of RNAi, besides siRNA delivery also stability and degradation of

siRNA receives increasing attention. An essential issue that needs to be taken into account when

applying siRNAs, is their susceptibility to nuclease attack in biological fluids. Naked siRNA is highly

vulnerable to exo‐and endonucleases, leading to a short half‐life in serum and a wealth of reports in

the literature describe chemical modifications and formulation strategies to improve siRNA stability

and pharmacokinetics. With regard to the intracellular siRNA persistence, some conflicting studies

exist on whether or not the siRNA integrity significantly influences RNAi kinetics and therefore a

direct correlation between siRNA stability and RNAi duration is not yet conclusive. Techniques for the

investigation of the siRNA integrity after cellular internalization are therefore of substantial interest

to address the intracellular fate of siRNA. In CHAPTER 2, we describe a method based on fluorescence

resonance energy transfer (FRET) to assess the stability of siRNA in the intracellular

microenvironment. The method was validated by analyzing the loss of FRET as a function of

degradation by RNase A or through incubation in cellular extracts. In these experiments we used the

ratio of acceptor fluorescence to donor fluorescence upon donor excitation (FA/FD ratio) as a

parameter to evaluate the cleavage of both single‐stranded (ss) as double‐stranded (ds) siRNAs. Such

techniques may prove useful to explore the correlation between the intracellular fate of delivered

nucleic acids and their biological effectiveness, as already partially demonstrated in the literature.[5‐7]

A concept to induce a more robust RNAi response, as first mentioned in chapter 1, is through a

controlled release of siRNA in the cytoplasm of the target cell. The design of carriers that are able to

function as an siRNA depot across the cellular barriers, may induce a more sustained gene silencing

effect by maintaining the intracellular siRNA concentrations above the level required for activity for a

longer period of time. Following the broad applicability of hydrogels as controlled drug delivery

matrices and their experience in a biopharmaceutical context, we focused in this thesis on dextran

based hydrogels as carriers for siRNA. We first evaluated if biodegradable dextran microgels could be

produced to achieve a controlled release of the encapsulated siRNA (CHAPTER 3). For the preparation

of the microgels, both a water‐in‐water emulsion as well as a water‐in‐oil emulsion were tested.

Preformed microscopic gel particles (microgels), composed of dextran hydroxyethyl methacrylate

(dex‐HEMA) and further functionalized with cationic methacrylate monomers (2‐

(dimethylamino)ethyl methacrylate or DMAEMA), were able to bind siRNA through electrostatic

interaction. Confocal microscopy revealed that the siRNA duplexes were mainly adsorbed onto the

microgel surface but that they could also penetrate deeper in the pores of the gel network at higher

concentrations. The HEMA moieties are coupled to the dextran backbone via a labile carbonate ester

209


linkage. Hydrolysis of this ester function, cleaves the crosslinks in between the dextran chains,

thereby again releasing the complexed siRNA. The degradation time of the dex‐HEMA microgels is

governed by their initial crosslink density and we showed in chapter 3 that by tailoring the network

properties we are able to control the release kinetics of the encapsulated siRNA over several days.

The vast majority of non‐viral delivery approaches makes use of (cationic) polymer or lipid‐based

nanosized particles to transfer complexed or encapsulated siRNA across the cellular membrane. A

principal requirement to assess the effectiveness of our siRNA delivery concept, was therefore to

scale down our cationic hydrogel particles to nanoscopic dimensions, in order to ensure efficient

cellular internalization. CHAPTER 4 briefly describes some interesting drug delivery applications based

on nanogels as the drug carrier. This chapter highlights several eye‐catching examples of nanogel

design and illustrates nanogels in terms of versatile stimuli‐responsive drug delivery systems.

Although the number of publications addressing this issue is growing exponentially since the

beginning of this century, to this date only a very limited number of reports has evaluated nanogels

for RNAi. In CHAPTER 5, we employed an inverse miniemulsion photopolymerization method for the

production of hydrolysable cationic dex‐HEMA nanogels. In contrast to chapter 3, this time we used a

methacrylate monomer carrying a quaternary ammonium group (trimethylammonium ethyl

methacrylate or TMAEMA) for copolymerization to incorporate positively charged amines into the

hydrogel network. Owing to the positive charge, the nanogels are able to complex siRNA with a high

loading capacity exceeding 70 wt%. Confocal microscopy and flow cytometry analysis revealed that

large numbers of siRNA loaded nanogels can effectively be internalized by human hepatoma cells

(Huh‐7 ) without relevant cytotoxicity. It is generally accepted that the majority of non‐viral carriers is

taken up through an endocytic process where most particles end up being trapped in acidified

(endo)lysosomes. We found that also our nanogel particles accumulated to a high extent in these

degradative organelles. Nevertheless, dependent on the amount of nanogels applied on the cells, still

a substantial suppression of luciferase expression could be obtained, indicating that a fraction of the

complexed siRNA reaches the cell cytoplasm where it is recognized by the RNAi machinery.

Moreover, this RNAi effect was almost absent for non‐degradable cationic dextran nanogels,

suggesting an influence of dex‐HEMA hydrolysis on the delivery of siRNA. On the other hand, the

extent of gene silencing could be markedly enhanced by applying validated endosomolytic tools, such

as photochemical internalization (PCI) and the co‐complexation of an influenza‐derived fusogenic

peptide (diINF‐7). These findings provide evidence that the endosomal sequestration of our siRNA

loaded nanogels partially abrogates the potential RNAi outcome.

While the magnitude of target gene suppression may be a good indicator for the success of an siRNA

treatment, evaluation of the kinetics of the RNAi effect is perhaps a more favourable parameter to

210


decide on the efficacy of the treatment. The time‐window that the siRNA therapy is able to keep the

intracellular protein titers below the threshold that is necessary to obtain the desired downstream

phenotype, will strongly determine the dosage frequency in clinical therapy. In CHAPTER 6, we

therefore analyzed the kinetics of gene inhibition for dex‐HEMA‐co‐TMAEMA nanogels with varying

crosslink density and compared the results with a commercially available lipid‐based transfection

reagent, i.e. lipofectamine™RNAiMAX. The transfection experiments revealed that the duration of

gene silencing (as assessed in Huh‐7 cells with firefly luciferase and EGFP as target respectively) is

largely independent on the type of carrier used. The magnitude of downregulation does however

differ for the investigated nanogel formulations with in general lower gene silencing efficiency for

higher crosslinked nanogels. Most likely, the extent of gene silencing is governed by the amount of

siRNA that is initially delivered in the cell cytoplasm while the recovery of gene expression to its

steady state level is caused by sequential cell divisions that dilute the intracellular siRNA pool until

insufficient active drug resides in the cytoplasm. In addition, we found that application of PCI, even

48 hours post‐transfection, was able to improve and lengthen the RNAi outcome for our dextran

nanogels, with up to 25% additional gene suppression. In conclusion, this result indicates that PCI is

able to liberate a fraction of the siRNA (‐loaded nanogels) that remain trapped in intracellular

compartments, which can be regarded as a potential strategy to prolong the therapeutic response

when endosomal escape is effect‐limiting, albeit that further optimization is required.

From the previous chapters it may be clear that to achieve an intracellular time‐controlled siRNA

delivery and to evaluate to what extent this may contribute to a more durable gene inhibition, a

transition of siRNA loaded nanogels from the endosomal lumen into the cytoplasm is imperative.

Following pioneering work by Van Thienen et al.,[8] CHAPTER 7 discusses the design of hybrid lipid‐

enveloped nanogels (LENs), where a multifunctional lipid shell is entrusted with the task to bypass

the major extracellular barriers to allow disposition of biodegradable siRNA loaded nanogels in the

target cell cytoplasm. These hybrid nanocomposites can be prepared by using liposomes as

nanoreactors for selective UV polymerization of dex‐HEMA in their aqueous core. SiRNA can be

loaded into LENs by passive hydration of a preformed lipid film with an encapsulation efficiency (non‐

optimized) of >30%. Both extrusion and probe sonication are successfully applied for LENs

production, as evidenced by dynamic light scattering. Although a majority of liposomal drug delivery

systems for RNAi typically comprises a cationic lipid, we chose to use a charge neutral lipid

combination for LENs development instead, consisting of an equimolar mixture of DOPC and DOPE.

In spite of this, preliminary cellular experiments suggested that Huh‐7 cells still quantitatively

internalize DOPC:DOPE coated LENs. Future experiments on LENs will be oriented toward a further

optimization of the siRNA encapsulation efficiency and the assessment of their RNAi potential.

211


212

GENERAL CONCLUSIONS

In conclusion, this thesis comprises a novel FRET based approach for the intracellular assessment of

small interfering RNA (siRNA) integrity, which could aid in clarifying the correlation between

intracellular siRNA fate and the eventual RNAi outcome. Besides siRNA stability, this thesis also

describes the design of cationic and biodegradable nanogels for the time‐controlled delivery of

siRNA. We provide evidence that these nanogels can effectively deliver active siRNA across the

cellular barrier, leading to substantial and durable gene silencing. Endosomal escape is identified as

the predominant barrier confining the full RNAi potential, opening up new opportunities to further

improve this delivery concept. It is conceivable that, although RNAi can generally be applied to

interfere with the expression of virtually any gene, several distinct in vivo delivery agents will be

needed depending on the disease target and the chosen route of administration. Our nanogels may

well claim a future position in this ensemble of siRNA delivery systems.

REFERENCE LIST

1. Haussecker D., The business of RNAi therapeutics, Hum. Gene Ther. 2008, 19: 451‐462. 2. Behlke M.A., Chemical modification of siRNAs for in vivo use, Oligonucleotides. 2008, 18: 305‐319. 3. de Fougerolles A.R., Delivery vehicles for small interfering RNA in vivo, Hum. Gene Ther. 2008, 19: 125‐132. 4. Whelan J., First clinical data on RNAi, Drug Discov. Today 2005, 10: 1014‐1015. 5. Jagannath A. and Wood M.J., Localization of double‐stranded small interfering RNA to cytoplasmic

processing bodies is Ago2 dependent and results in up‐regulation of GW182 and Argonaute‐2, Mol. Biol. Cell 2009, 20: 521‐529.

6. Jarve A., Muller J., Kim I.H., Rohr K. et al., Surveillance of siRNA integrity by FRET imaging, Nucleic Acids Research 2007, 1‐13.



213

214

samenvatting en algemene besluiten | APPENDIX A

215

APPENDIX A | SAMENVATTING EN

ALGEMENE BESLUITEN

APPENDIX A | samenvatting en algemene besluiten

216


A SAMENVATTING VAN HET DOCTORAAL PROEFSCHRIFT

Het is slechts 11 jaar geleden dat Andrew Fire and Craig Mello in hun Nature publicatie de vreemde

observatie beschreven dat de exogene toediening van dubbelstrengig RNA (dsRNA) aan de nematode

C. elegans resulteerde in de onderdrukking van het complementaire gen. Zij waren de eersten die de

term RNA interference (RNAi) hanteerden en werden voor hun ontdekking beloond met de Nobelprijs

voor Geneeskunde en Fysiologie in 2006. Dit doet reeds vermoeden dat het hier een belangrijke

wetenschappelijke doorbraak betreft. RNAi is dan ook daadwerkelijk een ‘hot topic’ in de exacte

wetenschap, hetgeen we verder ook kunnen afleiden uit de immense financiële investeringen van de

afgelopen jaren en de uitgebreide wetenschappelijke output omtrent RNAi die men vandaag reeds in

de vakliteratuur kan terugvinden.

In essentie is RNAi een intracellulair mechanisme dat actief is in eukaryote cellen en waarvan het

biochemisch raderwerk in de loop van de evolutie bewaard is gebleven. De RNAi machine wordt

geactiveerd door de introductie van dsRNA in het cytoplasma van de doelwitcel, hetgeen uiteindelijk

leidt tot de sequentie‐specifieke onderdrukking van een doelwitgen op het posttranscriptioneel

niveau (HOOFDSTUK 1). Naar aanleiding van de ontdekking in 2001 dat RNAi ook functioneel zou zijn in

menselijke cellen, nam het onderzoek naar hoe dit mechanisme zou kunnen gemanipuleerd worden

voor therapeutische doeleinden een hoge vlucht. RNAi biedt immers vele mogelijkheden voor de

behandeling van ziekten met een gekende genetische oorzaak, waaronder diverse kankers,

neurondegeneratieve aandoeningen, maar ook virale infecties. RNAi therapeutica bieden

daarenboven het voordeel messenger RNA (mRNA) als doelwit te hebben, in tegenstelling tot vele

conventionele geneesmiddelen waarvan de werking meestal beperkt is tot het eiwitniveau. Met het

ontrafelen van het menselijk genoom en onze toenemende kennis omtrent de functionaliteit van

menselijke genen, zou RNAi kunnen leiden tot een brede waaier van geneesmiddelen tegen eender

welk (genetisch) klinisch doelwit.

Intensief onderzoek heeft kleine intermediaire dsRNA fragmenten, afgekort als siRNAs (small

interfering RNAs), geïdentificeerd als de authentieke effectoren van de genetische interferentie. Deze

RNA moleculen kunnen eenvoudig op grote schaal aangemaakt worden via chemische synthese en in

het merendeel van de publicaties tot op heden omtrent RNAi therapeutica, staan siRNAs centraal.

217


Uiteindelijk is men nu in staat siRNAs te ontwerpen die, in functie van hun nucleotide sequentie,

mogelijks elk gentranscript kunnen onderdrukken. In iets meer dan 10 jaar sinds de ontdekking van

RNAi waren we reeds getuige van de overgang van siRNA naar de klinische testfase, hetgeen

nogmaals het belang en de mogelijkheden van RNAi onderstreept. Tabel 8.1 (zie hoofdstuk 8, pagina

208) toont een overzicht van de siRNA therapeutica die tot op heden reeds werden geëvalueerd in

klinische testen of in de toekomst aan evaluatie zullen onderworpen worden.

Deze data tonen aan dat de siRNAs in de meer gevorderde stadia van klinische evaluatie vooral lokaal

worden toegediend, geformuleerd als een 0.9% NaCl zoutoplossing. Wetenschappers zijn het er

echter over eens dat, om het therapeutisch potentieel van RNAi in de breedte te verzilveren, een

efficiënte afgifte van siRNA in het menselijk lichaam het belangrijkste obstakel blijft. Zoals

samengevat in HOOFDSTUK 1 bestaan er vele extra‐ en intracellulaire barrières voor nucleïnezuren als

geneesmiddel die overwonnen dienen te worden om hen een veilige doortocht naar het cytoplasma

van de doelwitcel te garanderen. In dit proefschrift wordt voornamelijk de focus geplaatst op de

hindernissen op cellulair niveau. Aangezien het RNAi raderwerk hoofdzakelijk is gelokaliseerd in het

cytoplasma, moeten siRNAs immers intact overheen de cellulaire membraan worden geloodst. Het

sterk negatief geladen karakter van siRNA (≥ 40 negatieve ladingen per molecule) en hun relatief

grote afmetingen (~14 kDa) laten niet zomaar toe dat deze spontaan door cellen worden

opgenomen. Onder meer om deze reden vormt het evalueren van nanoscopische dragers voor de

formulatie en afgifte van small RNAs een belangrijk segment in het onderzoek rond RNAi. Zowel

virale als niet‐virale afgifteystemen (of vectoren) werden reeds ontwikkeld voor de intracellulaire

vrijgave van RNAi effectoren, waarvan de virale vectoren ontegensprekelijk de meest efficiënte zijn.

Ondanks hun efficiëntie is het klinisch gebruik van virale vectoren gelimiteerd door tot op heden

onoverkoombare veiligheidsproblemen, hetgeen het onderzoek voedt naar niet‐virale alternatieven.

De meerderheid van wetenschappelijke publicaties beschrijft hierin het gebruik van synthetische

vetten en polymeren voor de complexatie en intracellulaire afgifte van siRNA, waarvan enkele

belangrijke voorbeelden ook in HOOFDSTUK 1 worden aangehaald.

Behalve siRNA afgifte, bespreekt dit hoofdstuk eveneens de meest voorname parameters die de

grootteorde en levensduur van het RNAi effect beïnvloeden. Genonderdrukking via siRNAs is volledig

reversibel en meer inzichten in de kinetiek van dit effect zullen cruciaal zijn voor het opstellen van

een geschikt doseringsschema. In de context van een meer duurzaam RNAi effect worden

genonderdrukking op het transcriptioneel niveau (transcriptional gene silencing of TGS), het gebruik

van mathematische modellen en combinatoriële RNAi (coRNAi) kort besproken als nieuwe

strategieën voor de verdere optimalisatie van genetische interferentie.

218


In het exponentieel groeiende veld van RNAi, gaat er behalve naar methodes voor verbeterde siRNA

afgifte ook aandacht uit naar de stabiliteit en afbraak van siRNA. Het is essentieel bij het gebruik van

siRNA als geneesmiddel dat de stabiliteit van de siRNA duplex en zijn vatbaarheid voor afbraak door

nucleasen in biologische media wordt nagegaan. Naakt siRNA (i.e. niet geformuleerd in virale of niet‐

virale nanoscopische dragers) wordt snel afgebroken door exo‐ en endonucleasen, hetgeen leidt tot

een zeer korte halfwaardetijd in de algemene circulatie. De literatuur bevat daarom ook vele

publicaties die chemische modificaties en formulatie strategieën bespreken met als doel de siRNA

stabiliteit en farmacokinetiek te verbeteren. Met betrekking tot het belang van de intracellulaire

siRNA stabiliteit zijn de meningen echter verdeeld. Tegenstrijdige rapporten of de siRNA integriteit al

dan niet een invloed zou hebben op het levensduur van het RNAi effect maakt de correlatie tussen

beide nog niet beslissend. Technieken om de siRNA integriteit na te gaan, volgend op cellulaire

opname, zijn daarom van belang om het intracellulair lot van siRNA te verifiëren. In HOOFDSTUK 2

wordt een methode voorgesteld, gebaseerd op fluorescentie resonantie energie transfer (FRET), om

de stabiliteit van dubbel fluorescent gemerkt siRNA te evalueren in de cellulaire micro‐omgeving. De

methode werd gevalideerd door het wegvallen van FRET te detecteren in functie van siRNA

degradatie door RNase A of door incubatie in celextracten. In deze experimenten werd de

verhouding van acceptor fluorescentie tot donor fluorescentie bij donor excitatie (FA/FD ratio)

gebruikt als parameter voor de stabiliteit van zowel enkelstrengig (ssRNA) als dubbelstrengig RNA

(dsRNA). Dergelijke technieken kunnen bruikbaar zijn om de correlatie tussen het intracellulair

gedrag van siRNA en het uiteindelijke biologische effect meer in detail te bestuderen, zoals reeds

gedeeltelijk aangetoond in de literatuur.

Zoals reeds aangehaald is het transiënte effect een mogelijk nadeel van het gebruik van siRNA als

therapeuticum, waardoor meerdere toedieningen noodzakelijk zullen zijn om het therapeutisch

effect gedurende langere tijd aan te houden. Een concept om een meer robust RNAi effect te

verkrijgen is via de gecontroleerde vrijstelling van siRNA in het cytoplasma van de doelwitcel. Het

ontwerpen van nanoscopische dragers die fungeren als een intracellulair depot, zou een verlengde

genonderdrukking kunnen verwezenlijken door de intracytoplasmatische siRNA concentratie

gedurende langere tijd in het therapeutisch venster te houden. In navolging van de brede

toepasbaarheid van hydrogelen als matrices voor gecontroleerde afgifte van geneesmiddelen en hun

veelvuldig gebruik in een biofarmaceutische context, werd in deze thesis de focus geplaatst op

dextraan hydrogelen als dragers voor siRNA. Hydrogelen kunnen gedefinieerd worden als 3D

netwerken, opgebouwd uit hydrofiele polymeren, die in staat zijn om grote hoeveelheden water of

fysiologische vloeistof op te nemen. Sinds hun introductie voor biologisch gebruik bijna 50 jaar

219


geleden, nemen hydrogelen een belangrijke plaats in voor allerhande biomedische en

farmaceutische toepassingen.

Hydrogelen kunnen worden aangemaakt in diverse geometrieën. HOOFDSTUK 3 beschrijft de aanmaak

van microscopische biodegradeerbare dextraan hydrogelen (i.e. microgelen) via zowel een w/w

emulsie in polyethyleenglycol als een w/o emulsie in minerale olie. Voorgevormde microgelen,

bestaande uit dextraan hydroxyethyl methacrylaat (dex‐HEMA) en verder gefunctionaliseerd met

kationische methacrylaat monomeren (2‐(dimethylamino)ethyl methacrylaat of DMAEMA), bleken in

staat siRNA op te nemen via elektrostatische aantrekking. De HEMA groepen zijn gekoppeld aan

dextraan via een onstabiele ester binding. Hydrolyse van deze ester functie, klieft de verbinding

tussen de dextraan ketens, waardoor het gelnetwerk langzaam uit elkaar valt en het gecomplexeerde

siRNA wordt vrijgesteld. Aangezien de degradatie snelheid van dex‐HEMA microgelen bepaald wordt

door de initiële dichtheid aan crosslinks, kan de vrijstellingskinetiek van siRNA gecontroleerd worden

door de dichtheid van het hydrogel netwerk te laten variëren.

Een primaire vereiste om de efficiëntie van het voorgestelde siRNA afgifte systeem te bepalen, is het

vertalen ervan naar nanoscopische dimensies om cellulaire internalisatie mogelijk te maken.

HOOFDSTUK 4 presenteert een overzicht van enkele opvallende applicaties waarin nanoscopische

hydrogelen (nanogelen) gebruikt worden als geneesmiddeldragers voor de gecontroleerde vrijstelling

van diverse therapeutica. Ondanks het toenemend aantal publicaties omtrent nanogelen for

farmaceutische toepassingen het voorbije decennium, zijn de rapporten die nanogelen beschrijven

voor siRNA afgifte tot op heden nog op één hand te tellen. De aanmaak van positief geladen en

hydrolyseerbare dex‐HEMA nanogelen via een inverse mini‐emulsie fotopolymerisatie staat

beschreven in HOOFDSTUK 5. In vergelijking met hoofstuk 3 werd ditmaal een methacrylaat

monomeer met een kwaternaire stikstofgroep (2‐(trimethylammonium)ethyl methacrylaat of

TMAEMA) gekozen voor copolymerisatie met dex‐HEMA. Dankzij de incorporatie van positieve

ladingen in het nanogel netwerk, hebben deze een grote ladingscapaciteit voor siRNA. Confocale

microscopie en flow cytometrie leverden het bewijs dat grote hoeveelheden siRNA beladen

nanogelen efficiënt konden opgenomen worden door humane hepatoma cellen (Huh‐7) onder

behoud van de celviabiliteit. Het wordt algemeen aangenomen dat voor het merendeel van de niet‐

virale dragers de cellulaire internalisatie gebeurt via endocytose, waarbij een grote fractie van de

opgenomen partikels uiteindelijk eindigt in verzuurde (endo)lysosomen. Dit bleek ook voor onze

kationische nanogelen het geval. Ongeacht deze nadelige intracellulaire eindbestemming, toonden

onze experimenten toch een substantiële en sequentie‐specifieke onderdrukking aan van het

luciferase gen na transfectie met afbreekbare dex‐HEMA nanogelen. Dit is een onrechtstreekse

aanwijzing dat een deel van het gecomplexeerd siRNA het cytoplasma bereikt alwaar het opgenomen

220


wordt in het RNAi raderwerk. Bovendien bleek het RNAi effect bijna afwezig wanneer de transfectie

werd uitgevoerd met stabiele (niet afbreekbaar door hydrolyse bij fysiologische pH) dextraan

nanogelen, hetgeen doet vermoeden dat dex‐HEMA hydrolyse een invloed heeft op de efficiëntie

van siRNA afgifte. Anderzijds kon het genonderdrukkende effect markant verbeterd worden door het

toepassen van gevalideerde endosomolytische hulpmiddelen, zoals fotochemische internalisatie

(PCI) en de co‐complexatie van siRNA beladen nanogelen met een fusogeen peptide afgeleid van de

hemagluttinine HA2 subeenheid van het influenza virus. Deze data leveren het bewijs aan dat de

endolysosomale accumulatie van de siRNA nanogelen nefast is voor een optimale siRNA afgifte in het

cytoplasma van de doelwitcel.

Terwijl de omvang van de genonderdrukking een goede indicator kan zijn voor het succes van een

siRNA behandeling, is de kinetiek van het RNAi effect hiervoor misschien beter geschikt. De tijd dat

de siRNA behandeling de intracellulaire concentratie van het doelwit proteïne beneden de drempel

kan houden die nodig is om het gewenste fenotype te bekomen, zal in belangrijke mate de

doseringsfrequentie bepalen in klinische therapie. In HOOFDSTUK 6 wordt o.a. om deze reden de

kinetiek van genonderdrukking nagegaan voor dex‐HEMA‐co‐TMAEMA nanogelen met variërende

crosslink densiteit. De transfectie experimenten toonden aan dat de kinetiek van genonderdrukking

grotendeels onafhankelijk is van de gebruikte nanogel formulatie. De grootteorde van het

onderdrukkend effect daarentegen bleek wel te verschillen met in het algemeen een kleiner effect

voor sterker gecrosslinkte nanogelen. Het lijkt aannemelijk te veronderstellen dat het maximaal

biologisch effect bepaald wordt door de hoeveelheid siRNA dat initieel het cytoplasma bereikt terwijl

het herstel van de genexpressie naar steady‐state gedicteerd wordt door sequentiële celdelingen die

de intracellulaire pool van actief siRNA gestaag reduceren. Daarbij werd eveneens aangetoond dat de

toepassing van PCI, zelfs 48 h post‐transfectie, het RNAi effect in belangrijke mate kon verbeteren

(tot 25% bijkomende genonderdrukking) en verlengen. Dit resultaat geeft aan dat PCI in staat is om

een fractie van de siRNA beladen nanogelen, die in een vesiculaire micro‐omgeving gestrand waren,

vrij te stellen. Dit vormt een mogelijke strategie om de therapeutische respons te verlengen in het

geval dat de endosomale barriere limiterend is, alhoewel deze strategie verdere optimalisatie vereist.

Uit de vorige hoofdstukken blijkt duidelijk dat om een intracellulaire gecontroleerde vrijstelling van

siRNA mogelijk te maken en de implicaties na te gaan in de context van een meer duurzame

genonderdrukking, een verbeterde overgang van siRNA beladen nanogelen van het endosomale

lumen naar het cytoplasma cruciaal is. In navolging van vernieuwend onderzoek, uitgevoerd door

Van Thienen et al. in 2005 bediscussieert HOOFDSTUK 7 het ontwerp van nanogelen omgeven door

een vetlaagje (lipid‐enveloped nanogels or LENs). De multifunctionele lipide mantel wordt in dit

concept belast met de opdracht om de belangrijkste extracellulaire barrières te overwinnen (i.e.

221


222

stabiliteit in de bloedstroom, vermijden van opsonizatie en een verbeterde mobiliteit in de

extracellulaire matrix). Bovendien anticiperen we op basis van literatuurgegevens een positieve

invloed op de accumulatie van intacte nanogelen in het cytoplasma, die daar als siRNA depot kunnen

fungeren. Hiertoe kan beroep gedaan worden op fusogene lipiden, die de eigenschap hebben te

versmelten met de endosomale membraan. Deze vorm van hybride nanopartikels werd aangemaakt

door liposomen in te schakelen als nanoreactor voor de selectieve UV polymerisatie van dex‐HEMA

in hun waterig lumen. Het opladen met siRNA gebeurt door passieve hydratatie van een

voorgevormde lipide film waarbij een encapsulatie efficiëntie werd bereikt van 30%. Zowel extrusie

als ultrasonicatie werden succesvol ingezet voor de aanmaak van LENs, zoals aangetoond via

dynamische lichtverstrooiing. In contrast met de vele liposomale formulaties voor RNAi die meestal

een kationisch lipide bevatten voor complexatie met het anionisch siRNA, werd in dit hoofdstuk

geopteerd voor een equimolaire mix van neutrale lipiden. Ondanks hun ladingsneutraliteit, kon via

flow cytometrie toch een duidelijke dosis afhankelijke Huh‐7 cellulaire opname waargenomen

worden in preliminaire experimenten. Naar de toekomst toe zullen LENs experimenten zich

hoofdzakelijk concentreren op de verdere optimalisatie van siRNA encapsulatie en het evalueren van

hun RNAi potentieel.

ALGEMENE BESLUITEN

Samenvattend beschrijft dit proefschrift een nieuwe methode, gebaseerd op FRET, met als

voornaamste doel de in situ evaluatie van siRNA stabiliteit in levende cellen. Deze methode moet het

mogelijk maken om bijkomende informatie te verkrijgen omtrent de correlatie tussen de

intracellulaire siRNA integriteit en het uiteindelijke RNAi effect. Daarnaast werd een nieuw concept

voorgesteld voor de intracellulaire gecontroleerde vrijstelling van siRNA. Hiertoe werd siRNA

gecomplexeerd in biodegradeerbare nanoscopische hydrogelen (i.e. nanogelen). Significante

onderdrukking van een reportergen vormde een onrechtstreekse aanwijzing dat deze nanogelen in

staat zijn actief siRNA in het cytoplasma van de doelwitcel af te leveren. Het ontsnappen van

geïnternaliseerde nanogelen uit de endosomen werd echter geïdentificeerd als de voornaamste

cellulaire barrière, hetgeen het RNAi potentieel gedeeltelijk beperkt. Tegelijkertijd biedt deze

observatie nieuwe opportuniteiten om het nanogel concept verder te verfijnen. Het is aannemelijk

dat afhankelijk van de te behandelen pathologie en de toedieningswijze, een grote verscheidenheid

aan siRNA dragers nodig zal zijn wanneer siRNAs de stap zetten naar nieuwe klinische toepassingen.

Het is best mogelijk dat ook nanogelen hierin hun plaats zullen opeisen.

curriculum vitae | APPENDIX B

223

APPENDIX B | CURRICULUM VITAE

APPENDIX B | curriculum vitae

224


B PERSONALIA

Name RAEMDONCK

First name KOEN

Nationality Belgian

Place and date of birth Dendermonde, 28‐08‐1981

Marital status Married

Private address Nieuwbrugkaai 70

9000 Gent

Telephone + 32 (0)485 28 08 12

Professional address Lab of General Biochemistry and Physical Pharmacy

Faculty of Pharmaceutical Sciences – Ghent University

Harelbekestraat 72

9000 Ghent

Telephone +32 (0)9 264 80 95

Fax +32 (0)9 264 81 89

E‐mail: [email protected]

URL http://www.biofys.ugent.be

225

mailto:[email protected]

http://www.biofys.ugent.be/


LANGUAGES

Dutch mother tongue

English fluent

French good

DEGREES

June, 2004 pharmacist with great distinction

Faculty of Pharmaceutical Sciences, Ghent University

INTERNATIONAL PEER REVIEWED PUBLICATIONS

Raemdonck K., Remaut K., Lucas B., Sanders N.N., Demeester J., and De Smedt S.C., In situ analysis of

single‐stranded and duplex siRNA integrity in living cells. Biochemistry 2006, 45: 10614‐10623.

Remaut K., Lucas B., Raemdonck K., Braeckmans K., Demeester J., and De Smedt S.C., Can we better

understand the intracellular behavior of DNA nanoparticles by fluorescence correlation

spectroscopy?. Journal of Controlled Release 2007, 121: 49‐63.

Remaut K., Lucas B., Raemdonck K., Braeckmans K., Demeester J., and De Smedt S.C., Protection of

oligonucleotides against enzymatic degradation by pegylated and nonpegylated branched

polyethyleneimine. Biomacromolecules 2007, 8: 1333‐1340.

Van Thienen T.G., Raemdonck K., Demeester J., and De Smedt S.C., Protein release from

biodegradable dextran nanogels. Langmuir 2007, 23: 9794‐9801.

Buyens K., Lucas B., Raemdonck K., Braeckmans K., Vercammen J., Hendrix J., Engelborghs Y., De

Smedt S.C., and Sanders N.N., A fast and sensitive method for measuring the integrity of siRNA‐

carrier complexes in full human serum. Journal of Controlled Release 2008, 126: 67‐76.

226


Raemdonck K., Van Thienen T.G., Vandenbroucke R.E., Sanders N.N., Demeester J., and De Smedt

S.C., Dextran microgels for time‐controlled delivery of siRNA. Advanced Functional Materials 2008,

18: 993‐1001.

Raemdonck K., Vandenbroucke R.E., Demeester J., Sanders N.N., and De Smedt S.C., Maintaining the

silence: reflections on long‐term RNAi. Drug Discovery Today 2008, 13: 917‐931.

Raemdonck K., Naeye B., Buyens K., Vandenbroucke R.E., Høgset A., Demeester J., De Smedt S.C.,

Biodegradable dextran nanogels for RNAi: focusing on endosomal escape and intracellular siRNA

delivery, Advanced Functional Materials 2009, 19: 1406‐1415. Press release (2009‐05‐28) on

www.pcibiotech.com.

Raemdonck K., Demeester J., De Smedt S.C., Advanced nanogel engineering for drug delivery, Soft

Matter 2009, 5: 707‐715. Invited.

Naeye B., Raemdonck K., Demeester J., and De Smedt S.C., PEGylation strategies for cationic dextran

nanogels as siRNA delivery agents. Manuscript in preparation.

CONFERENCE PROCEEDINGS

Raemdonck K., Remaut K., Lucas B., Sanders N.N., Demeester J., De Smedt S.C., Real‐time Monitoring

of Single‐Stranded and Duplex Short Interfering RNA Integrity inside living cells. Proceedings of the 9th

European Symposium on Controlled Drug Delivery 2006, 241‐243.

Raemdonck K., Remaut K., Lucas B., Sanders N.N., Demeester J., De Smedt S.C., Real‐time Monitoring

of Single‐Stranded and Duplex Short Interfering RNA Integrity inside living cells. Molecular Therapy

2006, 13: S273.

Raemdonck K., Demeester J., De Smedt S.C., Intracellular time‐controlled delivery of siRNA from

cationic nanogel depots. Molecular Therapy 2008, 16: S245.

227

http://www.pcibiotech.com/


ORAL PRESENTATIONS (presenting author)

Raemdonck K., Remaut K., Lucas B., Sanders N.N., Demeester J., De Smedt S.C., In situ analysis of

single‐stranded and duplex siRNA integrity in living cells. Spring Meeting of the Belgian‐Dutch

Biopharmaceutical Society, Janssen Pharmaceutica, Beerse (Belgium), May 22th 2006.

De Smedt S.C., Lucas B., Remaut K., Raemdonck K., Braeckmans K., Sanders N.N., Demeester J., Can

we better understand the intracellular behaviour of nucleic acid containing nanoparticles by

fluorescence correlation spectroscopy? Nanomedicine and drug delivery symposium, Omaha,

Nebraska (USA), October 8th‐10th 2006.

Raemdonck K., Demeester J., De Smedt S.C., Dextran micro‐and nanogels for controlled release of

siRNA. Pre‐satellite Meeting of the Pharmaceutical Sciences World Congres, Amsterdam (The

Netherlands), April 20th‐21st 2007.

Naeye B., Raemdonck K., Demeester J. and De Smedt S.C., Optimization of biodegradable dextran

nanogels for controlled siRNA delivery. MediTrans 2nd annual meeting, Rehovot (Israel), March 26th‐

29th 2009.

Raemdonck K., Naeye B., Demeester J., De Smedt S.C., Evaluation of a dextran nanogel concept for

intracellular siRNA delivery. EuroNanoMedicine, Bled (Slovenia), September 28th‐30th 2009.

POSTER PRESENTATIONS (presenting author)

Raemdonck K., Remaut K., De Smedt S.C., Demeester J., Studying the degradation of single‐ and

double‐stranded short interfering RNA in living cells by means of FCS‐FRET

• 1st Young Scientist Day of the Belgian Biophysical Society, Ghent (Belgium), May 19th 2005.

• Biopharmacy Day, Spring Meeting of the Belgian‐Dutch Biopharmaceutical Society, Ghent

(Belgium), May 20th 2005.

228


Raemdonck K., Remaut K., De Smedt S.C., Demeester J., Real‐time monitoring of single‐stranded and

uplex short interfering RNA integrity inside living cells

• RNAi Europe, Amsterdam (The Netherlands), September 28th‐29th 2005.

• Biopharmacy Day, Autumn Meeting of the Belgian‐Dutch Biopharmaceutical Society, Utrecht

(The Netherlands), December 2nd 2005.

Raemdonck K., Remaut K., Lucas B., Sanders N.N., Demeester J., De Smedt S.C.*, Real‐time

Monitoring of Single‐Stranded and Duplex Short Interfering RNA Integrity inside Living Cells

• 9th European Symposium on Controlled Drug Delivery, Noordwijk aan Zee (The Netherlands),

April 5th‐7th 2006.

• American Society of Gene Therapy annual meeting, Baltimore, MD (USA) May 31st‐June 4th

2006.

• 33rd Annual meeting of the Controlled Release Society, Vienna (Austria) July 22nd‐July 26th

2006 (*presenting author).

• 13th International symposium on recent advances in drug delivery systems, Salt Lake City, UT

(USA), February 26th‐28th 2007.

Raemdonck K., Demeester J., and De Smedt S.C., Towards dextran based nanogels for the

intracellular delivery of siRNA

• 13th International symposium on recent advances in drug delivery systems, Salt Lake City, UT


Raemdonck K., Demeester J., and De Smedt S.C.*, Cytosolic nanogel depots for time‐controlled

siRNA delivery

• 11th annual Meeting of the American Society of Gene Therapy (ASGT), Boston, MA (USA) May

28th‐June 1st 2008

• 2nd International Symposium on Cellular Delivery of Therapeutic Macromolecules, Cardiff

University (Wales), June 22nd‐June 25th 2008

• RNAi Europe, Stockholm (Sweden), September 16th‐18th 2008

229


230

• Right Symposium “RNA interference technology as Human Therapeutic Tool”, Brussels

(Belgium), November 3rd‐5th 2008 (* presenting author)

Raemdonck K., Demeester J., and De Smedt S.C., Biodegradable dextran nanogels for RNAi

• 14th international symposium on recent advances in drug delivery systems, Salt Lake City, UT


Naeye B., Raemdonck K., Demeester J., and De Smedt S.C., Pegylation of biodegradable dextran

nanogels for controlled siRNA release

• Nanomedicine, Sant Feliu de Guixols (Spain), September 20th‐24th 2008

Naeye B., Raemdonck K., Demeester J. and De Smedt S.C., Optimization of biodegradable dextran

nanogels for controlled siRNA delivery

• MediTrans 2nd annual meeting, Rehovot (Israel), March 26th‐29th 2009.

SCIENTIFIC COURSES

Training course ‐ Particle characterization & Innovation (organized by Analis), Sint‐Martens‐Latem

(Belgium), September 13th, 2007.

Flow cytometry course – learning and knowledge, BD Biosciences, Erembodegem (Belgium), February

10th, 2009.

Biogazelle course on qPCR experiment design and data‐analysis, Vlerick Business Center, Ghent

(Belgium), March 5th‐6th, 2009.

231

232

| DANKWOORD

DANKWOORD | ACKNOWLEDGMENT

Ik ga proberen het kort te houden. Proberen dus…

Eerst en vooral wil ik even melden dat ik een dankwoord beschouw als een nawoord, dat je hoort te

schrijven met een goed en volledig overzicht van (en inzicht in) de voorbije jaren en wie daar

allemaal een voorname rol in heeft gespeeld. Vandaar dus ook de keuze om het dankwoord

desgevallend een plaats te geven op hetgeen uiteindelijk pagina 233 is geworden. Desalniettemin

moet ik iedereen gelijk geven die in zijn of haar dankwoord vermeldt dat het niet vanzelfsprekend is

een dergelijk stukje tekst op papier te zetten. “Schrijven is schrappen”, vertrouwde Stefaan mij al

allitererend toe in de periode dat ik op het FFW het schrijven van mijn masterthesis aanvatte, en

deze stelling geldt ongetwijfeld ook voor een dankwoord, zo weet ik nu.

Een eerste woord van dank ben ik uiteraard verschuldigd aan Stefaan & Jo, promotoren van het

proefschrift dat u op dit punt volledig gelezen zou moeten hebben. Een wetenschappelijk labo steunt

grotendeels op een goed uitgeruste (infra)structuur die gevoed wordt vanuit gezonde financiële

fundamenten. Als directeur van het labo vervult Jo hierin een belangrijke en creatieve rol, waarvoor

onzer appreciatie! Deeltijds wetenschappelijk bezieler en deeltijds filosoof vat ongeveer het profiel

van Stefaan samen. Bedankt voor de professionele steun, motivatie en de extra dosis zelfvertrouwen

wanneer dat nodig was. Eveneens bedankt voor de vele anekdotes waarin je zelf een hoofdrol speelt

en die iedereen van ons nog regelmatig met plezier doorvertelt (carport en dergelijke, u weet het wel

;)). Sinds kort horen in deze paragraaf ook Niek en Kevin Br. thuis. Ik wens jullie beiden nog veel

succes met het verder uitbouwen van een wetenschappelijke carrière (dit geldt voor iedereen

trouwens…).

De laatste jaren is het labo exponentieel gegroeid. Dit vertaalde zich in een grote ‘voorraad’ aan

collega’s waaruit kon geput worden in geval van wetenschappelijke nood. Een welgekomen

extralegaal voordeel aangezien een doctoraat vervolmaken ook grotendeels steunt op advies van en

discussies met scientific peers. Bedankt allemaal (Bart L., Farzaneh, Roos, Lies, Tinneke, Barbara,

Stefaan D., Bruno DG, Joanna, Katrien, Ine, Sofie T., Elien, Marie‐Luce, Geertrui, Oliwia, Chaobo,

Broes, Hendrik, Dries, Nathalie, Kevin Bu. en Bart G.)! Er arriveerde zowaar een nanogel‐partner‐in‐

science (NPiS) onder de vorm van Broes. Met ons tweeën gaan we nu resoluut voor die Nature

publicatie, zeker nu we uit goede bron hebben vernomen dat die garant staat voor een fles

champagne. Het labo Algemene Biochemie en Fysische Farmacie is op meerdere vlakken uniek. Naar

mijn bescheiden mening vormen we een hechte groep gemotiveerde onderzoekers (beetje reclame

maken mag wel) die op en naast het werk zeer goed met elkaar kunnen opschieten. Moge het GRGN

Social Committee nog vele activiteiten op de kalender plaatsen!

233

DANKWOORD |

234

In ieder dankwoord wordt (en dit is meer dan terecht!) ook een paragraaf gereserveerd voor

Katharine en Bruno VdB, de beste secretariaatstandem die een onderzoekslabo zich kan wensen. Een

kort literatuuronderzoek in menig doctoraatsdankwoord leverde lyrische lofzangen op met als

voornaamste kernwoorden ‘werkijver’, ‘onmisbaar’, ‘glimlach’ en ‘grapjes’. Ik sluit mij hier volledig bij

aan. Misschien nog even opmerken dat Bruno VdB zich bovendien nuttig maakte als mijn

persoonlijke ‘ping‐pong’ coach. Volgend jaar gaan we voor de sportieve doorbraak!

Naast het wetenschappelijk werk (ping‐pongen werd onder de middag gepland red.) was er nog tijd

voor ontspanning na de uren en in de weekends onder de vorm van tennis, mountainbike, menig

cafébezoek, BBQ, wijndegustaties, festivals, weekendjes Ardennen, limericks en vooral veel gezellige

babbels! Special thanks to Bart W., Klaar, Tim, Bart L., Jo, Ine, Philippe, Sofie S., Olivier, Patricia,

Katrien, Björn, Birger, Line, Wim, Eva, Lieven, Julie, Griet, Nathalie, Steven, Annelies, Ruben,

Kathleen, Bert… .

Tot slot een overweldigende merci aan ma en pa, my brother and sister‐in‐law, ons meter en mijn

Meetjeslandse schoonfamilie voor vanalles en eigenlijk teveel om op te noemen (wij weten het!).

Details kan u lezen in mijn autobiografie die misschien ooit zal verschijnen. Iemand die hierin ook een

niet te onderschatten rol zal spelen is m’n vraaiken, Evelien, mijn steun en toeverlaat, die nu

ongetwijfeld ook tevreden zal zijn dat het doctoraat achter de rug is… ;). Dikke kus.

Oh ja, ook een woordje van dank aan iedereen die ik ben vergeten…

Gent, oktober 2009

Koen Raemdonck

| DANKWOORD

A PhD is a real way of life

That takes you beyond nine to five

But the effort’s in vain

If you try to explain

It’s ‘no pain, no gain’ to your wife

:)

(A PhD limerick© by Koen Raemdonck)

235

DANKWOORD |

236

Documents

Maintaining the silence: reflections on long-term RNAi