Development of a chitosan nanofibrillar scaffold for skin

repair and regeneration

Journal: Biomacromolecules

Manuscript ID: bm-2011-00680q

Manuscript Type: Article

Date Submitted by the Author:

18-May-2011

Complete List of Authors: tchemtchoua, victor; university of liège Atanasova, Ganka; Facultés Universitaires Notre-Dame de la Paix Aqil, Abdel; University of Liege, CERM Filée, Patrice; University of Liège Garbacki, Nancy; University of Liège Vanhooteghem, Olivier; Sainte-Elisabeth Hospital Deroanne, Christophe; University of Liège Noel, Agnès; University of Liege and CHU of Liege, GIGA-Research Jerome, Christine; University of Liege, CERM Nusgens, Betty; University of Liège Poumay, Yves; Facultés Universitaires Notre-Dame de la Paix Colige, Alain; University of Liège

ACS Paragon Plus Environment

Biomacromolecules

1

Development of a chitosan nanofibrillar scaffold for skin

repair and regeneration.

Victor T. Tchemtchoua1#

, Ganka Atanasova2#

, Abdel Aqil3, Patrice Filée

4, Nancy Garbacki

1, Olivier

Vanhooteghem5, Christophe Deroanne

1, Agnès Noël

6, Christine Jérome

3, Betty Nusgens

1, Yves

Poumay2, Alain Colige

1*

1 Laboratory of Connective Tissues Biology and

6 Laboratory of Biology Tumor and Development,

GIGA-R, University of Liège, 3 avenue de l’Hôpital, B-4000 Liège, Belgium

2 Laboratory of Cells and Tissues, Facultés Universitaires Notre-Dame de la Paix , 61 rue de Bruxelles,

B-5000 Namur, Belgium

3 Center for Education and Research on Macromolecules and

4Center of Proteins Engineering,

University of Liège, 3 Allée de la Chimie, B-4000 Liège, Belgium.

5 Sainte-Elisabeth Hospital, 15 Place Louise Godin, B-5000 Namur, Belgium.

RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required

according to the journal that you are submitting your paper to)

E-mail: acolige@ulg.ac.be

# These authors contributed equally to the work.

Page 1 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

2

ABSTRACT

The final goal of the present study was the development of a 3-D chitosan dressing that would shorten

the healing time of skin wounds by stimulating migration, invasion and proliferation of the relevant

resident cells. 3-D chitosan nanofibrillar scaffolds produced by electrospinning were compared to

evaporated films and freeze-dried sponges for their biological properties. The nanofibrillar structure

strongly improved cell adhesion and proliferation in vitro. When implanted in mice, the nanofibrilar

scaffold was colonized by mesenchymal cells and blood vessels. Accumulation of collagen fibrils was

also observed. By contrast sponges induced a foreign body granuloma. When used as a dressing

covering full-thickness skin wounds in mice, chitosan nanofibrils induced a faster regeneration of both

the epidermis and dermis compartments. Altogether our data illustrate the critical importance of the

structure of chitosan devices for their full biocompatibility and demonstrate the high beneficial effect of

chitosan as a wound-healing biomaterial.

KEYWORDS

Chitosan, biocompatibility, cell spreading, wound dressing, wound healing

Page 2 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

3

MANUSCRIPT TEXT

Introduction

Skin is the largest organ in the body, covering its entire external surface and forming about 15% of the

total body mass.1 It is composed of three layers: the epidermis, dermis and hypodermis. The epidermis is

a stratified highly cellular structure that provides skin with its vital barrier function. It is mainly

composed of keratinocytes and contains melanocytes in the lower layers. The dermis, constituting the

bulk of the skin is made of a vascularized abundant extracellular matrix rich in collagen, elastin,

hyaluronan and proteoglycans, conferring this tissue with elasticity and mechanical resistance.

Fibroblasts, the major cell type present in the dermis, play an important role in the wound healing

process. The hypodermis contains adipose tissue, is well vascularized and participates to both the

thermoregulatory and mechanical properties of the skin.2 These three layers contribute to protecting the

body from mechanical damage such as wounding which can result in a partial or total physical

disruption of the normal architecture of skin. The most common causes of significant skin loss are

thermal injury, trauma and chronic ulcerations secondary to diabetes, chronic pressure or venous stasis

among others.3 Wound healing is the specific biological process activated for the restoration of the

continuity of the skin. It includes a sequence of processes, starting with coagulation and platelets

activation, recruitment of inflammatory cells, granulation tissue formation and vascularization,

epithelialization, contraction and remodelling.4 Each year, millions of people need skin grafts due to

thermal injury. In the case of wounds that extend through the entire dermis, such as full-thickness burns

or deep ulcers, various skin substitutes such as autografts, allografts and xenografts have been used for

coverage and stimulating the healing process.3 However, these approaches have disadvantages, such as

high cost, limited availability of skin grafts in severely burned patients, risks of host rejection, disease

transmission and immune response.3, 4

One possible solution to such limitations relies on the use of

natural or synthetic biopolymers to design tissue-engineered skin substitutes.5

Page 3 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

4

Several biomaterials have been investigated for their suitability as matrices for wounds covering and

for their beneficial effect on skin repair.2 The therapeutic success of such advanced wound dressings

depends not only on their bioactivity but also on their architecture and mechanical properties, since the

cellular response to biological stimuli depends on the topography and mechanical strength of their

extracellular environment.6, 7

Nanofibre matrices are promising as tissue engineering scaffolds for skin

substitutes notably due, among other advantages, to their morphological similarity to the natural dermal

matrix in skin.8-10

Various techniques, such as phase separation,11

self-assembly12

and electrospinning,8

have been developed to fabricate nanofibrous scaffolds with unique chemical and mechanical properties.

Among these techniques, electrospinning is progressively becoming one of the most popular procedures

because it is a simple, rapid, efficient and inexpensive method for producing nanofibres by applying a

high voltage to electrically charged liquid. In this process, a continuous strand of a solution containing

the polymer is ejected from the open end of a spinneret by high electrostatic force and is finally

deposited randomly on a grounded collector as a nonwoven mat.13

Recently, various synthetic and

natural polymers have been electrospun13

and explored for potential use in a wide variety of

applications, including filters, composite reinforcements, drug or gene delivery vehicles and tissue

engineering.14-17

Chitosan, produced by partial deacetylation of chitin, is a linear polysaccharide composed of randomly

distributed β-(1,4)-linked N-acetyl-D-glucosamine (GlcNAc) and D-glucosamine (GlcN). It possesses

several relevant properties such as biocompatibility and progressive biodegradability. It has also been

claimed as being bacteriostatic and able to promote wound healing18-21

and is considered as one of the

most promising biomacromolecules for tissue engineered scaffolds and tissue repair.22

Several studies

have evaluated the survival and adhesion of different cell types, usually immortalized and less prone to

apoptosis, on various forms of chitosan, chitosan derivatives or mixed polymers scaffolds.14, 17, 19, 23-26

Initial characterizations of the biocompatility of lyophilyzed chitosan sponges have also been performed

in mice.20, 27

However, more comprehensive functional investigations using normal human cutaneous

cells on various types of chitosan scaffolds are required as a first step towards their use in advanced

Page 4 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

5

wound dressings. Furthermore, characterization of the biological properties of nanofibrous chitosan in

vivo and during cutaneous healing is strikingly lacking.

In this work, we analyzed the phenotype of primary human fibroblasts, keratinocytes and dermal

microvascular endothelial cells, the three main cell types of the skin, cultured in vitro on chitosan films

and on a 3D-scaffold of electrospun chitosan nanofibres. The biological properties of these nanofibres

scaffolds and chitosan sponges obtained by lyophilization were also compared in vivo, in order to

evaluate the importance of the 3D-architecture and topography of the biomaterial. Finally, dressings

fabricated with electrospun chitosan nanofibres were evaluated in full thickness wound model in mice.

Experimental section

Materials

Ultra-pure, medical grade chitosan produced from dedicated mushroom cultures (KiOmedine-

CsUPTM

, batch number L07293CsU) with a degree of acetylation of 16% and a molecular weight of

67000 was a gift from Kitozyme (Belgium). High molecular weight polyethylene glycol (PEG,

900000 g/mol) was from Aldrich. All chemical and biochemical reagents were of analytical grade.

Preparation of chitosan films, nanofibres scaffolds and sponges

Films. The solution of chitosan (1% w/v in 1% acetic acid) was filtered through 0.22 µm filters

(Millipore), poured in 12 wells culture plates (1 mL/well) and air-dried for 24 h in a laminar flow culture

hood under sterile conditions. After complete drying, films were immersed for 2 h in 1% NaOH for

neutralizing the residual acetic acid. Chitosan films were then extensively rinsed in sterile distilled water

and dried. Before use in culture, films were preequilibrated for 2 h in the culture medium at 37°C.

Sponges. Chitosan sponges were produced by Kitozyme by freeze-drying 2% chitosan solution in 1%

acetic acid. The chitosan sponges were sterilized by immersion in 70% ethanol and extensively rinsed in

sterile PBS before use.

Electrospun nanofibres scaffolds. The electrospinning solution was prepared by mixing 3 g of

chitosan solution (at 10.5% in 6.5% acetic acid) and 1 g of PEG solution (at 4% in distilled water). Two

mL were then poured into a 5 mL plastic syringe fitted with blunt tipped stainless steel needles (gauge

Page 5 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

6

18 and 21). The solution feed was driven using a syringe pump (Razel Scientific Instruments). An

electrospinning voltage, ranging from 15 to 30 kV, was applied between the needle and the collector

(aluminium foil) by the use of a Spellman SL10 power supply. The positive electrode of a high voltage

power supply was connected to a metal capillary by copper wires. The distance between the tip of the

needle and the surface of the aluminium foil used as a collector was 15 cm and the flow rate of the

solution was 0.75 mL/h. All the electrospinning procedures were performed at room temperature. The

scaffolds were then stabilized by immersion into 1 M NaOH before extensive washing in distilled water.

At this step, biophysical analyses revealed that nanofibres were formed from chitosan only, PEG having

been solubilised and removed during the successive washing steps (Figure S1, Supporting Information).

The electrospun scaffolds were then sterilized (70% ethanol for 2 h) and washed with sterile PBS before

in vivo assays or equilibrated at 37°C during 2 h in culture medium before in vitro assays.

Cell cultures

Fibroblasts were isolated from a human skin biopsy and used at passage 3-4. Endothelial cells

(HMEC) were from dermal microvascular origin and keratinocyte cultures were established from human

skin biopsies.28

Fibroblasts and HMEC were grown in Dulbecco’s Minimum Essential Medium (DMEM, Lonza)

supplemented with 10% fetal bovine serum (FBS, Lonza), antibiotics (100 U/mL penicillin and

100 µg/mL streptomycin, Lonza) and non-essential amino acids (Lonza). Keratinocytes were grown in

keratinocyte grown medium (KGM-2, Clonetics) supplemented with antibiotics (50 U/mL penicillin and

100 µg/mL streptomycin, Sigma) and with SingleQuots KGM-2 (Clonetics) at a final concentration of

50 µg/mL bovine pituitary extract, 10 ng/mL EGF, 5 µg/mL insulin, 5 µg/mL transferrin and 5x10-7

M

hydrocortisone. Cultures were maintained at 37°C in a humidified atmosphere containing 5% CO2.

In vitro evaluation

Preliminary testing showed that chitosan sponges were not an adequate substrate for cell culture since

they did not withstand cell attachment. They were not further evaluated in vitro and only chitosan films

Page 6 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

7

and electrospun nanofibres scaffolds were compared in terms of cell adhesion, morphology, proliferation

and differentiation.

Cell morphology. Time course evaluations of cell morphology, adhesion and spreading were

investigated by microscopic examination for the three cell types. Phase-contrast light microscopy was

used for cultures on chitosan films. Cells on the opaque electrospun scaffolds had to be observed by

scanning electron microscopy (SEM). Briefly, the chitosan scaffolds and adherent cells were fixed for

10 min at 4°C in 2.5% glutaraldehyde in sodium cacodylate buffer (0.1 M sodium cacodylate, pH 7.4,

0.1% CaCl2). After rinsing three times for 5 min in sodium cacodylate buffer, samples were dehydrated

in graded ethanol at 25%, 50%, 75%, 95% and 100% and then subjected to critical point drying. The

scaffolds were then sputter coated with a thin gold layer. The samples were examined and photographed

in a Phillips XL20 SEM. Keratinocytes cultured on electrospun chitosan scaffolds were fixed in 4%

paraformaldehyde for 10 min, dehydrated in methanol and toluol, and then embedded into paraffin. Six

µm-thick paraffin sections were used for histological staining with hematoxylin.

Proliferation. Cell proliferation was evaluated by measuring of 3H-thymidine incorporation into TCA-

insoluble DNA. Briefly, cells were seeded on films or electrospun scaffolds at a density of 15x103 cells

in 12 wells multiplates. The culture medium was renewed every other day. After 1, 3, 5 or 7 days,

2.5 µL/mL of [3H]-thymidine (2.5 Ci/mol) was added. Six hours later, cells were washed and sonicated

in PBS. DNA was collected by precipitation with cold TCA (10% final concentration, 20 min on ice)

and filtration on glass microfibres filter (GF/A, Whatman). After washing with 5% cold TCA and

drying, the TCA-insoluble tritiated material was quantified by liquid scintillation (Aqualuma plus,

LUMAC.LSC.B.V).

Differentiation of keratinocytes. Total RNA was purified from cultures at days 7, 14 and 21 after

seeding on chitosan sponges or nanofibres scaffolds according to an original procedure that we recently

developed, allowing the efficient recovery of proteins and nucleic acids from cells in contact with

chitosan.29

After reverse transcription of RNA (SuperScript II reverse transcriptase, Invitrogen,

Belgium), cDNAs were amplified using Power SYBR Green PCR Master Mix (Applied Biosystems,

Page 7 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

8

Belgium) in a 7300 real-time PCR machine (Applied Biosystems, Belgium). The primer pairs were

specific for keratin 14 (sense: 5’-CGATGGCAAGGTGGTGTC-3’; antisense: 5’-

GGGTGAAGCAGGGTCCAG-3’), keratin 10 (sense: 5’-AATCAGATTCTCAACCTAACAAC-3’;

antisense 5’-TTCCTCTTGCTTTGATGGG-3’), involucrin (sense: 5’-TGAAACAGCCAACTCCAC-

3’; antisense: 5’-CTCATCCAGCACCCTACG-3’) and 36B4 (sense: 5’-

ATCAACGGGTACAAACGAGTC-3’; antisense: 5’-CAGATGGATCAGCCAAGAAGG-3’), a house-

keeping gene used for normalisation of the results.

In vivo evaluation

All procedures were performed with the approval of the Animal Ethical Committee authorities of the

University of Liège (Belgium). Before starting the experiment, 8-10 week-old Balb/C mice received a

subcutaneous injection of Temgesic (0.05 mg/kg) to prevent pain. This treatment was continued twice

daily during 10 days. After anaesthesia with intramuscular injections of Domitor (500 µg/kg) and

Ketamine (60 mg/kg) the animals were treated as described below.

Biocompatibility of implanted chitosan sponges and electrospun nanofibres scaffolds. Eight mm

diameter chitosan electrospun scaffolds and sponges were sterilised in 70% ethanol and soaked in sterile

physiological saline solution. The back skin was shaved and the nanofibres scaffolds or sponges were

implanted subcutaneously at two lateral sites in the back through a 1 cm incision. Waking up was

induced by intramuscular injection of Antisedan (200 µg/kg). Groups of mice were sacrificed at 1, 2, 4,

8 or 12 weeks after implantation. Serum was collected for testing the presence of anti-chitosan

antibodies. Sponges or nanofibres scaffolds, including surrounding adherent tissues, were harvested,

fixed and used for histological examination or transmission electron microscopy as described below.

Wound healing assay. Ten week-old Balb/C mice, weighing 20 ± 3 g ,were obtained from the Animal

Facility of University of Liège. The back skin was carefully shaved and a full thickness (epidermis and

dermis) skin wound was performed with an 8 mm punch. In half of the mice, a chitosan nanofibres

dressing covering the entire wound was applied and moistened with one drop of sterile saline solution.

For the second half of the mice, used as the controls, the wound was only moistened. All the wounds

Page 8 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

9

were then covered successively by a TegadermTM adhesive dressing and by an Elastoplast bandaging

tape that remained in place throughout the experiment, which aimed at protecting the wound and

preventing its contraction. Mice were placed in individual cages and five animals of each group were

euthanized on day 7, 14 and 21. The entire wound areas, including a surrounding skin margin of 5 mm,

were harvested, cut in two identical pieces, fixed in 4% formalin solution and embedded in paraffin for

histological analysis as described below.

Histology and Transmission Electron Microscopy analysis. Pieces of implanted chitosan, in the form

of either sponges or nanofibres scaffolds, were fixed in 4% formalin and embedded in paraffin using

routine protocol. Sections of 5 µm thick were stained with haematoxylin/eosin or used for

immunohistochemistry analyses.

Different primary antibodies were used to identify leukocytes (monoclonal rat anti-mouse CD45, at

1:500, Pharmingen), mesenchymal cells (polyclonal guinea pig antibody against vimentin, at 1:20,

Qartett), type III collagen and type IV collagen (in-house developed rabbit antiserum, at 1:100). After

washing, sections were then incubated with HRP-coupled secondary antibodies. Staining was obtained

using diaminobenzidine (brown) or aminoethylcarbazole (red) as substrates.

For TEM observation, a second piece of the implant was fixed for 18 h at 4°C in 2.5% glutaraldelyde

in 0.1 M sodium cacodylate (pH 7.4) and then for 1 h at 4°C in 1% osmium tetroxide in 0.1 M sodium

cacodylate, pH 7.4. Samples were then dehydrated with increasing concentrations of ethanol and

propylene oxide and embedded in LX112 resin (LADD, USA) before polymerization at increasing

temperatures (37°C for 24 h; 47°C for 24 h; 60°C for 48 h). For orientation purpose, 2 µm sections were

performed, stained with toluidine blue and observed by light microscopy. Ultra-thin sections (50-70 nm)

were cut, stained with uranyl acetate, contrasted with lead citrate, mounted on uncoated grids and

observed by TEM (FEI Tecnai 10, The Netherlands).

For the excisional wounds, 5 µm sections were stained with haematoxylin/eosin or, for

immunohistochemistry analysis, using primary antibodies against type IV collagen according to the

protocol described above.

Page 9 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

10

Detection of anti-chitosan antibodies by ELISA. An ELISA assay was developed to evaluate the

potential presence of circulating antibodies against chitosan in mice implanted with chitosan sponges or

electrospun nanofiber scaffolds. Ninety-six wells plates were coated for 2 h with chitosan (180 µL per

well, 1% solution in 1% acetic acid) incubated for 2 h at room temperature, washed twice for 20 min

with 20 mM Tris buffer (pH 7.6) and then blocked overnight at 4°C by 5% bovine serum albumin in

PBS (BSA/PBS). Plates were washed three times with PBS containing 0.05% Tween 20 (PBS/Tween).

Serum samples from chitosan-implanted mice were serially diluted in BSA/PBS and dispensed in

duplicate. Negative and positive controls were, respectively, BSA/PBS and BSA/PBS containing anti-

chitosan antibody (generous gift of Daniel Hartmann, Laboratoire des Biomatériaux et Remodelages

Matriciels, EA 3090, Lyon, France). Plates were incubated overnight at 4°C. After washing with

PBS/Tween, the HRP-coupled secondary antibody was added (100 µL rabbit anti-mouse IgG at 1:1000

for experimental samples and goat anti-rabbit IgG at 1:1000 for positive control). Plates were incubated

for 2 h at room temperature, and then washed in PBS/Tween. Staining was obtained by the addition of a

solution of 1 mM ABTS containing 0.03% H2O2 and incubation during 1 h at 37°C in the dark. The

reaction was stopped by adding a solution of 1% SDS. OD at 405 nm was measured by using a

microplate spectrophotometer.

Results

Three-dimensional architecture of the chitosan sponges and nanofibrillar scaffold

Production of chitosan nanofibres by electrospinning can hardly be performed by using pure chitosan

solutions. This result from their inappropriate viscosity and from the presence of a large excess of

positive charges on the chitosan molecules at the pH required for its complete solubilization. Various

concentrations of PEG of different MW were tested as additive to improve these critical biophysical

properties. For the medical grade chitosan used here, the best conditions for electrospinning were

determined to be solutions containing 7.9% chitosan and 1% PEG, a voltage of 30 kV, a flow rate of

0.75 mL/h, and a needle to collector distance of 15 cm.

Page 10 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

11

The size, morphology and surface topography of the electrospun nanofibres and of lyophilized

sponges were investigated by scanning electron microscopy (SEM) (Figure 1). Individual electrospun

nanofibres had a mean diameter in the range of 50 to 200 nm. This non-woven material delineated small

pores with a diameter of less than 5 µm and had porosity greater than 90%. By contrast, lyophilized

sponges appeared as intertwined lamellae of chitosan delimiting irregular spaces that can extend up to

100 µm.

Figure 1. Scanning electron micrograph of the scaffold of chitosan nanofibres produced by

electrospinning (a) and of the chitosan sponge obtained by freeze-drying technique (b).

Cell attachment, spreading and proliferation in vitro

Cell Morphology and spreading. Morphology and kinetics of adhesion and spreading of fibroblasts,

endothelial cells and keratinocytes on tissue-culture polystyrene plates used as control and on films of

chitosan were evaluated by phase-contrast microscopy. On plastic, the three cell types were already fully

attached and spread at day 1 and reached confluence within one week (Figure 2). By contrast, fibroblasts

and endothelial cells seeded on chitosan films failed to fully spread, even at day 7, did not multiply, or

even regressed. This did not result from a potential toxicity of chitosan since cells recovered from

cultures on chitosan films at day 1 or 3 and seeded back on plastic in their own conditioned culture

medium rapidly adhered, spread and started to proliferate (not shown). For keratinocytes, a delay in

attachment and spreading on chitosan films was also observed, although to a lesser extent as compared

to the other cell types. Initially, keratinocytes seeded on chitosan films multiplied in a similar manner to

those seeded on plastic culture dishes, but they were not able to form a confluent monolayer. This

Page 11 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

12

growth arrest was correlated with atypical cell morphology. Keratinocytes were swollen, with condensed

nucleus and without nucleolus, an observation often associated with cell suffering and even cell death.

Page 12 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

13

Figure 2. Phase contrast microscopy observation of human fibroblasts, microvascular endothelial cells

and keratinocytes cultured on plastic or on a film of chitosan film for 1, 3 or 7 days.

When seeded on electrospun chitosan nanofibers, cell had to be observed by Scanning Electron

Microscopy because of the opacity of the material (Figure 3). The three cell types already adhered at day

1, were fully spread at day 3 and proliferated with time, in contrast with their behaviour on chitosan

films. It has to be noticed that keratinocytes at day 7 tend to form clusters of flat cells tightly joined

together, a physiological and essential property of keratinocytes in vivo.

Figure 3. Scanning electron microscopy observation of human fibroblasts (a), microvascular endothelial

cells (b) and keratinocytes (c) cultured on electrospun chitosan nanofibres for 1, 3 or 7 days.

Page 13 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

14

This observation was further confirmed by histochemistry showing a continuous layer of epithelial

cells at the surface of the electrospun nanofibres scaffold (Figure 4).

Figure 4. Confluent monolayer of keratinocytes cultured on electrospun chitosan nanofibres.

Keratinocytes were cultured for 7 days on chitosan nanofibres, fixed and embedded in paraffin.

Hematoxylin/Eosin staining of transversal sections showing a continuous layer of contiguous cells.

Closer look at fibroblast cultures (Figure 5a) also showed cellular extensions closely interacting with

individual nanofibres, explaining why electrospun chitosan induces cell spreading. It was also observed

that these cells were able to progressively invade the superficial layers of the scaffold (Figure 5b).

Figure 5. Fibroblasts interactions with chitosan nanofibres. Fibroblasts were cultured for 3 days on

electrospun chitosan nanofibres, fixed and processed for scanning electron microscopy. Some cell

extensions interacting with individual or small groups of nanofibres (a, arrows), or progressively

infiltrating the superficial layers of the biomaterial (b, arrow).

Cell proliferation. A [3H]-thymidine incorporation assay was used to evaluate the proliferation of the

three cell types cultured on chitosan films and electrospun nanofibres scaffold. Some increase of the

incorporation was observed between day 1 and day 5 for fibroblasts (Figure 6a) and endothelial cells

(Figure 6b) cultured on chitosan films. However, no significant increase occurred later demonstrating

Page 14 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

15

that cells stopped to proliferate, even though they were not confluent. By contrast, cells on chitosan

nanofibres proliferated faster and kept growing during the entire experiment, confirming the

microscopic observations (Figure 3). Keratinocytes adhered in a similar manner on chitosan films and

nanofibres and similarly proliferated during the first days (Figure 6c). A significantly increased rate of

[3H]-thymidine incorporation was noted on chitosan nanofibres on day 5. Incorporation decreased on

both substrates at day 7 but remained higher on nanofibres than on films.

Page 15 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

16

Figure 6. Cell proliferation on chitosan film and nanofibres. Human fibroblasts (a), microvascular

endothelial cells (b) and keratinocytes (c) were cultured for increasing time (1 to 7 days) on a chitosan

film or on a nanofibrous chitosan scaffold. A pulse of [3H]-thymidine was then performed for 6 h.

Proliferation was evaluated by measuring the amount of radioactivity incorporated in DNA.

Keratinocyte differentiation. A significant feature of keratinocytes is their differentiation program that

affects their proliferation rate and modifies their pattern of proteins expression. Proliferative

keratinocytes in culture express keratin 5 and keratin 14 that are markers for undifferentiated cell

phenotype. When committed towards differentiation, suprabasal keratin 10 and involucrin and, later,

filagrin and loricrin become expressed.28

RT-PCR was performed to determine the expression levels of

involucrin and keratins 10 and 14, as a way to evaluate the differentiation process of keratinocytes in the

different culture conditions (not shown). Keratin 14 expression remained constant with time (day 7 up to

day 21) in cultures on plastic, showing that these conditions do not induce differentiation. By contrast its

level slowly decreased when keratinocytes were cultured on chitosan nanofibres, indicating that they

were becoming more prone to differentiation. This hypothesis was further confirmed by data showing a

progressive increase of keratin 10 and involucrin expression in keratinocytes reaching confluence

(between day 14 and 21) when cultured on nanofibres.

In vivo evaluation

Biocompatibility. Following subcutaneous implantation of chitosan sponges or electrospun nanofibres

scaffold, mice recovered quickly and looked healthy during the entire experiment. Visual inspection of

the implantation site did not reveal any local infection or inflammation. Blood samples were collected at

increasing times after implantation. No circulating antibodies specific for chitosan were found by ELISA

even 12 weeks after implantation. Visual inspection after dissection showed that sponges and nanofibres

scaffolds adhere similarly to the inner face of the skin. No obvious swelling or fluid accumulation was

observed. Sponges appeared however reddish already at one week and were progressively surrounded by

a thick fibrous capsule while the nanofibres scaffolds were pale and seemed to be progressively

integrated as self material (Figure 7).

Page 16 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

17

Figure 7. Macroscopic appearance of chitosan sponge (a) and electrospun nanofibres (b) 12 weeks after

subcutaneous implantation. Sponges appear red and thickened by fibrous deposit while nanofibres

scaffold remained white and could be hardly distinguished from neighbouring mouse tissue.

Histological analyses were performed on the two biomaterials after increasing time of implantation

(Figure 8). In the case of chitosan sponges, cells and extracellular fibrous material accumulation were

detected as a capsule surrounding the biomaterial while the large pores formed by the chitosan lamellae

remained largely empty even 12 weeks after implantation (Figures 8a, c, e and g). When using a scaffold

composed of chitosan nanofibers, progressive cell invasion of the nanofibrillar material was already

observed after one week of implantation (Figures 8b, d, f and h). Remarkably, cells showed a

preferential orientation along the fibres that resembled the undulating overall structure of collagen

bundles in the skin (Figure 8, inserts). A surrounding fibrous capsule was never evidenced.

Page 17 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

18

Figure 8. Histological examination of implanted biomaterials. Chitosan sponges (a, c, e, g) and

electrospun scaffolds (b, d, f, h) were implanted subcutaneously and then collected after 1 (a, b), 4 (c, d),

8 (e, f) or 12 (g, h) weeks. After fixation and embedding in paraffin, sections were stained with

hematoxylin/eosin. Chitosan appeared stained in red, as loosely interconnected lamellae in sponges and

Page 18 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

19

as undulating bundles of thin structures in electrospun scaffold. A thick connective tissue capsule was

evidenced around sponges already one week after implantation, while the inner space between lamellae

remained essentially empty. Electrospun scaffolds were never encapsulated and were rapidly invaded by

cells that adopted a general orientation similar to that of chitosan nanofibres.

Transmission electron microscopy was performed to gain a better insight into the physical interactions

between cells and chitosan biomaterials. Only a few rounded cells, most probably leukocytes, were

detected within the sponges. By contrast, collagen accumulation and a highly cellular capsule

surrounding the scaffold were evidenced (not shown) as expected from histochemical analyses. In

electrospun scaffolds, chitosan nanofibres appeared as electron dense cylindrical or elongated structures,

depending upon their relative orientation (Figure 9). Twelve weeks after implantation, a significant

proportion of the interfibrillar space contained fibroblast-like cells as judged from their shape and the

deposition of tightly packed collagen fibrils in their surroundings. Remarkably these cells were in close

contact with chitosan nanofibres (Figures 9b and c). As compared to the pictures before implantation

(Figure 9a), the contour of nanofibres in implanted scaffolds was less sharply delineated suggesting

either ongoing degradation process or progressive coating by extracellular macromolecules. Dark and

irregular structures were also observed within some cells (Figure 9e), suggesting a process of

endocytosis and intracellular degradation of chitosan. Most interestingly, bundles of collagen fibrils

were in close contact with cells and with chitosan nanofibres (Figures 9b and d).

Page 19 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

20

Figure 9. Transmission electron microscopy evaluation of implanted electrospun chitosan scaffold. In

the chitosan scaffold before implantation, the nanofibres appear as elongated or circular black structures

(a), depending upon their orientation. Twelve weeks after implantation elongated cells were detected (b).

In most regions they are constricted between many nanofibres (c). These cells secrete collagen,

visualized as white structure (b, arrow). In many regions these collagen bundles are also tightly

associated with the chitosan scaffold (d). Some cells contain black disorganized structures (asterisks)

that could represent intracellularly degraded nanofibres (e). The contour of chitosan nanofibres is less

Page 20 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

21

sharply delimited 12 weeks after implantation (d) as compared to control samples (a), suggesting either a

progressive degradation of the nanofiber surface or coating by extracellular material.

Qualitative analyses were also performed in order to characterize the nature of invading cells.

Vimentin staining indicated that a large majority of cells invading the nanofibres scaffold were

mesenchymal cells, such as fibroblasts (Figures 10b and d). By contrast again, sponges were essentially

negative inside but strongly positive at the level of the surrounding connective tissue (Figures 10a and

c).

Figure 10. Colonization of implanted biomaterials by fibroblastic cells. Chitosan scaffolds were

recovered after 1 (a, b) and 12 (c, d) weeks, fixed, sectioned and analyzed for the presence of

fibroblastic cells stained for vimentin, a marker for cells of mesenchymal origin. A progressive

infiltration of the electrospun scaffolds (b and d) was clearly observed while only the fibrous capsule

was positive for sponges (a and c).

Page 21 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

22

These data were further confirmed by staining type III collagen, a macromolecule produced by

mesenchymal cells playing a key role in connective tissue formation and remodelling (Figure 11). Most

interestingly, the orientation of fibroblastic cells and newly formed collagen fibrils in electrospun

scaffold was strongly influenced by the undulating architecture of the chitosan nanofibres.

Figure 11. Collagen deposition in electrospun scaffolds. Electrospun scaffolds collected 12 weeks after

implantation was analyzed for its fibrillar collagen content by using an antibody specific for type III

collagen. Strong positivity was observed in regions colonized by fibroblastic cells. The newly deposited

collagen structures tend to have the same orientation as the chitosan nanofibrillar structures.

New blood vessels formation was also evaluated as it represents a critical step for full

biocompatibility of implanted scaffolds, by allowing long-term survival of invading cells. Type IV

collagen was stained, as it is an essential component of basement membranes surrounding mature blood

vessels (Figure 12). Vessels were detected within the connective tissue capsule surrounding the sponges

but never in the sponge itself. By contrast they were present inside the electrospun scaffold , confirming

an ongoing process of remodeling and integration of the nanofibrillar material.

Page 22 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

23

Figure 12. Blood vessel formation in chitosan scaffolds. Chitosan sponges and electrospun scaffolds

were recovered 12 weeks after subcutaneous implantation and analyzed by using an anti-type IV

collagen antiserum for visualization of the basement membrane that forms around blood vessels.

Staining was detected only in the connective tissue capsule for chitosan sponges (a) while progressive

formation of functional capillaries was identified in the electrospun scaffold (b).

Effect of electrospun chitosan nanofibres used as dressing during skin wound healing. Eight mm full

thickness wounds were performed on the back of mice. The wounded areas, control or covered by

chitosan scaffold, were collected at increasing time after skin excision and examined by microscopy

after HE staining. At day 7, the control wound beds were characterized mostly by the presence of

inflammatory cells in a poorly organized tissue (Figure 13a) while a progressive remodeling of the

granulation tissue could be already observed in chitosan-treated wounds (Figure 13b). Furthermore, the

neoepidermis covering the initial wound site has migrated more extensively. After 2 weeks, the

complete closure of the wound by the epidermal layer was observed in the treated mice, the

reepithelialization being far from complete in the control mice (Figure 13, compare c to d). Remarkable

also in treated mice is the presence of hair follicles and glands in area corresponding to the initial

wounds. At day 21, all the wounds were covered by an epithelium. However a faster regeneration was

still observed for chitosan-treated wounds again illustrating its beneficial effect.

Page 23 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

24

Page 24 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

25

Figure 13. Effect of chitosan nanofibers dressings on wound healing. Full thickness wounds were

performed in the back of mice (n=30) and covered by a chitosan nanofiber dressing (b, d, f) or not (a, c,

e). At day 7 (a, b), 14 (c, d) or 21 (e, f), the entire wounded areas, including sourrounding skin, were

harvested, fixed, embedded in paraffin and analysed after hematoxylin-eosin staining. The initial

marging of the wounds are localized at the two extremities of the pictures. The fibrin clot (FC)

sometime detached during the section or the staining of the samples. The dots lines mark the limit

between FC and the granulation tissue (GT). The neoepidermis (NE) that progressively covers the GT is

also indicated. The adnexa (hair bulbs, glands) that are present in normal skin appear as circular and/or

elongated structures in the dermis depending in the angle of the sections.

These data were further confirmed by immunohistochemistry using anti-collagen type IV antibodies.

At day 14, a well developed vascular network is already present in the chitosan treated wounds that is

almost completely absent in the control mice (Figure 14a and b). Note the restoration of a positively

stained dermo-epithelial basement membrane in the treated wounds (Figure 14b). The blood vessels

network is not yet completed restored by day 21 in control as compared to treated wounds (Figure 14c

and d).

Page 25 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

26

Figure 14. Blood vessel formation in wounded tissue. Blood vessels were visualised by using an anti-

type IV collagen antiserum. They were detected in the granulation tissue of chitosan-treated wounds (b,

d) by days 14 and 21, while in control group they were absents by day 14 (a) and poorly detected by day

21 (c).

Altogether these data demonstrate that dressing made of electrospun chitosan nanofibers can stimulate

the formation and the remodelling of the granulation tissue, enhance its covering by a neoepidermis and,

ultimately, favour a faster regeneration of the skin.

Discussion

Chronic dermal wounds, as venous ulcers, pressure sores and diabetic foot ulcers, represent a major

health problem that affects million of people worldwide and induces billions dollars of social and

economic costs30

explaining the large research efforts focused on the development of new therapeutic

approaches and dressings that could improve the impaired healing process.3, 31

Cutaneous wound healing

can be divided into different phases that can however interplay with a significant temporal overlap.32, 33

Page 26 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

27

After platelet aggregation and growth factors release upon degranulation, neutrophils and macrophages

are recruited to the site of injury in the fibrin clot constituting a provisional matrix support. Their

primary roles are the cleansing of the wound, by removing cell debris and phagocytosing bacteria, and

the secretion of soluble factors required for initiation of the proliferating phase. However, an excessive

inflammation depresses the healing rate by excessive production of ROS and proteases.34

During the

proliferative phase, recruited stem cells and (myo)fibroblasts multiply in response to growth factors and

start to synthesize extracellular matrix macromolecules such as fibrillar collagens.35

Capillary formation,

as a result of endothelial cells recruitment and proliferation, is also observed, which provides fibroblasts

with oxygen and nutrients.32

During the remodeling phase, the tensile strength of the repairing tissue

gradually increases as randomly deposited collagen fibrils within a fibrin-fibronectin network are

replaced by organized fibrils that align themselves in a specific orientation dictated by the environment

and by the mechanical forces developing in the wounded region. This is of crucial importance because it

restores the mechanical resistance of the skin but also because it strongly regulates the phenotype of

cells in the repaired tissue. The fourth phase consists in the restoration of the epithelial barrier through

processes of proliferation, migration and differentiation of keratinocytes.

Ideally, dressings developed for the treatment of chronic non-healing wounds should improve these

different phases. In addition, they must be non-toxic and non-immunogenic, maintain a moist

environment, absorb excess exudates, allow gas exchange and provide protection against bacteria

infection.36

Over the past decade efforts have been directed towards developing new dressings that would favor

healing by providing a provisional scaffold that would be progressively invaded and remodeled by cells

of the patient, thus favoring healing of deep wounds.3 In this medical context, wound dressings

consisting of nanofibrillar biomaterial would have several advantages such as potential to mimic the

nanofibrillar network forming the extracellular matrix in vivo, a high porosity with pores of small size

and a high specific surface area to volume ratio.25, 37, 38

Chitosan is one of the most promising

biomaterial for engineering such new dressings because it is innocuous, stable over extended period of

Page 27 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

28

time in various conditions and relatively cheap.15, 37

Some characterizations of its biological properties

have been described.17, 20, 39

Most of these studies, although providing valuable information, are not fully

relevant for evaluating the benefit of using pure chitosan for favoring the healing of “hard to heal”

wounds, such as ulcers, and for improving the mechanical and biological quality of the repaired skin. As

examples, some previous works evaluated chitosan in association with other macromolecules, such as

gelatin40

and collagen,41

preventing the characterization of their specific respective effects, while some

others used chitosan as films, granules or sponges,17, 42, 43

so structures lacking the nanofibrillar structure

found in the dermis. Finally, a limited number of studies have reported preliminary characterizations of

nanofibrillar chitosan as cell substrate using immortalized cells or primary cells that are not involved in

skin repair.19, 44

This large diversity of experimental conditions probably explains why some data about

chitosan are sometimes contradictory17, 24

and illustrates the need of a dedicated study.

Here we evaluated different types of chitosan scaffolds for their biocompatibility in vitro and in vivo,

especially focusing our study on human primary cells and processes involved in dermal wound healing.

In order to improve the relevance of our data for further use in clinic, we used chitosan produced from

dedicated mushroom cultures as it should facilitate GMP and medical grade production as compared to

chitosan produced from wastes of fishing industry and should be less susceptible to batch to batch

variability.45

Three types of chitosan-based material were used: evaporated films, sponges (produced by

freeze-drying of chitosan solutions and largely consisting in micrometric-thick lamellae) and scaffolds

of nanofibres produced by electrospinning. Our working hypothesis was that the nanofibrillar structures

could improve the biological properties of chitosan by increasing its specific surfaces and/or mimicking

the fibrillar collagen scaffold found in vivo in the skin.

In vitro, fibroblasts and endothelial cells attach, spread and significantly proliferate only when they are

seeded on chitosan in the form of electrospun nanofibres. This scaffold was also a better substrate for

keratinocytes. These data clearly demonstrate that interactions of cells with chitosan strongly depend on

the structural architecture of the biomaterial. This most probably explains the large diversity of data

reported in the literature regarding the use of chitosan for cell culture and demonstrates that electrospun

Page 28 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

29

chitosan nanofibres are an efficient adhesive substrate for the three most important cell types involved in

skin wound healing.39, 46, 47

This observation prompted us to evaluate its biological properties in mice in

vivo using models of subcutaneous implantation or of full-thickness skin wound, which allows

evaluation over a long and a short period of time, respectively.48

For subcutaneous implantation, chitosan nanofibres scaffolds were compared to sponges as they both

exhibit a 3D-structure but display a distinct architecture and porosity. Although it has been reported that

chitosan is not immunogenic, the potential production of antibodies was evaluated in blood, especially at

the longest time-points. All these assays remained negative confirming previous data and demonstrating

that neither chitosan sponges nor nanofibres elicit a humoral immune response.20, 21, 49

Histological

analyses of the implanted scaffolds indicate that chitosan sponges induce a strong inflammatory reaction

characterized by a massive recruitment of leukocytes and by the formation of a multilayered tissue

capsule resembling foreign body granuloma, a host defense reaction to inert foreign material that is too

large to be phagocytosed by macrophages. Sponges were never colonized by connective tissue’s host

cells and never contained blood vessels even 12 weeks after implantation. By sharp contrast, a moderate

infiltrate of leukocytes was observed in electrospun nanofibres material and granuloma was never

induced, demonstrating a much better biocompatibility of chitosan nanofibres. Immunohistological

analysis revealed that fibroblastic cells and numerous blood vessels are actively colonizing the

interfibrillar space which is progressively filled with extracellular matrix macromolecules, such as

collagen. The general orientation of cells and newly deposited collagen was strongly influenced by the

undulating topography of the chitosan nanofibres that remarkably mimick the structure of the dermis.

These data demonstrate that such type of scaffold is able to dictate the architecture of the repairing

tissues, therefore influencing its biomechanical properties, and could be most relevant for tissue

engineering.

Transmission electron microscopy was performed to have a better insight into the mechanisms used by

cells to invade a biomaterial with pores having a mean size of less than 5 µm. Although some potential

degradation of the nanofibres was suggested by a few pictures, degradation is obviously too limited to

Page 29 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

30

account for the fast colonization of the nanofibrillar network by the host cells. Invasion results probably

from two mechanisms. Firstly, pictures show cells constricted between several chitosan nanofibres

suggesting that migration could occur through progressive narrowing of the entire cell body, including

nucleus.50

Secondly, since the electrospun scaffold is a non-woven material formed by the accumulation

of non-cross-linked nanofibres, displacement of the chitosan nanofibres may be expected and should

favor cell migration. A further worth noting observation is the close interaction between the chitosan

nanofibres, the invading cells and the newly deposited collagen fibrils. This is in agreement with the

collagen III staining and further demonstrates the full biocompatibility of this type of scaffold that is

properly vascularized and progressively integrated as a host tissue.

Used as a dressing, chitosan nanofibres were found to adhere uniformly to freshly excised wound

surface, to absorb the exudates and to be fully biocompatible. As compared to controls it was found to

favour skin regeneration at different levels. It reduces the time required for covering the wound by a

neoepidermis, which is a critical property for limiting risks of infection in clinical situations, and it

accelerates the formation and the remodelling of the repairing/regenerating skin. By day 21 most of the

initial wound bed is replaced by a well vascularized newly formed skin containing hair follicles and

glands, as in a normal skin, while such differentiated structures are still lacking in controls. This

improved regeneration of a functional skin, rather than just a repair process generating a poorly

organized tissue, is of prime importance in several clinical situations characterized by extended skin loss

such as observed in deep ulcers and for patients with severe burns. The specific mechanisms involved in

this beneficial pro-healing activity remain to be firmly identified but they are certainly influenced by the

nanofibrillar structure of the chitosan, which improves its biocompatibility, as seen in this work, and

increases its specific surface, a feature that favours the trapping of factors detrimental to the repair

process when present in excess (proteases, reactive oxygen species...) and the progressive release of

chitosan fragments responsible for the recruitment and activation of leukocytes and mesenchymal

cells.18

Page 30 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

31

Conclusions

Despite many conflicting data, a review of the literature indicated that chitosan may be a valuable

biomaterial for cell culture and for improving wound healing. This work represents however the first

integrated study investigating specifically several complementary aspects of the use of chitosan as a

wound dressing for cutaneous lesions. A chitosan scaffold produced by accumulation of electrospun

nanofibres is a much better substrate for adhesion, growth and differentiation of the three main skin cell

types (keratinocytes, fibroblasts and endothelial cells) than other type of chitosan device (films, sponges,

gel…). Moreover, long term evaluations demonstrate its full biocompatibility and its progressive

integration and colonization in vivo by contrast with the lamellar three-dimensional chitosan sponge that

is considered as a “non-self” foreign body. Chitosan nanofibres also promote recovery of full thickness

wounds as illustrated by an improved re-epithelialization process, an enhanced vascularization and a

more efficient remodelling of the granulation tissue. These data clearly indicate that electrospun chitosan

scaffold is a promising wound dressing, especially for the treatment of deep ulcers that would benefit

from the use of a three-dimensional scaffold filling the entire wound and being progressively

incorporated in the repaired tissue. Moreover, our results show also that structure and orientation of the

nanofibrillar network dictate the repairing process, and therefore the architecture and mechanical

properties of the newly formed tissue, a highly desirable property for various applications in tissue

engineering.

Page 31 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

32

ACKNOWLEDGMENT

We thank the Région wallonne (DG06) for its support (GoCell project, Grants n° 516025 and 616262).

The anti-chitosan antibody is a generous gift of Daniel Hartmann, Laboratoire des Biomatériaux et

Remodelages Matriciels, EA 3090, Lyon, France.

SUPPORTING INFORMATION AVAILABLE

Figure S1. Thermal behaviour of pure PEG, pure chitosan and electrospun scaffold, before (BS) and

after stabilization (SCS).

This material is available free of charge via the Internet at http://pubs.acs.oorg

REFERENCES

(1) Kanitakis, J. Eur J Dermatol 2002, 12, 390-399; quiz 400-391.

(2) Metcalfe, A. D.; Ferguson, M. W. Biomaterials 2007, 28, 5100-5113.

(3) Clark, R. A.; Ghosh, K.; Tonnesen, M. G. J Invest Dermatol 2007, 127, 1018-1029.

(4) Singer, A. J.; Clark, R. A. N Engl J Med 1999, 341, 738-746.

(5) Lutolf, M. P.; Hubbell, J. A. Nat Biotechnol 2005, 23, 47-55.

(6) Engler, A. J.; Sen, S.; Sweeney, H. L.; Discher, D. E. Cell 2006, 126, 677-689.

(7) Vogel, V.; Sheetz, M. Nat Rev Mol Cell Biol 2006, 7, 265-275.

(8) Li, W. J.; Laurencin, C. T.; Caterson, E. J.; Tuan, R. S.; Ko, F. K. J Biomed Mater Res 2002, 60,

613-621.

(9) Teo, W. E.; He, W.; Ramakrishna, S. Biotechnol J 2006, 1, 918-929.

(10) Liao, S.; Li, B.; Ma, Z.; Wei, H.; Chan, C.; Ramakrishna, S. Biomed Mater 2006, 1, R45-53.

(11) Smith, L. A.; Ma, P. X. Colloids Surf B Biointerfaces 2004, 39, 125-131.

(12) Whitesides, G. M.; Boncheva, M. Proc Natl Acad Sci U S A 2002, 99, 4769-4774.

(13) Barnes, C. P.; Sell, S. A.; Boland, E. D.; Simpson, D. G.; Bowlin, G. L. Adv Drug Deliv Rev

2007, 59, 1413-1433.

(14) Sangsanoh, P.; Waleetorncheepsawat, S.; Suwantong, O.; Wutticharoenmongkol, P.;

Weeranantanapan, O.; Chuenjitbuntaworn, B.; Cheepsunthorn, P.; Pavasant, P.; Supaphol, P.

Biomacromolecules 2007, 8, 1587-1594.

(15) Khor, E.; Lim, L. Y. Biomaterials 2003, 24, 2339-2349.

(16) Chen, G.; Sato, T.; Ohgushi, H.; Ushida, T.; Tateishi, T.; Tanaka, J. Biomaterials 2005, 26,

2559-2566.

(17) Chatelet, C.; Damour, O.; Domard, A. Biomaterials 2001, 22, 261-268.

(18) Ueno, H.; Mori, T.; Fujinaga, T. Adv Drug Deliv Rev 2001, 52, 105-115.

(19) Bhattarai, N.; Edmondson, D.; Veiseh, O.; Matsen, F. A.; Zhang, M. Biomaterials 2005, 26,

6176-6184.

(20) VandeVord, P. J.; Matthew, H. W.; DeSilva, S. P.; Mayton, L.; Wu, B.; Wooley, P. H. J Biomed

Mater Res 2002, 59, 585-590.

(21) Mi, F. L.; Tan, Y. C.; Liang, H. F.; Sung, H. W. Biomaterials 2002, 23, 181-191.

Page 32 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

33

(22) Kim, I. Y.; Seo, S. J.; Moon, H. S.; Yoo, M. K.; Park, I. Y.; Kim, B. C.; Cho, C. S. Biotechnol

Adv 2008, 26, 1-21.

(23) Chang, Y. Y.; Chen, S. J.; Liang, H. C.; Sung, H. W.; Lin, C. C.; Huang, R. N. Biomaterials

2004, 25, 3603-3611.

(24) Hamilton, V.; Yuan, Y.; Rigney, D. A.; Puckett, A. D.; Ong, J. L.; Yang, Y.; Elder, S. H.;

Bumgardner, J. D. J Mater Sci Mater Med 2006, 17, 1373-1381.

(25) Jayakumar, R.; Prabaharan, M.; Nair, S. V.; Tamura, H. Biotechnol Adv 2010, 28, 142-150.

(26) Yeh, L. K.; Chen, Y. H.; Chiu, C. S.; Hu, F. R.; Young, T. H.; Wang, I. J. J Biomed Mater Res A

2008.

(27) Barbosa, J. N.; Amaral, I. F.; Aguas, A. P.; Barbosa, M. A. J Biomed Mater Res A 2010, 93, 20-

28.

(28) Minner, F.; Herphelin, F.; Poumay, Y. Methods Mol Biol 2010, 585, 71-82.

(29) Tchemtchoua, V. T.; Atanasova, G.; Aqil, A.; Maquet, V.; Jerome, C.; Poumay, Y.; Colige, A.

Anal Biochem 2009, 393, 145-147.

(30) Rippon, M.; Davies, P.; White, R.; Bosanquet, N. J Wound Care 2008, 17, 224-227.

(31) Falanga, V. J. Adv Skin Wound Care 2000, 13, 15-19.

(32) Davidson, J. M. J Hand Ther 1998, 11, 80-94.

(33) Theoret, C. L. Vet Clin North Am Equine Pract 2005, 21, 1-13.

(34) Eming, S. A.; Krieg, T.; Davidson, J. M. J Invest Dermatol 2007, 127, 514-525.

(35) Hinz, B. J Invest Dermatol 2007, 127, 526-537.

(36) Theoret, C. Vet Surg 2009, 38, 905-913.

(37) Ignatova, M.; Manolova, N.; Markova, N.; Rashkov, I. Macromol Biosci 2009, 9, 102-111.

(38) Lee, K. Y.; Jeong, L.; Kang, Y. O.; Lee, S. J.; Park, W. H. Advanced Drug Delivery Reviews

2009, 61, 1020-1032.

(39) Howling, G. I.; Dettmar, P. W.; Goddard, P. A.; Hampson, F. C.; Dornish, M.; Wood, E. J.

Biomaterials 2001, 22, 2959-2966.

(40) Huang, Y.; Onyeri, S.; Siewe, M.; Moshfeghian, A.; Madihally, S. V. Biomaterials 2005, 26,

7616-7627.

(41) Ma, L.; Gao, C.; Mao, Z.; Zhou, J.; Shen, J.; Hu, X.; Han, C. Biomaterials 2003, 24, 4833-4841.

(42) Mi, F.-L.; Shyu, S.-S.; Wu, Y.-B.; Lee, S.-T.; Shyong, J.-Y.; Huang, R.-N. Biomaterials 2001,

22, 165-173.

(43) Khan, T. A.; Peh, K. K. J Pharm Pharm Sci 2003, 6, 20-26.

(44) Shim, I. K.; Suh, W. H.; Lee, S. Y.; Lee, S. H.; Heo, S. J.; Lee, M. C.; Lee, S. J. J Biomed Mater

Res A 2009, 90, 595-602.

(45) Pochanavanich, P.; Suntornsuk, W. Lett Appl Microbiol 2002, 35, 17-21.

(46) Zhou, Y.; Yang, D.; Chen, X.; Xu, Q.; Lu, F.; Nie, J. Biomacromolecules 2008, 9, 349-354.

(47) Neamnark, A.; Sanchavanakit, N.; Pavasant, P.; Rujiravanit, R.; Supaphol, P. European Polymer

Journal 2008, 44, 2060-2067.

(48) Vogt, J. C.; Brandes, G.; Kruger, I.; Behrens, P.; Nolte, I.; Lenarz, T.; Stieve, M. J Mater Sci

Mater Med 2008, 19, 2629-2636.

(49) Sorlier, P.; Hartmann, D. J.; Denuziere, A.; Viton, C.; Domard, A. J Biomed Mater Res A 2003,

67, 766-774.

(50) Friedl, P.; Wolf, K. J Cell Biol 2010, 188, 11-19.

SYNOPSIS TOC

Page 33 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

34

Page 34 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

82x39mm (300 x 300 DPI)

Page 35 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

172x230mm (300 x 300 DPI)

Page 36 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

177x138mm (300 x 300 DPI)

Page 37 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

53x37mm (300 x 300 DPI)

Page 38 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

82x39mm (300 x 300 DPI)

Page 39 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

171x172mm (300 x 300 DPI)

Page 40 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

82x33mm (300 x 300 DPI)

Page 41 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

127x209mm (300 x 300 DPI)

Page 42 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

148x172mm (300 x 300 DPI)

Page 43 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

148x114mm (300 x 300 DPI)

Page 44 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

52x76mm (300 x 300 DPI)

Page 45 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

60x79mm (300 x 300 DPI)

Page 46 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

177x233mm (300 x 300 DPI)

Page 47 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

148x114mm (300 x 300 DPI)

Page 48 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

87x34mm (300 x 300 DPI)

Page 49 of 49

ACS Paragon Plus Environment

Biomacromolecules

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960