Design of Hyaluronic Acid Hydrogels to Promote Neurite Outgrowth in

Subscriber access provided by Northern Illinois University

Article

Design of hyaluronic acid hydrogels to promote neurite outgrowth in three dimensions Dominte Tarus, Lauriane Hamard, Flavien Caraguel, Didier Wion, Anna Szarpak-Jankowska, Boudewijn van der Sanden, and Rachel Auzely-Velty ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b06446 • Publication Date (Web): 06 Sep 2016 Downloaded from http://pubs.acs.org on September 7, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Design of hyaluronic acid hydrogels to promote neurite outgrowth in three dimensions Dominte Tarus, ‡,¶ Lauriane Hamard, §,¶ Flavien Caraguel, § Didier Wion, † Anna SzarpakJankowska, ‡ Boudewijn van der Sanden*,§ and Rachel Auzély-Velty*,‡ ‡

Grenoble Alpes University, Centre de Recherches sur les Macromolécules Végétales

(CERMAV-CNRS), 601, rue de la Chimie, BP 53, 38041 Grenoble Cedex 9, France §

Platform Intravital Microscopy, France Life Imaging, Grenoble Alpes University, INSERM

U1205, 17 rue des Martyrs, 38054 Grenoble, France †

Grenoble Alpes University, INSERM U1205, 17 rue des Martyrs, 38054 Grenoble, France

KEYWORDS: Hyaluronic acid, hydrogel, neural progenitor cells, two-photon microscopy, neurite outgrowth

ABSTRACT: A hyaluronic acid (HA)-based extracellular matrix (ECM) platform with independently tunable stiffness and density of cell-adhesive peptide (RGD, arginine-glycineaspartic acid) that mimics key biochemical and mechanical features of brain matrix has been designed. We demonstrated here its utility in elucidating ECM regulation of neural progenitor cell behavior and neurite outgrowth. The analysis of neurite outgrowth in 3-D by two-photon

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 32

microscopy showed several important results in the development of these hydrogels. First, the ability of neurites to extend deeply into these soft HA-based matrices even in the absence of celladhesive ligand further confirms the potential of HA hydrogels for central nervous system (CNS) regeneration. Secondly, the behavior of hippocampal neural progenitor cells differed markedly between the hydrogels with a storage modulus of 400 Pa and those with a modulus of 800 Pa. We observed an increased outgrowth and density of neurites in the softest hydrogels (G’ = 400 Pa). Interestingly, cells seeded on the surface of the hydrogels functionalized with the RGD ligand experienced an optimum in neurite outgrowth as a function of ligand density. Surprinsingly, neurites preferentially progressed inside the gels in a vertical direction, suggesting that outgrowth is directed by the hydrogel structure. This work may provide design principles for the development of hydrogels to facilitate neuronal regeneration in the adult brain.

1. INTRODUCTION Brain injury, as occurs in stroke, often results in tissue loss and subsequent formation of a lesion cavity. Although the body attempts to repair itself by proliferation and neuronal differentiation of endogenous neural stem or progenitor cells (NSCs or NPCs),1 functional recovery is limited, especially in adults2-3. Therefore, significant effort is underway to develop therapies to improve functional recovery. In this regard, an increasing effort has been put in the design and development of hydrogels to facilitate neuronal regeneration. The benefit provided by hydrogels may partly be attributed to their mechanical properties that can be tailored to mimic native brain tissue.4-7 This offers a means to modulate and control neural stem cell behavior. Indeed, it was reported that in the case of hydrogels based on synthetic or natural polymers


2

Page 3 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(poly(acrylamide), chitosan), NSCs/NPCs preferentially differentiate into neurons on very soft substrates (elastic modulus (E) < 1 kPa) yet into astrocytes on stiffer substrates (E > 7 kPa).6, 8 The mechanical properties of hydrogels were also reported to influence neural cell differentiation in three-dimensional (3-D) cultures. It was found for hyaluronic acid (HA) hydrogels that the majority of NPCs cultured in scaffolds with similar compressive modulus to that of neonatal brain tissue differentiated into neurons with extended long processes, while those cultured in hydrogels with similar compressive modulus to that of adult brain tissue (being more than twice that of neonatal brain tissue) mostly differentiated into astrocytes.7 Hyaluronic acid is a major extracellular matrix component in the fetal mammalian brain.9-10 In this matrix, the cell-surface receptor for HA, CD44 as expressed in NPCs, can act as a mechano-transduction sensor without cell adhesion motif derived from natural proteins of the extracellular matrix (ECM).11-12 This result, together with previous reports on the significant effect of HA on the migration of neurons13 and astrocytes14 indicate that HA itself plays a key role in central nervous system (CNS) development. To further enhance axonal regeneration, several strategies for the chemical modifications of HA were implemented.15 In this regard, ECM proteins together with their peptide motifs, namely, laminin, RGD and IKVAV were covalently linked to HA hydrogels. HA hydrogels modified with laminin improved neuronal regeneration mainly by reducing reactive astrocytes gathering around the lesion boundary and promoting new fibre formation within the scaffolds.16 RGD modification was demonstrated to promote cell migration into implants, resulting in the formation of collagen-like bundles and neurofibrils in the hydrogel implants.17 IKVAV modification was shown to have similar effects.18 Although these results are promising, the extent to which ligand density influences neurite outgrowth was not explored. However, it was found that this may be a critical regulator of neurite outgrowth rate. Indeed, using fibrin


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 32

matrices that contained variable amounts of RGD peptides, Hubbell et al. demonstrated that both 2-D and 3-D neurite outgrowth showed a near Gaussian dependence on RGD concentration, with intermediate RGD adhesion site densities leading to the fastest outgrowth.19 Therefore, to get a better understanding of the relative and combined effects of stiffness and biochemical characteristics of HA hydrogels on neuronal regeneration, a systematic approach to the design of HA matrices with independently tunable crosslinking and biochemical ligand density is required. The aim of the study was to optimise adhesion and proliferation of exogenous neural progenitor cells and neurite outgrowth in hyaluronic acid hydrogels during neuronal differentiation by modifying the shear elastic modulus as well as RGD ligand density (Figure 1). To this end, thiolene chemistry was applied for the sequential bioconjugation and crosslinking of HA,20 affording hydrogels of which the bioactive and mechanical properties can be finely manipulated. When neural progenitor cells were seeded on the HA gels, these scaffolds appeared to promote neuronal differentiation and 3-D neurite outgrowth into the gels as observed by two-photon microscopy. Through the use of an in-house-designed, semi-automated image analysis method to quantify neurite density in the gels, we show that the mechanical and cell-adhesive biochemical parameters of HA hydrogels both have profound effects on neurite outgrowth. It was thus determined that HA hydrogels modified with RGD ligand could promote cell adherence and enhance neurite outgrowth, suggesting that it may become a useful therapeutic approach for CNS injury.


4

Page 5 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Synthesis of hyaluronic acid-based hydrogels with independently tunable crosslinking and RGD ligand density using thiol-ene photochemistry and their use as biomimetic platforms for 3-D culture of HNPCs.

2. MATERIALS AND METHODS 2.1. Materials. Sodium hyaluronate (HA) with a molar mass Mw of 400 kg/mol (HA400) was supplied by HTL Biotechnology (France). The molecular weight distribution and the weightaverage molecular weight of this HA sample were determined by size exclusion chromatography using a Waters GPC Alliance chromatograph (USA) equipped with a differential refractometer and a light scattering detector (MALLS) from Wyatt (USA); the solutions were injected at a concentration of 5 × 10-4 g/mL in 0.1 M NaNO3. The polydispersity index (PDI) of the sample is Mw/Mn ~ 2. The GRGDS (Gly-Arg-Gly-Asp-Ser) peptide functionalized with mercaptopropionic acid with 77 % purity (from manufacturer HPLC analysis) was obtained from GeneCust Europe (Luxemburg). Poly(ethylene glycol)-bis(thiol) (Mn = 3400 g/mol) was purchased from PEGWorks (USA). Irgacure 2959 was kindly provided by Ciba Specialty Chemicals (Basel, Switzerland). N-Hydroxysulfosuccinimide sodium salt (Sulfo-NHS) was purchased from


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 32

Chemrio (Ningbo Zhejiang, P.R. CHINA). Bovine testes hyaluronidase, pentenoic anhydride, 2(N-morpholino)ethanesulfonic acid (MES), phosphate buffer saline (PBS, pH 7.4), N-ethyl-N'(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC), (3-aminopropyl)trimethoxysilane (APTS) and other chemicals were purchased from Sigma-Aldrich-Fluka and were used without further purification. The water used in all experiments was purified by an Elga Purelab2 purification system, with a resistivity of 18.2 MΩ cm. 2.2. Synthesis of HA-p. HA (0.30 g, 0.75 mmol) was dissolved in ultrapure water (15 mL) at 4 °C and the resulting mixture was kept at 4 °C under continuous stirring overnight for complete dissolution. DMF (10 mL) was then added dropwise in order to have a water/DMF ratio of (3:2, v/v). Pentenoic anhydride (0.136 g, 0.75 mmol) was added while maintaining the pH between 8 and 9 (by adding 1 M NaOH) for 4 hours at 4 °C. The reaction was kept at 4 °C under continuous stirring for one night. After this time, NaCl was added to the reaction mixture to have a NaCl concentration of 0.5 M. The polymer was precipitated by addition of ethanol (water/ EtOH (v/v) ratio of 2:3). After removal of the supernatant, the precipitate was successively washed with mixtures of water/EtOH (1:4, 1:9, v/v) and finally dissolved in ultrapure water for a final purification by diafiltration (ultramembrane Amicon YM10) with ultrapure water. The purified product was recovered by freeze-drying and characterized by 1H NMR spectroscopy. The DS was determined to be 25 %. 2.3. Synthesis of HA-GRGDS derivatives. HA-GRGDS (DS = 5 %) was prepared by dissolving HA-p (0. 15 g, 0.36 mmol) in ultrapure water at a concentration of 10 g/L. 75 µL of an aqueous solution of Irgacure 2959 (10 mg/mL) was then added to the polysaccharide solution to obtain a final photoiniator concentration of 5 % (w/v), followed by HS-CH2-CH2-CONHGRGDS (0.0159 g, 0.021 mmol). The mixture was exposed to UV light (λ= 365 nm) with an


6

Page 7 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


intensity of 20 mW/cm2 for 5 min under stirring. The modified HA was purified by diafiltration (ultramembrane Amicon YM 100) with ultrapure water and was recovered by freeze-drying. The chemical integrity and purity of the final product were checked by 1H NMR. Digital integration of the NMR signals arising from the H2 protons of HA and the significant peptide protons gave a DS of 5 %. The syntheses of the HA-GRGDS derivatives with DS = 3 and 7 % were performed using a similar procedure but by modifying the amount of the GRGDS peptide. The amount of peptide added was 0.0080 g (0.011 mmol) and 0.0185 g (0.032 mmol) to synthesize the derivatives with DS of 3 and 7 %, respectively. 2.4. NMR spectroscopy. The 1H NMR spectra were recorded on a Bruker AVANCE III HD spectrometer operating at 400 MHz with deuterium oxide (D2O from SDS, Vitry, France) as a solvent. Chemical shifts (δin ppm) are given relative to external tetramethylsilane (TMS = 0 ppm) and calibration was performed using the signal of the residual protons of the solvent as a secondary reference. 2.5. In situ monitoring of gel formation by photorheometry. An AR2000Ex rheometer (TA Instruments Inc.) fitted with a UV-curing cell (λ = 365 nm) and an aluminum plate (diameter 20 mm) was used for the in situ monitoring of gel formation of HA-p/PEG and HA-GRGDS/PEG mixtures at 25 °C. Following deposition of 300 µL of a mixture of HA-p or HA-GRGDS, PEG(SH)2 and photoinitiator, the gap between the flat quartz plate and the aluminum plate was initially 0.85 mm (measuring ambient temperature). It was controlled during the experiments by maintaining the normal force at 0 ± 0.1 N.


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 32

The oscillatory time sweep experiments were carried out during a period of 22 min at a fixed frequency of 1 Hz and a fixed deformation of 3.5 %. Typically, after deposition of the solution of the mixture of polymers in PBS between the plates and equilibration for 1 min, the solution was exposed to light (λ = 365 nm) for 22 min at a fixed light power (20 mW/cm2) leading to gelation. All measurements were done in triplicate. 2.6. Scanning electron microscopy. The HA/PEG hydrogels prepared as described above were frozen at -20 °C for 12 h and then freeze-dried. The samples were coated by approximately 2 nm of sputtered Au-Pd. They were observed in secondary electron imaging mode with a ZEISS Ultra 55 FEG-SEM (Grenoble INP - CMTC). Images were acquired at low voltage 3 kV using the Everhart-Thornley detector to accentuate the topography. 2.7. Swelling studies. Hydrogels prepared as above were swollen in PBS for 1 day at room temperature. The estimated equilibrium swelling ratio (SR) was defined as the ratio of the mass of the swollen hydrogel to that of the dry polymer obtained after drying at 40 °C in an oven for 48 h. 2.8. In vitro degradation of hydrogels by hyaluronidase. Hydrogels with storage moduli of 400 and 800 Pa were prepared as described above from precursor solutions (400 µL) containing I2959, HA-p, and PEG-(SH)2. The precursor solutions were transferred into 12 mm diameter, 4 mm deep molds and exposed to UV light (20 mW/cm2, λ = 365 nm) for 20 min. The resulting hydrogel cylinders (diameter = 12 mm, height = 3 mm) were cut into four pieces which were accurately weighed in order to determine the amount of HA contained inside. Each hydrogel sample was suspended at 37 °C in PBS (10 mL). Degradation experiments were performed in vitro in 50 U/mL bovine testes hyaluronidase in PBS at 37 °C. At regular time points (1 h), an


8

Page 9 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


aliquot of the incubation buffer medium (200 µL) was removed and replaced. The amount of HA released during degradation was measured using a previously established carbazole assay21. The total amount of HA released from the hydrogels was calculated from the concentrations of the different aliquots. The dilution of the release medium caused by sampling was taken into account using the following equation22:

= +

∗ ∑

(eq. 1)

where, CHA RM = true concentration of HA in the release medium CHA sample n = HA concentration in a sample taken at a selected time Vsample = sample volume VRM = total volume of the release medium at a selected time Csample = HA concentration of the previous samples

2.9. Analysis of the diffusion of fluorescently-labeled latex nanoparticles into the HA hydrogels. For these experiments, nanoparticles (NPs) with three different sizes were tested: 100 nm, 250 nm and 500 nm. The HA hydrogel samples with a G’ value of 400 and 800 Pa were synthesized and swollen in PBS till equilibrium swelling (12 h). Each hydrogel cylinder (diameter: 6 mm, height: 2.7 mm) was cut into two equally-sized pieces. The pieces were immersed in PBS (3 mL) and 200 µL of a solution of fluorescein-labeled latex nanoparticles (Micromer GreenF (surface: COOH; concentration: 10 mg/mL), Micromod Partikeltechnologie GmbH, Germany) were added. After stirring at room temperature for 2 or 7 days using an orbital


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 32

stirrer, the hydrogel samples were washed with fresh PBS and the soaking solutions were kept for fluorescence analysis. The washed hydrogel samples were weighed and then treated with 0.1 M NaOH (1 mL) to hydrolyze the crosslinked network and release the nanoparticles. The fluorescence of the resulting samples was measured using a Perkin Elmer LS 50B luminescence spectrometer (λexc = 475 nm and λem = 514 nm). A volume of soaking solution equal to the hydrogel sample was diluted with 0.1 M NaOH (1 mL) and was used as a reference to assess particle diffusion inside the gels. The penetration of the nanoparticles was calculated as the fluorescence ratio between the hydrogel and the soaking solution intensity. There were no significant differences in the results obtained with NPs of 100 and 250 nm between day 2 and day 7, which indicates that a two-day period is sufficient for NP to penetrate the gels. The NPs of 500 nm did not penetrate the gels. The penetration of NP in the gels was additionally observed by two-photon laser scanning microscopy (TPLSM). The hydrogels were covalently immobilized during crosslinking on glass coverslips as described below (section 2.10). The HA hydrogels were prepared by deposition of the gelling mixture (HA-p, PEG(SH)2, and Irgacure 2959 in PBS (300 µL)) on the HA-pmodified coverslip to which a teflon ring (diameter: 20 mm) was fixed. The mixture was then exposed to UV light (20 mW/cm2, λ = 365 nm) for 20 min. The resulting hydrogels with an elastic modulus of 400 and 800 Pa, fixed on the coverslip and having a thickness of ∼0.8 mm were labeled with RhodamineB-PEG-thiol (RB-PEG-SH, Mw= 3 kg/mol, Nanocs Inc, USA) using a thiol-ene reaction, in order to generate contrast between the surrounding medium and the gel. To this end, the gels were immersed in PBS (2 mL) containing the dye (0.05 mg/mL) and Irgacure 2959 (20 µL, 10 g/L), and irradiated at 20 mW/cm2 for 2 min. After 3 washings with PBS for 24 h in total, a PBS solution (3 mL) and 200 µL of a solution of fluorescein-labeled


10

Page 11 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


latex nanoparticles (NPs of 100 and 250 nm, concentration: 10 mg/mL) was added. The hydrogels were incubated at room temperature for 48 h on an orbital stirrer at 60 rpm, and then observed by TPLSM using a LSM 7 MP microscope (ZEISS, Jena, Germany) equipped with a 20x water-immersion objective (NA 1.0, maximum spatial resolution is 100 nm) and ZEN 2010 software. 2.10. Hydrogel Immobilization. For the cell culture experiments, hydrogels were covalently linked to a glass coverslip during crosslinking. The glass coverslip was first modified with APTS as previously described.23 Then, the modified coverslip (d = 20 mm) with −NH2 groups was immersed in PBS (1 mL) containing HA-p (3 g/L). EDC (50 µL, 50 g/L) and sulfo-NHS (50 µL, 30 g/L) were added. The coverslip was gently stirred at room temperature for 4 h and then, rinsed with ultrapure water and dried under a nitrogen flow before use. The HA-based hydrogels were prepared by deposition of the gelling mixture (HA-p or HA-GRGDS, PEG(SH)2, and Irgacure 2959 in PBS (300 µL)) on the HA-p-modified coverslip to which a teflon ring (diameter: 20 mm) was fixed. The mixture was then exposed to UV light for 20 min at a fixed light power (20 mW/cm2, λ = 365 nm). Under such conditions, hydrogels with different stiffness (G' modulus of 400 or 800 Pa) were prepared. The resulting hydrogels fixed on the coverslip and having a thickness of ∼0.8 mm were then transferred to sterile PBS, before seeding with HNPCs. 2.11. Extraction and cell culture of multi-potent hippocampal neural progenitor cells. In all experiments, multi-potent hippocampal neural progenitor cells (HNPCs) were freshly prepared from the embryonic mice hippocampus (CD1 mice at 14 days post-fertilization, Charles River, Saint Germain sur l'Arbresle France). Their capacity to differentiate into neurons (anti-β3 tubulin staining) and astrocytes (anti-GFAP) was tested in parallel to the all experiments using AntiNeuronal Class III β-tubulin (Stem Cell, catalogue number 01409, France) and Anti-Polyclonal


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 32

rabbit Anti-Glial Fibrillary Acidic Protein (Dako France SAS, catalogue number Z0334). This was tested on poly(D-lysine) coated slides (Figure S1) before plating the hippocampal neural progenitor cells on the HA hydrogels. In all experiments, the differentiation of HNPCs in neurons (β3 tubulin staining) and astrocytes (GFAP staining) was observed. For the different series of experiments, the plating density of the HNPCs on the surface of the HA gels was 35000 cells/cm2 (n= 3 for each hydrogel condition in 3 independent experiments). The HNPCs were incubated

for

analysis

of

neurites

outgrowth

using

two-photon

microscopy

after

immunohistochemical staining of the neuronal β3 tubulin and astrocytes during maximum 21 days. The incubation was performed with standard culture medium Dulbecco’s Modified Eagle Medium (DMEM) containing 5 % Fetal Bovine Serum (FBS) in order to stimulate cell adherence during two hours. Then, this medium was replaced by Neurobasal™ medium/B27/Glutamax supplement to support the neurites growth at 5 % CO2, 20 % O2 at 37 °C. Neurobasal medium was changed every 2-3 days. At D17 or D21 after plating, the HA gels were fixed for immunohistochemical and two-photon microscopic analysis between D17 and D21. 2.12. Immunohistochemistry. After maximum 21 days, Neurobasal medium was replaced by a 4 % paraformaldehyde solution for cell fixation for 2 h. After rinsing two times with PBS, PBS containing 0.5 % triton X-100 (Sigma-Aldrich) and 5 % FBS were added for 2 h to permeabilize the cell membranes. The following immunostaining was applied on each hydrogel to detect neurons and astrocytes. Samples were incubated with primary antibody overnight at room temperature. After thoroughly washing in PBS, samples were incubated in secondary antibody for 2 h. Primary antibodies against ß-III tubulin (monoclonal anti-mouse β3-tubulin, Stem Cell) and glial fibrillary acidic protein (polyclonal anti-rabbit GFAP) were used. Polyclonal goat anti-mouse IgG conjugated to


12

Page 13 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


FITC (Sigma, catalogue number F5897) and polyclonal donkey anti-rabbit conjugated to Alexa 647 (Life Technologies, catalogue number A31573) were used as secondary antibodies. In most protocols, all nuclei were stained with Hoechst 33342 (Sigma-Aldrich, France). 2.13. Two-photon microscopy. TPLSM was performed using a LSM 7 MP microscope (ZEISS, Jena, Germany) equipped with a 20x water-immersion objective (NA 1.0, maximum spatial resolution is 100 nm) and ZEN 2010 software. For all experiments, the excitation wavelength of the pulsed infrared laser (Ti:Sapphire, Chameleon Vision II; Coherent, UK) was 800 nm and fluorescence emissions were epi-collected by three photomultiplier tubes with a short pass filter < 480 nm (Semrock, USA) for the blue channel, a 542/50 nm filter (Semrock, USA) for the green channel and a 617/73 nm filter (Semrock, USA) for the red channel. For 3 – 4 volumes of interest per gel, z-stack acquisitions were performed using the whole gel thickness with a field of view of 600 × 600 microns (512 × 512 pixels) and a minimum step-size of 2 microns. Image processing was performed with NIH ImageJ software. 2.14. 3-D analyses of the mean neurites density. In this study, only 3D neurites growth in the gels (and not on the surface) was analyzed, because their capacity to grow inside the gels is important for 3D neuroregeneration in vivo applications. Neurites semi-automatic segmentation was performed by a random forest classification method implemented in Ilastik software. The classifier was trained to detect neurites using images pixel intensity features. We specifically developed an imageJ Java plugin to quantify neurites growth in the gels by counting the number of detected neurites per focal plane in a z-series. The neurites were quantified by counting every detected object within an area from 1 to 20 pixels Objects of more than 20 pixels were considered as artifacts and automatically removed from the analysis.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 32

3. RESULTS AND DISCUSSION 3.1. Synthesis of soft HA/PEG scaffolds and analysis of the effect of their mechanics on NPC phenotype by two-photon microscopy. The development of biomaterials for neural regeneration applications requires fabrication of substrates with an elastic modulus (E) in the range of ~ 0.5-1 kPa.8, 24-25 In a previous study, we demonstrated the synthesis of dextran-based hydrogels with tunable storage moduli (G') ranging from 2 to 10 kPa26 (corresponding to a range of young moduli (E) of ~ 6-60 kPa, according to E = 2G’ (1 + ν) and assuming a Poisson’s ratio (ν) of 0.511). These hydrogels were prepared using a radical thiol-ene reaction to crosslink the polysaccharide modified with alkene (pentenoate) groups with poly(ethylene glycol)-bis(thiol) (PEG-(SH)2) as a crosslinker. The storage modulus was tuned by varying the [SH]/[alkene group] ratio. As neural stem cells preferentially differentiate into neurons when cultured on soft materials, we adjusted here the conditions of the HA/PEG gel synthesis to obtain substrates with low stiffness (G’ < 1 kPa). The UV-induced crosslinking reaction performed from a HApentenoate (HA-p) derivative having a degree of substitution of 25 % (DS, average number of substituents per 100 repeating units, derived from 1H NMR analysis, see Figure S2) and PEG(SH)2 was monitored in situ by photorheometry. The time required for the G’ modulus to reach a plateau, indicating the end of the gelation11 was 20 min (Figure S3). By varying the [SH]/[alkene group] ratio from 0.1 to 1.5, hydrogels with G’ values ranging from 380 to 5965 Pa could be obtained (Figure 2). As seen from Figure 2, above a [SH]/[alkene group] ratio of 1.2, the G’ values decreased, indicating a decrease of effective interchain cross-links, likely due to the thiolene coupling of PEG-(SH)2 with HA through only one extremity. Such a reaction is indeed promoted when PEG-(SH)2 is in excess, as reported in other systems27.


14

Page 15 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Control of the storage moduli (G’) of HA-p hydrogels by varying the thiol to alkene group ratio ([SH]/([=]). The values correspond to the steady-state G’ value measured after 20 min of UV irradiation at 20 mW/cm2. Here, because the maximum of G' was observed for a [SH]/[alkene group] ratio of 1.2, it is apparent that the crosslinking of two HA-p chains by one PEG-(SH)2 does not proceed with 100 % efficiency. Nevertheless, as this maximum is close to a [SH]/[alkene group] ratio of 1, this indicates that HA crosslinking is far the most important reaction that occurs between HA-p and PEG-(SH)2. The linear relationship between G' and [SH]/[alkene group] ratio up to 1.2 allowed us to define the crosslinking conditions leading to the formation of hydrogels with desired mechanical properties. From Figure 2 (zoom), it can be observed that soft HA gels (G’ < 1 kPa) can be obtained using [SH]/[alkene group] ratios in the range 0.1-0.2, resulting in shear moduli ranging from 387 to 920 Pa. As expected, the hydrogels could be degraded in vitro by hyaluronidase. By comparing the degradation of gels with storage moduli of 400 and 800 Pa by the bovine testicular hyaluronidase (used at a concentration of 50 U/mL), it was found that the time required to reach 95 % degradation (derived from the quantification of the HA released from the gel) was 10 h and 16 h for the 400 Pa and 800 Pa gels, respectively (Figure S4). These


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 32

differences in the degradation of gels correlate with the different crosslinking densities of the gels. In order to test how neurons behave in our 3-D tissue model, we cultured hippocampal neural progenitor cells (HNPCs) on HA gels with storage moduli of 400 and 800 Pa and analyzed their behavior in term of differentiation and neurite outgrowth by two-photon microscopy after immunohistochemical staining of the neuronal β3 tubulin and the intermediate filament Glial Fibrillary Acidic Protein (GFAP) in astrocytes at D21. The HNPCs formed neurospheres at the surface of the gels from which neurites emerged at the surface and extended into the matrix in a vertical direction (Figure 3A,B). Immunostaining demonstrated that the neurite outgrowth was positive for the neuronal marker β-tubulin, but little or no GFAP positive staining of astrocytes was detected at D21 after plating. At this time, the neurite density appears to be higher in the softest hydrogel (G’ = 400 Pa). In this gel, neurites were observed until 1000 µm inside the matrix. Quantification of the neurite outgrowth inside the gels using image analysis showed statistically significant differences between both hydrogels. Previous quantitative methods used to evaluate neurite outgrowth were not well-suited to characterize the complex 3-D neurite extension in our hydrogels. Parameters like neurite length and number of neurites per neuron cannot be estimated,28-29 because the gel volumes were heterogeneous and change over time. Indeed, the HA hydrogels can be dynamically remodeled and degraded by cellular hyaluronidase, resulting in gel volume changes during cell culture. In a previous study, it was reported that NPCs began to degrade HA-based hydrogels during the three-week culture period and that the majority of hydrogels completely degraded by the fourth week.7 Moreover, most neuronal cell bodies were located within the neurospheres and could not be quantified by two-photon microscopy, because the diffusion of photons is too high in these


16

Page 17 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


spheres. Therefore, we analyzed the mean neurite density in the gels using the smallest volume that contain neurites at D21 (bleu transparent box in Figure 3C) as reference for the whole series including gel moduli of 400 and 800 Pa. As a consequence, neurites at larger depths than the smallest volume in the series were not taken into account for the calculation of the mean neurite density. Moreover, in depth analysis underneath large neurospheres at the gel surface were avoided, because they absorbed the fluorescent emission of neurites underneath these spheres (see pink cylinder in Figure 3C). After all previous corrections, the neurite density (number/ mm2) was calculated for every focal plane in the z-series (3 – 4 per gel) with a stepsize of 2 microns. Finally, the mean neurite density ± the SD per z-stack, as shown in Figure 3D, was calculated for every z-series.


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 32

Figure 3. A) and B) 3D two photon microscopy images of neurites outgrowth (β3 tubulin staining) at D21 after plating of HNPCs on the surface of HA/PEG hydrogels with a storage modulus of 400 Pa (A) and 800 Pa (B); top: z-projections (top view, bar is 100 µm) of mean fluorescence intensities; bottom: side views (y-projections of mean fluorescence intensities). C)


18

Page 19 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Method for the calculation of the mean neurite density in the gels using the smallest volume that contain neurites at D21 (transparent blue box) as reference for the whole series including moduli of 400 and 800 Pa. In depth analysis underneath large neurospheres at the gel surface were avoided, as they absorbed the fluorescent emission of neurites underneath these spheres (see pink cylinder). D) Analyses of the mean neurites densities at D21 in the 400 and 800 Pa gels. Left: every point in the graph represents the mean neurites density for one z-stack of interest; 3-4 zstacks per gel were analyzed for 4 different gels per elastic modulus. The horizontal line is the mean neurites density. Right: mean neurites density values and SD: 400 Pa: 394 ± 193, N=27, 800 Pa: 47 ± 42, N=32. The difference between the mean neurites density values was highly significant: p

Design of Hyaluronic Acid Hydrogels to Promote Neurite Outgrowth in

Recommend Documents