Neuron-Targeted Nanoparticle for siRNADelivery to Traumatic Brain InjuriesEster J. Kwon,† Matthew Skalak,† Riana Lo Bu,† and Sangeeta N. Bhatia*,†,‡,§,∥,⊥,#

†Koch Institute for Integrative Cancer Research, ‡Harvard−MIT Division of Health Sciences and Technology, Institute for MedicalEngineering and Science, and §Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge,Massachusetts 02139, United States∥Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts 02115, United States⊥Broad Institute of Massachusetts Institute of Technology and Harvard, Cambridge, Massachusetts 02139, United States#Howard Hughes Medical Institute, Cambridge, Massachusetts 02139, United States

*S Supporting Information

ABSTRACT: Traumatic brain injuries (TBIs) affect 2.5 millionAmericans per year, and survivors of TBI can develop long-termimpairments in physical, cognitive, and psychosocial functions.Currently, there are no treatments available to stop the long-termeffects of TBI. Although the primary injury can only be prevented,there is an opportunity for intervention during the secondaryinjury, which persists over the course of hours to years after theinitial injury. One promising strategy is to modulate destructivepathways using nucleic acid therapeutics, which can downregulate“undruggable” targets considered difficult to inhibit with small molecules; however, the delivery of these materials to thecentral nervous system is challenging. We engineered a neuron-targeting nanoparticle which can mediate intracellulartrafficking of siRNA cargo and achieve silencing of mRNA and protein levels in cultured cells. We hypothesized that, soonafter an injury, nanoparticles in the bloodstream may be able to infiltrate brain tissue in the vicinity of areas with acompromised blood brain barrier (BBB). We find that, when administered systemically into animals with brain injuries,neuron-targeted nanoparticles can accumulate into the tissue adjacent to the injured site and downregulate a therapeuticcandidate.

KEYWORDS: traumatic brain injury, nucleic acid delivery, nanoparticle, peptide, neuron targeting

Traumatic brain injury (TBI) is a leading cause of deathand disability for young people, and it is estimated that5.3 million people in the U.S. live with a TBI-relateddisability.1 Survivors of TBI can develop life-long impairment inphysical, cognitive, and psychosocial functions. Currently,clinical care of TBI patients is palliative due to the lack oftherapeutics available to prevent long-term effects, indicatingthere is a need to develop new technology. In the sequence ofevents after an injury, the primary injury can only be prevented;however, there is an opportunity for intervention during thesecondary injury. Secondary injuries can develop over thecourse of hours to years after the initial insult, during which acascade of destructive pathways (e.g., glutamate excitotoxicityand inflammation) ultimately leads to cell death of neurons andglia, either as a direct result of the injury or as collateraldamage.2,3 These destructive pathways can be chronicallyactivated, leading to long-term deterioration of cell populationsin the brain.4 As more molecular responses to dysregulatedhomeostasis in the brain are uncovered, one promising strategyis to mitigate destructive pathways or promote protectivepathways using nucleic acid therapeutics. In particular, delivery

of siRNA can downregulate gene candidates which do not havesmall-molecule antagonists in order to exert neuroprotectiveproperties or be used as a tool to screen for potentialtherapeutic targets in animal models before time- and cost-intensive small-molecule development is initiated. In additionto the delivery of siRNA, an increasing number of therapeuticmicroRNAs have been identified for intervention in braininjuries.5−7

Traversing the intact blood brain barrier (BBB) in a healthybrain remains a significant hurdle for macromoleculeinterventions designed to function in the central nervoussystem.8 Beyond entry into the tissue, siRNA cargos, inparticular, must also navigate into cell types of interest andescape from intracellular vesicles into the cytoplasm, wherethey can engage the RNA-induced silencing complex (RISC)machinery. Nanoparticles offer a solution to achieve traffickingof systemically administered nucleic acids, as they can protect

Received: June 10, 2016Accepted: July 18, 2016Published: July 18, 2016

Artic

lewww.acsnano.org

© 2016 American Chemical Society 7926 DOI: 10.1021/acsnano.6b03858ACS Nano 2016, 10, 7926−7933

www.acsnano.orghttp://dx.doi.org/10.1021/acsnano.6b03858

the sensitive cargo in the vasculature and encode functions toovercome bottlenecks in both the extracellular and intracellulartransport pathway. Viruses such as herpes simplex virus9 andrabies virus10 have evolved mechanisms to bypass the tightlyregulated layer of endothelial cells that separates the brain fromthe vasculature by hijacking retrograde transport machinery toinfiltrate the brain. Even without this BBB-bypassing activity, inthe context of traumatic brain injury, resulting physical breachesin vasculature may offer a window during which small-moleculedrugs can diffuse through the otherwise difficult-to-traverseBBB.11 This strategy can also apply to the transport ofnanoparticulate materials.12,13 Our group has previouslyengineered a peptide-based nanoparticle system for siRNAdelivery that is both targeted and can mediate endosomalescape after cellular internalization using components inspiredby biological systems such as viruses.14,15 This nanoparticlesystem was built by complexing siRNA with a peptide thatincorporates in tandem two components: a membrane-interactive peptide and a targeting peptide. The application ofa tumor-targeting version of this technology decreased tumorburden when applied to a model of orthotopic disseminatedovarian cancer.14 We hypothesized that these tandem peptidenanoparticles could be adapted for use in delivery of nucleicacid cargos to impact secondary cellular damage that arises inareas surrounded by compromised vasculature following TBIs.After a TBI, neurons are a particularly vulnerable population,

and their loss is thought to be responsible for the manynegative repercussions found in TBI survivors.16 Nanoparticlescan be encoded with the ability to target specific cell types17−20

and, once in these cells, to traffic cargo to the cytosol, which isan important consideration for siRNA delivery to engage theRISC machinery. In order to target therapeutic nucleic acidcargos to neurons, we re-engineered our tumor-targetedpeptide-based delivery system to include a neuron-targetingpeptide that was taken from the coat protein of rabies virus.21,22

Our re-engineered tandem peptide nanoparticle system wastargeted to neurons in a modular design and retained its abilityto form sub-100 nm particles while attaining the ability todownregulate genes in neuronal cells in culture. In a mousemodel of penetrating TBI, we delineated the transient timewindow during which nanoparticles administered intravenouslyare able to infiltrate into the injury site and translocate intoneurons. Furthermore, nanoparticles carrying a potentialtherapeutic, siRNA against caspase 3, achieved downregulationof protein levels in the injured brain.

RESULTS AND DISCUSSIONOur goal was to build a nanoparticle system that could infiltratebrain tissue by leveraging the transient access to the brain tissuethat arises after a TBI23 to deliver siRNA cargo into neurons(Figure 1A). In order to target neuronal populations, a tandempeptide nanoparticle system developed in our lab was re-engineered to incorporate the targeting peptide from rabiesvirus, RVG.21 The optimal configuration between targetingpeptide and an intracellular trafficking peptide, transportan(TP), was investigated by synthesizing either an N-terminaltransportan (TP-RVG) or an N-terminal targeting peptide(RVG-TP; Figure 1A). These cationic peptides were used todetermine amounts of peptide required to complex siRNA intonanoparticles. The hydrodynamic diameter of nanoparticles wasmeasured after formulation at 20:1, 10:1, and 5:1 peptide/siRNA ratios (charge ratios of 2.86, 1.43, and 0.71,respectively). Regardless of which peptide was at the N-

terminus of the tandem peptide, particles formed at the sameratio resulted in the same hydrodynamic diameters (Figure 1B),which was expected because the physicochemical properties ofthe peptides remain the same in both orientations. Particlesincreased in size at a peptide/siRNA ratio of 10:1, likely due to

Figure 1. Optimization of neuron-targeted nanoparticles. (A)Schematic of neuron-targeted nanoparticle formulation withtandem peptides and siRNA and entry into injured tissue. (B)Hydrodynamic diameter of nanoparticles formulated at variouspeptide/siRNA ratios. Inset: TEM image of TP-RVG particleformed at 20 peptide/siRNA ratio. Scale bar represents 100 nm.(C) Knockdown activity of nanoparticles formulated at variouspeptide/siRNA ratios analyzed by RT-qPCR. Error bars representSD; one-way ANOVA *p < 0.05, **p < 0.01, ***p < 0.001.

ACS Nano Article

DOI: 10.1021/acsnano.6b03858ACS Nano 2016, 10, 7926−7933

7927

http://dx.doi.org/10.1021/acsnano.6b03858

the phenomenon of colloid aggregation that occurs with neutralsurface charges.24 Next, the particles were evaluated for theirability to mediate gene silencing of a model endogenous gene,peptidylprolyl isomerase B (Ppib), in the mouse neuroblastomacell line, Neuro-2a (Figure 1C). Despite the two peptidesforming particles at the same ratio, the formulation thatincorporated transportan at the N-terminus had significantlyimproved gene silencing activity compared to when this peptidewas located at the C-terminus.A series of experiments were designed to determine why

peptide with N-terminally located transportan and C-terminallylocated targeting peptide was more efficient at deliveringsiRNA. We first tested the hypothesis that this configuration oftransportan and targeting peptide led to a greater binding anduptake of material by cells. Cells were incubated with thedesignated concentrations of nanocomplexes to determine theamount of material associated with cells when internalizationpathways are active (37 °C, binding and uptake; Figure 2A) ornot active (4 °C, binding only; Figure 2B). The amount ofnanoparticle association was measured using flow cytometry bydetecting the fluorescence of 5-carboxyfluorescein (FAM)-labeled peptide. We observed elevated cell association ofnanocomplexes with N-terminal-targeting peptides at bothtemperatures, indicating that increased particle binding anduptake was not likely to be responsible for the improvedsilencing efficiency mediated by N-terminal transportanconformation. Next, we tested whether the orientation oftransportan influences nanoparticle escape from intracellularvesicles after internalization, by performing the assay in thepresence or absence of the small-molecule chloroquine. It isknown that chloroquine disrupts endosomal acidification andcan therefore aid in escape of nanoparticles trapped inendocytic vesicles after internalization.25 The addition ofchloroquine resulted in similar levels of gene silencing forboth peptide configurations with no statistical differencebetween the two formulations at matched peptide/siRNAratios, as determined by one-way ANOVA analysis (Figure 2C),indicating that the N-terminal position of transportan plays arole in cytosolic delivery. The importance of N- or C-terminalorientation has been observed for other membrane-activepeptides, such as melittin,26 presumably because membraneinteraction is dictated by certain residues, and naturallyoccurring peptides are found in specific orientations of thenative protein. We also noted that, in previous work, RVG waslocated on the N-terminus,21 but the activity of the intracellular

trafficking peptide was the driving factor that dictated ∼50% ofthe gene silencing activity in our targeted delivery system, asestimated based on our chloroquine transfection study.Using the optimal configuration of transportan and targeting

peptide, the importance of the targeting ligand was evaluated ina neuronal cell line, Neuro-2a. Neuro-2a cells express thenicotinic acetylcholine receptor (Figure 3A), a receptor that isthought to be primarily responsible for rabies virus glycoproteinbinding.27 Nanoparticles were targeted with either RVG or acontrol peptide (MAT), a sequence taken from a differentportion of the same rabies virus glycoprotein as RVG, and

Figure 2. Investigation of intracellular trafficking mechanism of tandem peptide nanoparticles. Flow cytometric analysis of Neuro-2a cellsincubated with fluorescently labeled nanoparticles at (A) 4 °C and (B) 37 °C. (C) Knockdown activity of nanoparticles in the presence ofchloroquine analyzed by RT-qPCR. There was no statistical difference between knockdown with or without chloroquine at matched peptide/siRNA ratios as determined by analysis with one-way ANOVA.

Figure 3. RVG peptide targeting of nanoparticles. (A) Staining ofNeuro-2a cells for expression of α-acetylcholine receptor (green)counterstained with Hoechst (blue). Scale bar represents 20 μm.(B) Knockdown mediated by RVG- and MAT-targeted nano-particles in Neuro-2a cells analyzed by RT-qPCR. Error barsrepresent SD; one-way ANOVA *p < 0.05, ***p < 0.001. (C)Accumulation of nanoparticles made with RVG or MAT peptide(green) and siRNA (red) in Neuro-2a cells after 2 h. Scale barrepresents 50 μm.

ACS Nano Article

DOI: 10.1021/acsnano.6b03858ACS Nano 2016, 10, 7926−7933

7928

http://dx.doi.org/10.1021/acsnano.6b03858

applied to Neuro-2a cells. MAT-targeted nanoparticlesmediated significantly less knockdown of Pbib, comparedwith RVG-targeted nanoparticles (Figure 3B). When uptake ofnanoparticles was evaluated using fluorescence microscopy bytracking FAM-labeled peptides and Dylight-647-labeled siRNA,significantly less uptake of both peptide and siRNA was seenwhen particles were targeted with MAT compared with RVG(Figure 3C).The optimized targeted nanoparticle formulation was then

applied to deliver siRNA against a potential therapeutic gene. InTBIs, the secondary injuries that arise in response to theprimary event include dysregulation of the BBB, edema, ectopicrelease of excitatory neurotransmitters, and inflammation.4

These indirect effects can occur chronically and causewidespread death of neurons in the brain over the course ofhours to years after the initial injury. To prevent the apoptosisof neurons caused by the secondary injury, we delivered siRNAagainst caspase 3. Caspase 3 is known to be a main player inapoptosis after TBIs,28 and small-molecule and peptidetherapies targeted to caspase 3 have been explored.29,30 Wefirst validated siRNA sequences against caspase 3 and foundthat levels of mRNA after caspase 3 knockdown were decreasedto approximately 50% after 48 h (Figure 4A). Since the

mediator of damage is the CASP3 protein, we then evaluatedthe ability of our targeted nanocomplex to suppress CASP3protein levels by Western blot (Figure 4B). We found a moreextensive decrease in protein levels compared with mRNAlevels, with over 90% of the CASP3 protein levels decreasedafter siRNA treatment. To simulate neurons in apoptotic states,we probed the levels of CASP3 protein in the presence of thechemotherapeutic staurosporine, a well-documented inducer ofapoptosis. Activation of the apoptotic cascade can be inferredbased on the presence of the activated fragments of cleavedCASP3 (17 and 19 kDa) in Neuro-2a cells treated withstaurosporine (Figure 4B). When siCASP3 was delivered, both

the zymogen and activated forms of CASP3 were significantlydecreased.In the presented platform, neuron-targeted nanoparticles

were designed to leverage the transiently compromisedvasculature after a traumatic brain injury to gain access tobrain tissue. Because this breach in the BBB offers onlytemporary access to the brain tissue, a nanoparticle treatmentwould be best administered within 24 h after injury. To enablea formulation that could be deployable in the field, or whereverpatients experience trauma, we tested the stability of our RVG-targeted nanoparticles when stored in ambient conditions.Nanoparticles were able to maintain their size (Figure 4C) andtheir knockdown activity (Figure 4D) after 24 h of storage atroom temperature. Furthermore, nanoparticles stored for up to7 days maintain their size and likely their activity because thecargo should be protected in nanoparticle form (SI Figure 1).With the basic criteria of our platform credentialed in vitro,

we applied our siRNA delivery carrier to an animal model ofTBI. We modeled the TBI to mimic penetrating injuries using aseries of needle stick wounds applied in a controlled manner viaa stereotaxic apparatus (Figure 5A). The wounds were inflictedon the cortex and hippocampus of mice on only onehemisphere of the brain, whereas the contralateral hemisphereof the brain was not directly injured to serve as a control. Oneimportant caveat is that the uninjured contralateral tissue maybe indirectly affected by signaling pathways that are activated inresponse to injury; for example, changes in vascularpermeability at the whole brain level have been observed inresponse to injury, although the mechanisms are currentlyunknown.12,23 The siRNA/peptide nanocomplexes werelabeled with a far-red dye, VivoTag 750, to enable sensitivedetection of material in animal tissue. Sham injuries wereperformed by removing the skull but not inducing shear injuriesin the brain tissue, in order to discriminate surgery-mediatedparticle accumulation from that attributable to the deeper shearinjuries (SI Figure 2). When nanoparticles were administeredinto the circulation 5 min after the sham injury, we observed amodest signal increase concentrated in a ring where the skullwas removed, likely caused by damaged vasculature on the duramade by the drill. There was no evidence of peptideaccumulation in the deeper brain tissue in sham injuredanimals. In order to quantitatively observe the kinetics ofvascular permeability to nanoparticles, nanoparticles weredelivered intravenously 5 min, 1, 3, 6, and 24 h after injury,and brains were collected 1 h after each delivery to assessnanoparticle infiltration (Figure 5B,C). Notably, the contrala-teral side of the injured brain displayed levels of nanoparticleaccumulation similar to that observed in completely uninjuredmice, supporting the interpretation that material uptake islargely driven by the conditions at the immediate injury site atleast in the early response. We do note a slight increase inaccumulation on the contralateral side of the brain at 1 h afterinjury. Although this result was not statistically significant, itmay be that the biological response to injury increases vascularpermeability beyond the local site of the damaged tissue.12 Wealso noted that free siRNA accumulated into injured tissue togreater levels than observed with complexed siRNA, likely dueto the small size and negative charge of free siRNA whichprevents binding and sequestration into off-target tissue (SIFigure 3). However, free unmodified siRNA has negligiblesilencing activity (SI Figure 4) and is vulnerable to degradationwhen unprotected in the vasculature. The differentialdistribution between free siRNA and siRNA−nanoparticles

Figure 4. Knockdown of therapeutic protein, CASP3. RVG-targetednanoparticle-mediated knockdown of caspase 3 (A) mRNAanalyzed by RT-qPCR and (B) protein analyzed by Westernblotting compared to α-tubulin loading control (TUBA). Stabilityof (C) hydrodynamic diameter and (D) silencing activity of RVG-targeted nanoparticle after 24 h incubation at room temperature.Error bars represent SD; one-way ANOVA *p < 0.05, ***p < 0.001.

ACS Nano Article

DOI: 10.1021/acsnano.6b03858ACS Nano 2016, 10, 7926−7933

7929

http://pubs.acs.org/doi/suppl/10.1021/acsnano.6b03858/suppl_file/nn6b03858_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acsnano.6b03858/suppl_file/nn6b03858_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acsnano.6b03858/suppl_file/nn6b03858_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acsnano.6b03858/suppl_file/nn6b03858_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acsnano.6b03858/suppl_file/nn6b03858_si_001.pdfhttp://dx.doi.org/10.1021/acsnano.6b03858

also suggests that the nanoparticles are stable in vivo, such thatthe majority of siRNA remains complexed with peptide in thevasculature. On the injured hemisphere of the brain, weobserved nanoparticle accumulation at the injury site up to 6 h,followed by a significant decrease at 24 h. This observation isconsistent with other models of brain injury, where infiltrationof macromolecules into the brain tissue from the blood ismaximal at 6 h after the injury, likely mitigated by progressionof edema that increases local pressure.31

To more specifically localize the delivered material within thebrain tissue, brains were sectioned and stained for peptide usingan anti-fluorescein antibody combined with staining for gliausing the marker glial fibrillary acid protein (GFAP; Figure5D), microglia using the marker ionized calcium bindingadaptor molecule 1 (Iba1; Figure 5E), or neurons using themarker NeuN (Figure 5F). There was no detectable peptidematerial observed in glia and microglia (Figure 5D,E). Incontrast, inspecting the neuronal population in the cortex nearsites of the injury revealed that, although not all neurons at theinjured site are associated with detectable levels of peptide, themajority of peptide signal localized with neurons (Figure 5F).The reputed receptor for rabies virus, and presumably RVGpeptide, is the nicotinic acetylcholine receptor.27 We stainedsections with a fluorophore-labeled α-bungarotoxin, a high

affinity ligand for the nicotinic acetylcholine receptor, andfound that while some colocalization of peptide with α-bungarotoxin positive cells was evident, we also observedexamples of peptide localized in cells not positive for α-bungarotoxin (SI Figure 5). Possible explanations for thisobservation are alternative protein complexes thought to have arole in rabies virus binding and entry,32 and the peptide fromthe rabies virus used in our particles may not be sufficient toconfer the same specificity as the full protein. In order toquantitatively support the microscopy images, we countedGFAP+, Iba1+, and NeuN+ cells that costained peptide+ cellsobserved in several fields of view from brain sections of threedifferent mice (Figure 5G). The peptide+NeuN+ doublepositive cells were 80% of the total peptide+ population,whereas peptide+GFAP+ and peptide+Iba1+ double positivecells were less than 2% each of the total peptide+ population,supporting that the majority of material was trafficked toneuronal cells.Based on the observations that neuron-targeted nano-

complexes accumulated at the site of injury and deliveredmaterial was taken up by neurons, we administered nano-complexes carrying a therapeutic gene and evaluated theirability to mediate protein knockdown in regions proximal tothe injury site. We employed the same penetrating injury model

Figure 5. Accumulation of RVG-targeted nanoparticles in the injured brain. (A) Schematic of the simulated TBI in mice. (B) Accumulation offluorescently labeled particles in the injured hemisphere (right) when administered intravenously 5 min, 1, 3, 6, and 24 h after injury.Uninjured brain administered particles serves as a control. (C) Quantification of nanoparticle accumulation in the contralateral (C) versusinjured (I) hemisphere. Dotted line represents an uninjured (U) animal injected with nanoparticles. Error bars represent SEM; one-wayANOVA, n = 3 *p < 0.05, **p < 0.01. Sections of 10 μm taken from brain administered nanoparticles 5 min after injury and stained for nuclei(blue), FAM-labeled peptide (green), and (D) astrocytes (red; GFAP), (E) microglia (red; Iba1), or (F) neurons (red; NeuN). White dashedline marks the boundary of shear injury; the box indicates magnified inset (bottom), and arrows denote cells which stain positively for bothNeuN and peptide. Scale bar represents 100 μm. (G) Quantification of total counts and percentages of NeuN, Iba1, and GFAP positivepopulations from images.

ACS Nano Article

DOI: 10.1021/acsnano.6b03858ACS Nano 2016, 10, 7926−7933

7930

http://pubs.acs.org/doi/suppl/10.1021/acsnano.6b03858/suppl_file/nn6b03858_si_001.pdfhttp://dx.doi.org/10.1021/acsnano.6b03858

of TBI used above, and nanoparticles carrying either controlsiRNA or siRNA against CASP3 were delivered intravenously 5min after the injury. Three days post-injury, brains weredissected to isolate the injured area of the cortex, as well as thecorresponding tissue region on the contralateral uninjuredhemisphere. The resulting tissue samples were processed foranalysis of protein expression by Western blotting. A decreasein CASP3 protein levels was observed in the injuredhemisphere when siCASP3 was administered, relative to thedelivery of siCtrl (Figure 6A). Quantification of these results

showed ∼80% knockdown of CASP3 in the damaged region ofthe injured hemisphere (Figure 6B). By contrast, in thecontralateral hemisphere, no knockdown was observedfollowing either nanoparticle treatment (Figure 6C,D). Lackof silencing in the contralateral hemisphere was consistent withthe limited particle accumulation observed in this side of thebrain (Figure 5C).

CONCLUSIONSTBI remains a major cause of disability with no therapeutictreatments available. The delivery of nanoparticles into thecentral nervous system for nucleic acid delivery has remained achallenge due to the poor permeability of the BBB tonanoparticles and the requirement to localize nucleic acidcargos to the cytosol or nucleus of cells. Existing approacheshave focused on local delivery of material, either into thecerebrospinal fluid reservoirs33,34 or direct administration intothe brain tissue.35,36 Other approaches include the use of smallmolecules37 or ultrasound38,39 to alter the BBB to inducetransient permeabilization of the BBB for nanoparticletransport. In the presented work, we envisioned a therapeuticthat could take advantage of the BBB breach arisingimmediately after a penetrating injury and therefore could beadministered in the field via an intravenous delivery route. Wedesigned a targeted nanoparticle inspired by previoustechnology from our lab developed for cancer therapeuticswhich utilizes an intracellular trafficking peptide and a targetingpeptide joined in tandem to self-assemble with siRNA intonanometer-sized structures. Re-engineering these nanoparticlesto target neurons using a peptide ligand produced a systemcapable of mediating silencing in cultured cells that express theproposed target receptor. Intravenous application of these

nanoparticles to an in vivo model of TBI transiently allowssignificant accumulation of nanoparticles if administered withina 6 h window after the onset of injury. Observations in theliterature suggest that this window is limited in the short termby biological cascades that are produced by the injury such asintravascular coagulation and edema.31,40 Furthermore, targetednanoparticles were found to accumulate in neurons and wereable to silence a candidate therapeutic gene in the regionproximal to the injury. The RVG-targeting ligand has been usedpreviously to mediate transport of nanostructured materialsacross an intact BBB when decorated on electrostaticcomplexed nanoparticles21,37 and exosomes.22,41 Anotheraspect of central nervous system delivery that is being exploredis the delivery of targeted materials in disease states,42,43 wherethere may be altered BBB permeability. In this work, we soughtto characterize nanoparticle infiltration from the systemicvasculature when administered in the context of a TBI and,furthermore, the ability to deliver nucleic acid cargo to targetedcell populations. We observed that there was significantlyincreased accumulation of nanoparticles in injured brain tissuecompared to uninjured brains after intravenous delivery (Figure5C), and in this context, a targeting ligand could achieveneuron-specific uptake. Future studies could explore long-termeffects of therapeutic siRNA delivery in animals, such as lookingat retention of neuronal populations and performance inbehavioral tasks. Altering vascularization to coordinate nano-particle accumulation has been harnessed in the context oftumor delivery in our lab and others,43−45 but this strategy hasnot been explored extensively in the context of the centralnervous system, to date. Because our tandem peptide system ismodular, siRNA against new targets can be rapidly synthesizedand incorporated into nanoparticles to mitigate long-termeffects of TBI.

METHODSCell Culture. Neuro-2a cells were purchased from the American

Type Culture Collection and maintained in EMEM supplementedwith 10% FBS.

Nanoparticle Complexation. The appropriate molar ratio ofpeptide/siRNA was formulated into nanoparticles by adding peptidesolutions to equal volume of siRNA in water and mixing rapidly. TP-RVG (GGWTLNSAGYLLGKINLKALAALAKKILGGK(FAM)GGY-T IWMPENPRPGTPCDIFTNSRGKRASNG) , RVG-TP(YTIWMPENPRPGTPCDIFTNSRGKRASNGGGK (FAM)GGGGWTLNSAGYLLGKINLKALAALAKKIL), and TP-MAT(GGWTLNSAGYLLGKINLKALAALAKKILGGK (FAM)GGMNLLRKIVKNRRDEDTQKSSPASAPLDDG) were synthesizedby CPC Scientific with N-terminal myristic acid. siRNA wassynthesized by Dharmacon (siCtrl = CUUACGCUGAGUACUUCGA(sequence against firefly luciferase), siPPIB 1 = CAAGUUC-CAUCGUGUCAUC, siPPIB 2 = GAAAGAGCAUCUAUGGUGA,CASP3 1 = GGAUAGUGUUUCUAAGGAA, CASP3 2 =CGCACAAGCUAGAAUUUAU).

Dynamic Light Scattering and Transmission ElectronMicroscopy. Nanoparticle solutions were incubated for 10 minafter formulation. Hydrodynamic diameters were measured in aMalvern Zetasizer in water. Particles were deposited on a 200 meshFormvar/carbon mesh (Ted Pella Inc.) and dried. Particles wereimaged on a JEOL 2100 FEG TEM.

Transfection. Neuro-2a cells were plated at 80K cells per well in24-well plates coated with 10 μg/mL poly-D-lysine 24 h beforetransfection. Nanocomplexes were added to cells in OptiMEM at afinal concentration of 100 nM, and 4 h after addition, the medium wasreplaced with full culture medium. Cells were collected fordownstream RT-qPCR or Western blot analysis 48 h after transfection.For relative protein quantification of Western blots, films were

Figure 6. In vivo knockdown of CASP3 in a mouse model oftraumatic brain injury. Western blot of lysates and quantification ofblots from the (A,B) injured hemisphere or the (C,D) contralateralhemisphere. Error bars represent SEM; Student’s t test **p < 0.01,ns = not significant.

ACS Nano Article

DOI: 10.1021/acsnano.6b03858ACS Nano 2016, 10, 7926−7933

7931

http://dx.doi.org/10.1021/acsnano.6b03858

analyzed using the Gels tool in ImageJ by calculating the area underthe curve of sample lanes converted to 2D plots. For transfectionsdone in the presence of chloroquine, nanoparticles were co-incubatedwith 100 nM chloroquine for 4 h before the medium was changed.Cell Association Studies. Cells were co-incubated with nano-

particles with the designated final peptide concentrations as describedin the transfection. For 4 °C association studies, plates were incubatedon ice for 30 min before the addition of nanoparticles. Cells wereharvested using trypsin and analyzed for positive FAM fluorescence onthe peptides. For microscopic examination, Dylight-647 siRNA wasused to make complexes. After 4 h of incubation, cells were fixed,counterstained with Hoechst (Sigma-Aldrich), and visualized on aNikon Eclipse Ti.Injury Model. All animal protocols were done in accordance with

the Massachusetts Institute of Technology’s Institutional Animal Careand Use Committee. Six−eight week old C57BL/6 mice wereobtained from Taconic and were fitted into a stereotaxic device. A 5mm diameter portion of the skull was removed, and a 19 gauge needlewas used to make a series of shear injuries 3 mm deep and 1 mm apart.For biodistribution, 1 nmol of siRNA labeled with VivoTag 750 on the5′ end of the sense strand was used. At the times designated, micewere anesthetized with isoflurane and perfused with 10% formalin, andthe brains were imaged on a LI-COR Odyssey using the 800 channel.For knockdown studies, 2 nmol of siRNA was administered via the tailvein after the injury in a solution of 5% dextrose 5 min following theinjury. Mice were sacrificed via CO2 asphyxiation 72 h afteradministration of particles, and fresh tissue was collected andimmediately frozen for protein analysis. The injured cortex wasdissected along with the uninjured contralateral tissue for Westernblotting. Western blots were probed with antibodies against caspase 3(Santa Cruz Biotechnology, Inc., 1:500), cleaved caspase 3 (CellSignaling Technology, 1:1000), and α-tubulin (Invitrogen, 1:1000).Staining of Brain Sections and Quantification. Mice were

perfused with fixative, and brains were fixed for an additional 16 h.Brains were equilibrated into 30% sucrose solution, embedded intoOptimum Cutting Temperature compound (Sakura Finetek), frozenon dry ice, and cut into 10 μm sections. Sections were blocked in 10%goat serum, 2% BSA, and 0.1% TritonX-100 in PBS and probed withanti-NeuN (Millipore, 1:800), anti-GFAP (Abcam, 1:500), anti-Iba1(Abcam, 1:500), and antifluorescein (Thermo Fisher, 1:200) antibod-ies. Appropriate fluorophore-conjugated secondary antibodies (Jack-son Immunolabs) were used to visualize primary antibodies. Severalfields of view were imaged using fluorescence microscopy, andcounting of NeuN, Iba1, and GFAP and peptide double positivepopulations was done in ImageJ.

ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acsnano.6b03858.

SI Figures 1-5 (PDF)

AUTHOR INFORMATIONCorresponding Author*E-mail: sbhatia@mit.edu.NotesThe authors declare no competing financial interest.

ACKNOWLEDGMENTSWe thank the Koch Institute Swanson Biotechnology Center(MIT) for assistance with TEM imaging, specifically Dr. DongSoo Yun, and tissue sectioning, specifically Michael Brown andKathleen Cormier. We thank Dr. Heather Fleming (MIT) forcritical reading and editing of the manuscript. This work wassupported in part by the Marie D. & Pierre Casimir-LambertFund, and a Koch Institute Support Grant P30-CA14051 from

the National Cancer Institute (Swanson Biotechnology Center)and a Core Center Grant P30-ES002109 from the NationalInstitute of Environmental Health Sciences. This work was alsosupported by the Defense Advanced Research Projects Agencyunder Cooperative Agreement HR0011-13-2-0017. Thecontent of the information within this document does notnecessarily reflect the position or the policy of the government.E.J.K. acknowledges support from the Ruth L. KirschsteinNational Research Service Award (1F32CA177094-01). S.N.B.is an HHMI Investigator.

REFERENCES(1) Roozenbeek, B.; Maas, A. I.; Menon, D. K. Changing Patterns inthe Epidemiology of Traumatic Brain Injury. Nat. Rev. Neurol. 2013, 9(4), 231−6.(2) Xiong, Y.; Mahmood, A.; Chopp, M. Animal Models ofTraumatic Brain Injury. Nat. Rev. Neurosci. 2013, 14 (2), 128−42.(3) Cernak, I.; Noble-Haeusslein, L. J. Traumatic Brain Injury: AnOverview of Pathobiology with Emphasis on Military Populations. J.Cereb. Blood Flow Metab. 2010, 30 (2), 255−66.(4) Werner, C.; Engelhard, K. Pathophysiology of Traumatic BrainInjury. Br. J. Anaesth. 2007, 99 (1), 4−9.(5) Ge, X. T.; Lei, P.; Wang, H. C.; Zhang, A. L.; Han, Z. L.; Chen,X.; Li, S. H.; Jiang, R. C.; Kang, C. S.; Zhang, J. N. miR-21 Improvesthe Neurological Outcome after Traumatic Brain Injury in Rats. Sci.Rep. 2014, 4, 6718.(6) Bhalala, O. G.; Srikanth, M.; Kessler, J. A. The Emerging Roles ofMicroRNAs in CNS Injuries. Nat. Rev. Neurol. 2013, 9 (6), 328−39.(7) Saugstad, J. A. MicroRNAs as Effectors of Brain Function withRoles in Ischemia and Injury, Neuroprotection, and Neurodegenera-tion. J. Cereb. Blood Flow Metab. 2010, 30 (9), 1564−76.(8) Pardridge, W. M. Drug Transport across the Blood-Brain Barrier.J. Cereb. Blood Flow Metab. 2012, 32 (11), 1959−72.(9) Sodeik, B.; Ebersold, M. W.; Helenius, A. Microtubule-MediatedTransport of Incoming Herpes Simplex Virus 1 Capsids to theNucleus. J. Cell Biol. 1997, 136 (5), 1007−21.(10) Schnell, M. J.; McGettigan, J. P.; Wirblich, C.; Papaneri, A. TheCell Biology of Rabies Virus: Using Stealth to Reach the Brain. Nat.Rev. Microbiol. 2010, 8 (1), 51−61.(11) Shlosberg, D.; Benifla, M.; Kaufer, D.; Friedman, A. Blood-BrainBarrier Breakdown as a Therapeutic Target in Traumatic Brain Injury.Nat. Rev. Neurol. 2010, 6 (7), 393−403.(12) Smith, N. M.; Gachulincova, I.; Ho, D.; Bailey, C.; Bartlett, C.A.; Norret, M.; Murphy, J.; Buckley, A.; Rigby, P. J.; House, M. J.; St.Pierre, T.; Fitzgerald, M.; Iyer, K. S.; Dunlop, S. A. An UnexpectedTransient Breakdown of the Blood Brain Barrier Triggers Passage ofLarge Intravenously Administered Nanoparticles. Sci. Rep. 2016, 6,22595.(13) Cruz, L. J.; Stammes, M. A.; Que, I.; van Beek, E. R.; Knol-Blankevoort, V. T.; Snoeks, T. J.; Chan, A.; Kaijzel, E. L.; Lowik, C. W.Effect of PLGA NP Size on Efficiency to Target Traumatic BrainInjury. J. Controlled Release 2016, 223, 31−41.(14) Ren, Y.; Cheung, H. W.; von Maltzhan, G.; Agrawal, A.; Cowley,G. S.; Weir, B. A.; Boehm, J. S.; Tamayo, P.; Karst, A. M.; Liu, J. F.;Hirsch, M. S.; Mesirov, J. P.; Drapkin, R.; Root, D. E.; Lo, J.; Fogal, V.;Ruoslahti, E.; Hahn, W. C.; Bhatia, S. N. Targeted Tumor-PenetratingsiRNA Nanocomplexes for Credentialing the Ovarian CancerOncogene ID4. Sci. Transl. Med. 2012, 4 (147), 147ra112.(15) Ren, Y.; Hauert, S.; Lo, J. H.; Bhatia, S. N. Identification andCharacterization of Receptor-Specific Peptides for siRNA Delivery.ACS Nano 2012, 6 (10), 8620−31.(16) Raghupathi, R. Cell Death Mechanisms Following TraumaticBrain Injury. Brain Pathol. 2004, 14 (2), 215−22.(17) Davis, M. E.; Zuckerman, J. E.; Choi, C. H.; Seligson, D.;Tolcher, A.; Alabi, C. A.; Yen, Y.; Heidel, J. D.; Ribas, A. Evidence ofRNAi in Humans from Systemically Administered siRNA via TargetedNanoparticles. Nature 2010, 464 (7291), 1067−70.

ACS Nano Article

DOI: 10.1021/acsnano.6b03858ACS Nano 2016, 10, 7926−7933

7932

http://pubs.acs.orghttp://pubs.acs.org/doi/abs/10.1021/acsnano.6b03858http://pubs.acs.org/doi/suppl/10.1021/acsnano.6b03858/suppl_file/nn6b03858_si_001.pdfmailto:sbhatia@mit.eduhttp://dx.doi.org/10.1021/acsnano.6b03858

(18) Schroeder, A.; Heller, D. A.; Winslow, M. M.; Dahlman, J. E.;Pratt, G. W.; Langer, R.; Jacks, T.; Anderson, D. G. TreatingMetastatic Cancer with Nanotechnology. Nat. Rev. Cancer 2012, 12(1), 39−50.(19) Choi, C. H.; Alabi, C. A.; Webster, P.; Davis, M. E. Mechanismof Active Targeting in Solid Tumors with Transferrin-Containing GoldNanoparticles. Proc. Natl. Acad. Sci. U. S. A. 2010, 107 (3), 1235−40.(20) Ruoslahti, E.; Bhatia, S. N.; Sailor, M. J. Targeting of Drugs andNanoparticles to Tumors. J. Cell Biol. 2010, 188 (6), 759−68.(21) Kumar, P.; Wu, H.; McBride, J. L.; Jung, K. E.; Kim, M. H.;Davidson, B. L.; Lee, S. K.; Shankar, P.; Manjunath, N. TransvascularDelivery of Small Interfering RNA to the Central Nervous System.Nature 2007, 448 (7149), 39−43.(22) Alvarez-Erviti, L.; Seow, Y.; Yin, H.; Betts, C.; Lakhal, S.; Wood,M. J. Delivery of siRNA to the Mouse Brain by Systemic Injection ofTargeted Exosomes. Nat. Biotechnol. 2011, 29 (4), 341−5.(23) Price, L.; Wilson, C.; Grant, G. Blood-Brain BarrierPathophysiology Following Traumatic Brain Injury. In TranslationalResearch in Traumatic Brain Injury; Laskowitz, D., Grant, G., Eds.;CRC Press: Boca Raton, FL, 2016.(24) El Badawy, A. M.; Luxton, T. P.; Silva, R. G.; Scheckel, K. G.;Suidan, M. T.; Tolaymat, T. M. Impact of Environmental Conditions(pH, Ionic Strength, and Electrolyte Type) on the Surface Charge andAggregation of Silver Nanoparticles Suspensions. Environ. Sci. Technol.2010, 44 (4), 1260−1266.(25) Cheng, J.; Zeidan, R.; Mishra, S.; Liu, A.; Pun, S. H.; Kulkarni, R.P.; Jensen, G. S.; Bellocq, N. C.; Davis, M. E. Structure-FunctionCorrelation of Chloroquine and Analogues as Transgene ExpressionEnhancers in Nonviral Gene Delivery. J. Med. Chem. 2006, 49 (22),6522−31.(26) Boeckle, S.; Wagner, E.; Ogris, M. C- versus N-TerminallyLinked Melittin-Polyethylenimine Conjugates: The Site of LinkageStrongly Influences Activity of DNA Polyplexes. J. Gene Med. 2005, 7(10), 1335−47.(27) Hanham, C. A.; Zhao, F.; Tignor, G. H. Evidence from the Anti-Idiotypic Network That the Acetylcholine Receptor Is a Rabies VirusReceptor. J. Virol 1993, 67 (1), 530−42.(28) Clark, R. S.; Kochanek, P. M.; Watkins, S. C.; Chen, M.; Dixon,C. E.; Seidberg, N. A.; Melick, J.; Loeffert, J. E.; Nathaniel, P. D.; Jin,K. L.; Graham, S. H. Caspase-3 Mediated Neuronal Death afterTraumatic Brain Injury in Rats. J. Neurochem. 2000, 74 (2), 740−53.(29) Cheng, Y.; Deshmukh, M.; D’Costa, A.; Demaro, J. A.; Gidday,J. M.; Shah, A.; Sun, Y.; Jacquin, M. F.; Johnson, E. M.; Holtzman, D.M. Caspase Inhibitor Affords Neuroprotection with DelayedAdministration in a Rat Model of Neonatal Hypoxic-Ischemic BrainInjury. J. Clin. Invest. 1998, 101 (9), 1992−9.(30) Han, B. H.; Xu, D.; Choi, J.; Han, Y.; Xanthoudakis, S.; Roy, S.;Tam, J.; Vaillancourt, J.; Colucci, J.; Siman, R.; Giroux, A.; Robertson,G. S.; Zamboni, R.; Nicholson, D. W.; Holtzman, D. M. Selective,Reversible Caspase-3 Inhibitor Is Neuroprotective and RevealsDistinct Pathways of Cell Death after Neonatal Hypoxic-IschemicBrain Injury. J. Biol. Chem. 2002, 277 (33), 30128−36.(31) Baskaya, M. K.; Rao, A. M.; Dogan, A.; Donaldson, D.;Dempsey, R. J. The Biphasic Opening of the Blood-Brain Barrier in theCortex and Hippocampus after Traumatic Brain Injury in Rats.Neurosci. Lett. 1997, 226 (1), 33−36.(32) Lafon, M. Rabies Virus Receptors. J. NeuroVirol. 2005, 11 (1),82−7.(33) Kwon, E. J.; Lasiene, J.; Jacobson, B. E.; Park, I. K.; Horner, P. J.;Pun, S. H. Targeted Nonviral Delivery Vehicles to Neural ProgenitorCells in the Mouse Subventricular Zone. Biomaterials 2010, 31 (8),2417−24.(34) Bharali, D. J.; Klejbor, I.; Stachowiak, E. K.; Dutta, P.; Roy, I.;Kaur, N.; Bergey, E. J.; Prasad, P. N.; Stachowiak, M. K. OrganicallyModified Silica Nanoparticles: A Nonviral Vector for In Vivo GeneDelivery and Expression in the Brain. Proc. Natl. Acad. Sci. U. S. A.2005, 102 (32), 11539−44.(35) Rungta, R. L.; Choi, H. B.; Lin, P. J.; Ko, R. W.; Ashby, D.; Nair,J.; Manoharan, M.; Cullis, P. R.; Macvicar, B. A. Lipid Nanoparticle

Delivery of siRNA to Silence Neuronal Gene Expression in the Brain.Mol. Ther.–Nucleic Acids 2013, 2, e136.(36) Al-Jamal, K. T.; Gherardini, L.; Bardi, G.; Nunes, A.; Guo, C.;Bussy, C.; Herrero, M. A.; Bianco, A.; Prato, M.; Kostarelos, K.;Pizzorusso, T. Functional Motor Recovery from Brain Ischemic Insultby Carbon Nanotube-Mediated siRNA Silencing. Proc. Natl. Acad. Sci.U. S. A. 2011, 108 (27), 10952−7.(37) Hwang, D. W.; Son, S.; Jang, J.; Youn, H.; Lee, S.; Lee, D.; Lee,Y. S.; Jeong, J. M.; Kim, W. J.; Lee, D. S. A Brain-Targeted RabiesVirus Glycoprotein-Disulfide Linked PEI Nanocarrier for Delivery ofNeurogenic MicroRNA. Biomaterials 2011, 32 (21), 4968−4975.(38) Nance, E.; Timbie, K.; Miller, G. W.; Song, J.; Louttit, C.;Klibanov, A. L.; Shih, T. Y.; Swaminathan, G.; Tamargo, R. J.;Woodworth, G. F.; Hanes, J.; Price, R. J. Non-Invasive Delivery ofStealth, Brain-Penetrating Nanoparticles across the Blood-BrainBarrier Using MRI-Guided Focused Ultrasound. J. Controlled Release2014, 189, 123−32.(39) Mulik, R. S.; Bing, C.; Ladouceur-Wodzak, M.; Munaweera, I.;Chopra, R.; Corbin, I. R. Localized Delivery of Low-DensityLipoprotein Docosahexaenoic Acid Nanoparticles to the Rat BrainUsing Focused Ultrasound. Biomaterials 2016, 83, 257−68.(40) Chodobski, A.; Zink, B. J.; Szmydynger-Chodobska, J. Blood-Brain Barrier Pathophysiology in Traumatic Brain Injury. Transl. StrokeRes. 2011, 2 (4), 492−516.(41) Liu, Y.; Li, D.; Liu, Z.; Zhou, Y.; Chu, D.; Li, X.; Jiang, X.; Hou,D.; Chen, X.; Chen, Y.; Yang, Z.; Jin, L.; Jiang, W.; Tian, C.; Zhou, G.;Zen, K.; Zhang, J.; Zhang, Y.; Li, J.; Zhang, C. Y. Targeted Exosome-Mediated Delivery of Opioid Receptor Mu siRNA for the Treatmentof Morphine Relapse. Sci. Rep. 2015, 5, 17543.(42) Lee, J.; Min, H. S.; You, D. G.; Kim, K.; Kwon, I. C.; Rhim, T.;Lee, K. Y. Theranostic Gas-Generating Nanoparticles for TargetedUltrasound Imaging and Treatment of Neuroblastoma. J. ControlledRelease 2016, 223, 197−206.(43) Liu, Y.; Guo, Y.; An, S.; Kuang, Y.; He, X.; Ma, H.; Li, J.; Lv, J.;Zhang, N.; Jiang, C. Targeting Caspase-3 as Dual Therapeutic Benefitsby RNAi Facilitating Brain-Targeted Nanoparticles in a Rat Model ofParkinson’s Disease. PLoS One 2013, 8 (5), e62905.(44) Lin, K. Y.; Kwon, E. J.; Lo, J. H.; Bhatia, S. N. Drug-InducedAmplification of Nanoparticle Targeting to Tumors. Nano Today2014, 9 (5), 550−559.(45) Chauhan, V. P.; Stylianopoulos, T.; Martin, J. D.; Popovic, Z.;Chen, O.; Kamoun, W. S.; Bawendi, M. G.; Fukumura, D.; Jain, R. K.Normalization of Tumour Blood Vessels Improves the Delivery ofNanomedicines in a Size-Dependent Manner. Nat. Nanotechnol. 2012,7 (6), 383−8.

ACS Nano Article

DOI: 10.1021/acsnano.6b03858ACS Nano 2016, 10, 7926−7933

7933

http://dx.doi.org/10.1021/acsnano.6b03858

Supporting Information

A Neuron-Targeted Nanoparticle for siRNA

Delivery to Traumatic Brain Injuries

Ester J. Kwon

1, Matthew Skalak

1, Riana Lo Bu

1, Sangeeta N. Bhatia

1-6,*

1Koch Institute for Integrative Cancer Research, Massachusetts Institute of Technology,

Cambridge, MA 02139

2Harvard–MIT Division of Health Sciences and Technology, Institute for Medical Engineering

and Science, Massachusetts Institute of Technology, Cambridge, MA 02139

3Electrical Engineering and Computer Science, Massachusetts Institute of Technology,

Cambridge, MA 02139

4Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston,

MA 02115

5Broad Institute of Massachusetts Institute of Technology and Harvard, Cambridge, MA 02139

6Howard Hughes Medical Institute, Cambridge, MA 02139

SI Figure 1. Hydrodynamic diameters of nanoparticles sized after storage at room temperature

in the dark for 7d, 3d, 1d, or prepared fresh.

SI Figure 2. Accumulation of nanoparticles administered 5 miutes after an injury or a sham

procedure when administered nanoparticles. Uninjured brain administered nanoparticles included

as a control. Sham injured animals have minimal amounts of accumulation that is located in the

ring where the skull was removed. Injured brains have significantly more material accumulated

in the 3x3 grid of shear injuries. Scale bar represents 1 cm.

SI Figure 3. Fluorescently labeled free siRNA or siRNA formulated in RVG-targeted

nanoparticles were administered to mice bearing two needle-shear injuries one hour after injury.

One hour after administration of siRNA, mice were perfused and brains were harvested. After

imaging, it is observed that free siRNA can diffuse more freely than siRNA formulated in

nanoparticles. Scale bar represents 1 cm.

SI Figure 4. Knockdown levels of cells that were treated with no treatment (Unt), free siRNA,

RVG-targeted NPs, or Lipofectamine. No knockdown is observed when free siRNA is added to

cells compared to RVG-targeted NPs or Lipofectamine control.

SI Figure 5. Three example fields of view from brain sections stained for nuclei (Hoechst; blue),

FAM-labeled peptide (green), and nicotinic acetylcholine receptor (α−bungarotoxin Alexa

Fluor® 647 conjugate (αB; ThermoFisher 1:500) red). Some peptide co-localized with

α−bungarotoxin staining, but in some fields of view, peptide localization did not co-localize with

α−bungarotoxin staining. Scale bars represent 20 µm.