Adipose derived mesenchymal stem cells (ADMSCs)
Adipose tissue is also an abundant source of adult MSC, which also exhibit self-renew ability, plasticity and multilineage differentiation potential ability. Due to its abundant sources and easy accessibility for cell harvesting, ADMSCs has become an attractive cell source as alternative to BM-MSCs for regenerative medicine (Sterodimas et al. 2010). ADMSCs were seeded on 3D polycaprolactone fumarate (PCLF) and showed increased cell proliferation and expression of a ligamentous matrix (collagen and TNC) by the stimulation of platelet lysate and FGF-2 (E. R. Wagner et al. 2015). The suitability of ADMSCs for ACL tissue engineering under the growth factor treatment was investigated. Results showed that the expressions of Col I, Col III and TNC were increased when exposed to growth factors EGF and bFGF; while no significant difference was observed with TGF-β1 or IGF-1 at 4 weeks (Eagan et al. 2012).
Periodontal ligament is a specialized connective tissue aimed at connecting cementum and alveolar bone to support teeth and maintain the homeostasis. PDL-MSCs were first discovered by Seo et al in 2004 (Seo et al. 2004). It was reported that PDL-MSCs hold better proliferation ability than BM-MSCs: the population doubling of BM-MSCs is 50, while population doubling of PDL-MSC is over 100 (Shi et al. 2005). Previous studies revealed that PDL-MSCs possessed capacities to differentiate into osteoblast/cementoblast-like cells, osteogenic, adipogenic chondrogenic, neurogenic and angiogenic cells (Seo et al. 2004; Jinping Xu et al. 2008).
The application of PDL-MSCs for regenerative medicine was firstly reported for cementum/PDL-like tissue regeneration (Seo et al. 2004). PDL-MSCs were reported to possess greater potential for tendon regeneration than BM-MSCs and gingival MSCs (GMSCs) in vivo and in vitro (Moshaverinia et al. 2014). Results from the subcutaneous transplantation of PDL-MSCs encapsulated in TGF-β3-loaded RGD-coupled alginate microspheres into immunocompromised mice showed that PDL-MSCs expressed higher mRNA level of tendon-related gene expression (Scx, Dcn, TNMD, and Bgy) than BM-MSCs after 4 weeks, and in vivo histological examination showed the increased expression of tendon-related protein (Moshaverinia et al. 2014). PDL-MSCs exhibited better healing effect on tendon defect than Achilles tendon-derived cells (AT) with similar morphology as AT, advanced tissue maturation, less ectopic fibrocartilage formation, more organized collagen synthesis, and tendon specific matrix expression at 16 weeks after surgery. Thus, it confirmed the feasibility of PDL-MSCs application for tendon/ligament regeneration (Hsieh et al. 2016). Among these promising cell sources for regenerative medicine, WJ-MSCs and BM-MSCs were selected as cell sources for research because of its easy processability and stable donner supply since our lab had a collaboration with CHRU Hospital Nancy.
Ligament reconstruction is a sophisticated and highly-regulated process. In the process of tissue engineering, once an appropriate scaffold and cell source have been selected, tissue regeneration may point to supply biochemical and/or mechanical stimuli to promote ECM synthesis. For the biochemical stimulation, growth factors may be considered to promote tissue regeneration such as Insulin-like Growth Factor-I (IL-1), Vascular Endothelial Growth Factor (VEGF), Platelet-derived Growth Factor (PDGF), basic Fibroblast Growth Factor (bFGF), Transforming Growth Factor-β (TGF-β), and Growth Differentiation Factor 5 (GDF-5). Recently, platelet lysate (PL) consisting in a cocktail of growth factors has gained increased attractions for biomedical and clinical application. For the mechanical stimulation, the item bioreactor has been designed to support a mechanical stimulation environment and basically encourage the cells to proliferate and differentiation on the scaffold. These different solutions for providing biochemical and biomechanical stimuli will be detailed in the following sections.
Biomechanical stimulation with bioreactors
Bioreactors aim at providing a controllable environment to direct cell proliferation, differentiation, and neo-tissue formation towards a specific tissue (T. Wang et al. 2012). For ligament/tendon tissue engineering, a bioreactor should not only play a role of traditional incubator by supplying an appropriate environment (temperature, humidity, waste removal), a suited biochemical environment (pH, pO2, concentrations of nutrients
and growth factors) to support cell proliferation and matrix deposition, but should also provide scaffolds with the regulation of mechanical stimulation to promote ligament differentiation (Altman, Lu, et al. 2002). The stimulation of scaffolds within dynamic bioreactors mimicking the native physiological environment of the tissue may also permit to draw conclusions concerning the evolution of the properties of the construct when subject to loading, without demanding in vivo implantation. First proposed bioreactors provided uniaxial cyclic loading via a piston, and more recent bioreactors are capable of applying multiaxial and cyclic strains, such as 2-dimensional (2D) cyclic mechanical strain bioreactor, 3-dimensional (3D) dynamic shear and compression bioreactor. Recently, commercial ligament/tendon bioreactors as the Bose® ElectroForce® BioDynamicsy® system and the LigaGen system (Figure 6) were developed and expected to provide more accurate and complex environment for tissue regeneration (T. Wang et al. 2012). However, such bioreactors are limited by the complicated scaffold fixation or the type of mechanical stimulation (T. Wang et al. 2012).
Isolation and expansion of BM-MSCs and WJ-MSCs
Human bone marrow was harvested from patient’s femoral head during surgery and supplied by Orthopedic Surgery and traumatology center of CHRU Hospital (Nancy, France) with the informed consent of patients. Firstly, 25 mL α-MEM (Lonza, BE12-169F) complete medium contained 10% Fetal bovine serum (FBS) (Dominique Dutscher), 2 mM L-glutamine (Sigma, A-4034), 100 U/mL penicillin /streptomycin (Gibco, 15140-122) and 1μg/mL amphoterin B (Gibco, 15290-026) was added to 20 mL BM, and then centrifuged at 300 g for 5 min. The supernatant was removed, and 20 mL α-MEM complete medium was added, aspirated and blowed. 10 μL cell suspension was mixed with190 μL leucoplate for 3 min, red blood cells were lysed, and BM-MSCs were counted. Cells were seeded on tissue culture plates at the density of 50000 cells/cm2, then incubated in 37�, 5% CO2, 90% humidity for 3 days. The culture medium was changed twice a week.
Colony Forming Unit-fibroblast (CFU-F)
MSCs were detached from culture flask and seeded on petri dishes at 100 cells/ dish, then cultured in an incubator at 37°C, 5% CO2 and 90% humidity. MSCs were supplied with α-MEM complete medium for 14 days. At the end of day 14, 3 mL crystal violet was added to each petri dish and stained for 15 min, then dishes were washed with distilled water. The colonies petri dishes were recorded with a scanner (Scanner GS-800, BIORAD).
The typical cell surface epitope markers on MSCs at passage 2 (P2) were determined by flow cytometer (Beckman Coulter, Gallios AN24085). MSC (P2) were defreezed in 5 mL α-MEM complete medium, centrifuged at 400 rmp/min for 5 min, and then resuspended in 5 mL 0.5% PBS-BSA. Isotype FITC /PE were used as negative control, classical markers like CD 146, CD106, CD200, CD90, CD34, HLA-DR, CD73, CD45, CD166, CD44, and CD105 were detected. The antibodies related were shown in Table 4.
Silk and silk/PLCL multilayer braided scaffolds
Raw silk fibers (Bombyx mori, 20/22D) were supplied by Trudel Limited (Zurich, Switzerland). Prior to fabricate scaffold, silk was firstly collected on the bobbins, and boiled in 0.02 M Na2CO3 solution at 100 � for more than 1.5 h to remove sericin. Silk was then rinsed under running water overnight and finally washed with distilled water for 3 times.
16 bobbins of silk fibers were installed on the custom braided machine described previously to fabricate one-layer silk braided scaffold (S1), two-layers silk braided scaffold (S2), three-layers silk braided scaffold (S3) and Silk-Polymer (SP) three-layer composite scaffold made of a layer of PLCL between two layers of silk, with optimized and consistent speed. The ends of the scaffold were fixed with silk fiber and cut into 1 cm for the following study.
Scaffold modification with Layer-By-Layer (LBL) technology
Firstly, NaCl-Tris buffer (0.9% w/v NaCl in 10 mM Tris buffer) was prepared at pH 6.0-6.5 and pH 7.4 respectively. PLL (Sigma, P2636-1G) and HA (ACROS, 251770250) were dissolved in the NaCl-Tris buffer at the concentration of 1mg/mL. The scaffold pieces were firstly immersed in PLL solutions for 7 min, then gently rinsed in NaCl-Tris buffer for 1 min. They were subsequently introduced into HA solution for 7min and then rinsed in NaCl-Tris buffer for 1 min. This process was repeated (n=1, 3, 5) at the 2 different pH conditions, then ended with PLL solution to get scaffold-(PLL-HA) n-PLL (n refers to the number of bilayers “PLL-HA”). The term scaffold-blank (SB) refers to the scaffold without LBL modification, and scaffold-PLL refers to scaffold modified only by PLL solution. The names of all the scaffolds in this study are listed in Table 5.
Morphology by Scanning Electronic microscopy (SEM)
The multi-layer braided structure was observed by SEM. Scaffold were firstly dehydrated with a series of gradient alcohol solution (50°, 10 min; 70°, 10 min; 90°, 10 min; and 100°, 10 min for 2 times). Scaffolds were then coated with gold and viewed with SEM (JEOL 5400 LV).
Morphology by confocal laser macroscopy (CLSM)
PLCL fibers were cut into 1 cm-long pieces, PLL-FITC (Sigma, P3096) was dissolved in NaCl-tris buffer at pH 6.3 with the concentration of 1mg/mL. Fibers were then immersed in PLL-FITC/HA solution as described above, with the replacement of PLL with PLL-FITC solution, then washed with NaCl-tris buffer. Finally, fibers were observed with CLSM.
Table of contents :
1. Chapter I: State of art
1.1 General Introduction
1.2 Ligament structure
1.2.1 Gross structure
1.2.2 Microscopic organization
1.3 Ligament composition
1.3.1 Cellular component
1.3.2 Molecular components
1.4 Ligament function
1.5 Ligament biomechanics
1.6 Ligament injuries
1.6.1 Epidemiology of ligament injuries
1.6.2 Ligaments response to injury
1.6.3 Solutions for ligament repair
1.7 Ligament tissue engineering
1.7.1 Requirements for defining scaffolds for ligament tissue engineering
1.7.2 Biomaterial resources
1.7.3 Scaffold design
1.7.4 Cell sources for ligament tissue engineering
1.8 Biomechanical/Biochemical stimulation
1.8.1 Biochemical stimulation
1.8.2 Biomechanical stimulation with bioreactors
1.9.1 Summary, issues and hypothesis
Objective and content of the thesis
2 Chapter II : Materials and methods
2.1 Biological study of MSCs
2.1.1 Isolation and expansion of BM-MSCs and WJ-MSCs
2.1.2 Colony Forming Unit-fibroblast (CFU-F)
2.2 Scaffold fabrication
2.2.1 PLCL multilayer braided scaffolds
2.2.2 Silk and silk/PLCL multilayer braided scaffolds
2.3 Scaffold modification with Layer-By-Layer (LBL) technology
2.3.1 Scaffold with PLL/HA modification
2.3.2 Fibers modification with PLL-FITC/HA
2.4 Scaffold characterization
2.4.1 Structure by Fourier transform infrared spectroscopy (FTIR)
2.4.2 Morphology by Scanning Electronic microscopy (SEM)
2.4.3 Morphology by confocal laser macroscopy (CLSM)
2.4.4 Topology by atomic force microscopy (AFM)
2.4.5 Mechanical properties
2.4.6 Morphology and Porosity by micro computed tomography (μCT)
2.5 Evaluation of biocompatibility of MSCs on scaffolds
2.5.1 Scaffold sterilization
2.5.2 Cell seeding on the scaffold
2.5.3 Cell proliferation evaluated by AB test
2.5.4 Cell location and morphology detection by SEM
2.5.5 Cell morphology observation by CLSM and fluorescent microscopy
2.5.6 Live/dead staining of MSCs on scaffolds
2.5.7 Detection of Extracellular matrix synthesis by fluorescent microscopy
2.5.8 Cell migration stimulated by scaffolds
2.5.9 Histology and immunohistochemistry of MSCs on scaffolds
2.5.10 Bioreactor sterilization and assembly
2.5.11 MSCs seeded on scaffolds with mechanical stimulation and biological evaluation
2.6 Conclusion of the present chapter
3 Chapter III : Biological study of cell sources selected for ligament tissue engineering
3.2 Results and discussions
3.2.1 Biological characteristics of WJ-MSCs and BM-MSCs
4 Chapter IV : Surface modification of a braided multilayer PLCL scaffold for ligament tissue engineering
4.2 Results and discussion
4.2.1 Physicochemical properties of PLCL
4.2.2 Biocompatibility of MSCs on PLCL scaffold
5 Chapter V : Proposition of silk and silk/PLCL scaffold for ligament tissue engineering
5.2 Results and discussion
5.2.1 Physiochemical properties of silk-based braided scaffold
5.2.2 Biological properties of silk-based braided scaffold
6 Chapter VI : An attempt to study effect of dynamic mechanical stimulation on MSCs-construct in a tension-torsion bioreactor
6.2 Results and discussion
6.2.1 MSCs metabolic activity of MSCs on scaffold
6.2.2 MSCs morphology and location on scaffolds
7 Chapter VII: Discussion
7.1 Initial PLCL braided scaffold
7.2 PLL/HA LBL modification
7.3 Limitations of the initial PLCL braided scaffold
7.4 Silk as an alternative to PLCL
7.5 Proposition of a silk-based braided scaffold
7.7 Preliminary study of the effect of mechanical stimulation on MSC-scaffold differentiation
8 Chapter VIII: Conclusion, limitation and perspectives
8.1 Conclusion of the present work
8.2 Limitations of the proposed scaffolds and required further characterization
8.2.1 Biodegradation properties
8.2.2 In vivo implantation
8.2.3 Quantitative characterization
8.3 Perspectives of the present work