Influence of perfusion and compression on the proliferation and differentiation of bone mesenchymal stromal cells seeded on polyurethane scaffolds

Influence of perfusion and compression on the proliferation and differentiation of bone mesenchymal stromal cells seeded on polyurethane scaffolds

Biomaterials 33 (2012) 1052e1064 Contents lists available at SciVerse ScienceDirect Biomaterials journal homepage: www.elsevier.com/locate/biomateri...

3MB Sizes 1 Downloads 81 Views

Biomaterials 33 (2012) 1052e1064

Contents lists available at SciVerse ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Influence of perfusion and compression on the proliferation and differentiation of bone mesenchymal stromal cells seeded on polyurethane scaffolds Chaoxu Liu a, b, *, Reza Abedian c, Roland Meister a, Carl Haasper a, Christof Hurschler c, Christian Krettek a, Gabriela von Lewinski d, Michael Jagodzinski a a

Department of Orthopedic Trauma, Hanover Medical School (MHH), OE 6230 Carl-Neuberg-Straße 1, D-30625 Hanover, Germany Department of Orthopedics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Jiefang Dadao 1095, 430030 Wuhan, PR China Laboratory for Biomechanics and Biomaterials, Department of Orthopedics, Hanover Medical School, Anna-von-Borries-Straße 1-7, D-30625 Hanover, Germany d Department of Orthopedics, Hanover Medical School, Anna-von-Borries-Straße 1-7, D-30625 Hanover, Germany b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 August 2011 Accepted 17 October 2011 Available online 5 November 2011

In the present study, a porous meniscal-shaped scaffold consisting of polyurethane (PU)-based 1, 4-butanediisocyanate (BDI), which provided a 3-D culture condition for human bone mesenchymal stromal cells (hBMSC) was employed. A bioreactor was utilized to produce perfusion and mechanical stimulations. The viability, proliferation and fibro-cartilaginous differentiation of the hBMSC cultured on the PU-based meniscal scaffold were investigated during the perfusion and mechanical stimulation process. In addition, the mechanical properties of the cell-laden scaffolds were examined as well. Our finding indicated that the perfusion (10 ml/min) and on-off cyclic compressions mechanical stimulation (10% strain, 0.5 Hz, 4 times/day, 2 h/time with 4 h of rest thereafter) maintained the viability and promoted the proliferation of hBMSC over 2 weeks. The on-off cyclic compression caused a 1.85 fold increase in equilibrium modulus. Meanwhile, type I procollagen produced by hBMSC was increased for 3.02-fold after 2 weeks culture. On the other hand, the irrigating medium enhanced the synthesis of type III procollagen for 2.24-fold after 2 weeks. Tensile modulus was elevated for 2.02-fold in perfusion group after 1 week, which was decreased after 2 weeks unexpectedly. Our study suggests that the perfusion and on-off compression are promising to enhance the functional properties of the hBMSC-laden PUbased meniscal scaffold. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Polyurethane scaffold Mesenchymal stromal cells Physical stimulations Mechanical properties Cytotoxicity Meniscus tissue engineering

1. Introduction The menisci of the knee joint are made of a pair of crescentshaped fibro-cartilaginous tissues, which play a key role in directing the load transmission and restraint on the knee joint. Injury or deficiency of meniscus can cause the osteoarthritis and an irreversible joint damage [1]. The meniscus is only vascularized in the outer one-third zone (red zone), so that the lesions in the inner two-third zone (white zone) are rarely healed spontaneously. Partial or total removal of the torn meniscus, the so-called meniscectomy, has been accepted as the standard therapy in the past [2]. It has been demonstrated that meniscectomy significantly reduced the acute symptoms of a meniscus lesion in short-term follow-ups

* Corresponding author. Department of Orthopedic Trauma, Hanover Medical School (MHH), OE 6230 Carl-Neuberg-Straße 1, D-30625 Hanover, Germany. Tel.: þ49 511 532 8970; fax: þ49 511 532 8969. E-mail addresses: [email protected], [email protected] (C. Liu). 0142-9612/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2011.10.041

[3], but several chronic syndromes were found to be present in long-term studies [4e6]. In addition, meniscectomy is highly associated with the development of the early osteoarthritis [7]. As arthroscopy techniques for performing meniscectomies have advanced, both inside-out and outside-in sutures have been advocated. However, the use of sutures is associated with a high risk of injury to the neurovascular structures in the white zone [8,9]. For decades, autologous and allogenic transplantation using tendons [10], cartilages [11], synovial flaps [12] and deep-frozen meniscal grafts [13] to reconstruct the meniscus has been of great interest. However, these meniscal replacements have had only a short-term effect in the prevention of degenerative cartilage [10,14], and the mechanical properties of the autologous materials were still inferior to those of the natural meniscus [15]. Since the traditional human auto- or allograft-materials have not resulted in satisfying meniscus regeneration, emphasis has been placed on developing alternative materials. With the help from the techniques in tissue engineering, many different kinds of

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

1053

Table 1 Characteristics of the seven donors that contributed to the cell pool used for this study. Donor number

Gender

Age

Diagnosis

Operative procedure

Drugs involved at the time of harvesting

1 2 3 4 5 6 7

Female Female Female Female Male Male Male

27 23 27 30 30 31 34

Burst fracture L3 Burst fracture C5 Burst fracture L1 Burst fracture L1 Osteochondritis dissecans Open fracture of the patella Tibial plateau fracture

Dorsal stabilization L2e4 Dorsal stabilization C4-6 Dorsal stabilization T12-L2 Dorsal stabilization T12-L2 Arthroscopic microfracturing Internal fixation Internal fixation

Fentanyl Sevoflurane Droperidole Fentanyl Sevoflurane Droperidole Fentanyl Sevoflurane Droperidole Fentanyl Sevoflurane Droperidole NSAID (COX-2 selective) NSAID NSAID

materials have be utilized to construct meniscal scaffold, among them agarose [16], hyaluronan [17] and collagen [18]. In order to obtain better mechanical properties and geometry of the scaffold, some polymers were employed because of their promising degradation rate and mechanical properties, such as poly(lactic-coglycolic acid) (PLGA) [19] and polyvinyl alcohol-hydrogel (PVA-H) [20]. However, no consensus was reached on the biocompatibility of these materials. As a biodegradable material, polyurethane (PU) has been used in clinic for several years. Commonly used polyurethanes are based on aromatic diisocyanates, which are noncytotoxic due to its possible toxic degradation products. For engineered meniscus, source cell is the other element as essential as scaffold. Meniscus cells were employed in some previous researches [21e24]. However, the amount of these cells is insufficient [25], which limited the wide-utilization of meniscus cells in the meniscus regeneration. Mesenchymal stromal cells (MSC), self-renewing multipotent stem cells, have recently aroused an interest for regenerative medicine. It has been demonstrated that MSC hold enormous potentials for reconstructing the tissues, such as bone, skin, articular cartilage, intervertebral disc and tendon [26e30]. In the light of this, MSC have been considered as an alternative cell source for regenerating meniscus [31]. Recently, dynamic cultures have been employed with the purpose of functional tissue engineering [32e35]. It is evidenced that the differentiation potential of MSC can be changed by the mechanical and topographical properties of their environment [36,37]. Even though the physiological stimulations were employed in tendon and articular cartilage tissue engineering researches [38,39], rare physiological stimulations were investigated in the meniscus tissue engineering field. In the present study, we for the first time developed a bioreactor, which mimicked the physiological environment of the knee joint. In this study, a meniscal scaffold composed of an aliphatic polyester-polyurethane, 1, 4-butanediisocyanate (BDI)-derived and human bone mesenchymal stromal cells (hBMSC) were utilized. The dynamic culture system that we developed produced a perfusion and cyclic compression on the hBMSC-laden PU scaffold. The goal of this study was to investigate the biocompatibility of the scaffold and the influence of the dynamic culture system to the viability, proliferation and fibro-cartilaginous differentiation of hBMSC growing in the scaffold. We hypothesized that the dynamic culture could alter the biochemical and biomechanical properties of tissue engineered menisci constructed with hBMSC and polyurethane scaffolds. 2. Materials and methods

solution at 80  C using a BDO. BDI. BDO urethane block-extender. The glass transition temperature of the resulting polyurethane was 54  C. The melting endotherm corresponding to the soft segment was 16.7  C, and the melting endotherm corresponding to the hard segment was 130.4  C. The HSC of polyurethane was 14.3%, determined by 1H NMR using the 2.25 and 3.13 ppm peaks [41,42]. The polyurethane was precipitated in water and dried under vacuum. Afterwards, the polyurethane was dissolved in 1, 4-dioxane at a concentration of 20%. The pores were created by mixing the polymer solvent solution with salt crystals ranging in size from 150 to 355 mm. After complete polymerization of the scaffold, salt crystals were removed by washing the polymer/salt mixture with distill water. 2.2. Effective porosity measurement To evaluate the porosity of the scaffold, dry weight (Wd) and apparent volume (Va) of each scaffold (n ¼ 6) was determined. Then these scaffolds were rinsed with 95% ethanol, washed with distilled water, and immersed in distilled water over night. After these steps, the wet scaffold weight (Ww) was obtained. The scaffold porosity was obtained on the basis of the following formula: Porosityð%Þ ¼

Ww  Wd 100 Va

where the units of Ww, Wd and Va are gram, gram and milliliter respectively. 2.3. Sudan Black B staining Scaffolds were embedded in OCTÔ compound (Sakura Finetek Europe B.V., Netherlands) and sectioned at 20  C to produce cross sections with a thickness of 80 mm. After rinsing the sections with water, they were stained with Sudan Black B as described previously [43]. Sections were observed using a light microscope (Olympus, Hamburg, Germany) at 100 magnification. Images were analyzed by Image-Pro Plus 5.0 (Media Cybernetics, USA). 2.4. hBMSC harvest and culture The Institutional Ethical Committee approved all procedures and written informed consent was obtained from all subjects (Table 1). Bone marrow aspirates were harvested from seven healthy donors (4 males, 3 females, median age: 29  3.5 years) by iliac crest aspiration during routine orthopedic procedures from February 2007 to March 2009. Isolation and cultivation of hBMSC were performed according to a modified protocol as previously described [44]. Experiments were performed with a single cell pool of all donors after three passages. 2.5. Cell seeding and culturing hBMSC of passage 3 were collected, resuspended and seeded onto the hydrated scaffolds, which were equilibrated in PBS over night with gentle shaking, at a density of 6  106 cells/scaffold under static conditions. After 4 h of attachment, the scaffolds were cultured under different conditions: static free swell culture group; perfusion culture group; mechanical stimulation I group and mechanical stimulation II group (Table 2).

Table 2 Groups and culture conditions investigated in this experiment: hBMSC were stimulated by perfusion or a combination of perfusion and mechanical stimulation. Static cultures served as controls.

2.1. Composite meniscus scaffold The polyurethane-based meniscus scaffolds were manufactured by Orteq (ActifitÒ, Groningen, The Netherlands) according to the procedures described by van Tienen et al. [40]. Briefly, a prepolymer of 50/50 ε-caprolactone/L-lactide was synthesized by ring opening polymerization of the monomers with 1, 4-butanediol (BDO) initiator. The obtained macrodiol was end-capped with 1, 4butanediisocyanate (BDI) at 80  C. Chain extension was performed in 50% DMSO

Groups

Perfusion rate (10 ml/min)

Mechanical stimulation (10% cyclic compression at 0.5 Hz)

Static free swell Perfusion Mechanical stimulation I Mechanical stimulation II

None Continuous Continuous Continuous

None None 1 time/day, 8 h/time 4 times/day, 2 h/time, 4 h of rest

1054

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

Fig. 1. Schematic showing the perfusion/mechanical stimulation bioreactor system used in this experiment.

2.6. Dynamic culture system Scaffolds seeded with cells were placed in a custom made bioreactor (Fig. 1). The bioreactor was continually perfused with a perfusion rate of 10 ml/min by a peristaltic pump (Ismatec IPC 16, ISMATEC SA, Glattbrugg, Germany). Cyclic confined compression was applied by means of a magnetic field engine. The compression was monitored by an electronic length gauge. The amplitude of the piston could be increased up to 2 mm and the stimulation frequency could be elevated up to 5 Hz. The parameters were adjusted according to Table 2. Fifty percent of the culture media was changed every 3 days and the system was kept in an incubator at 37  C. After 24 h, 1 week and 2 weeks, the constructs were collected from all groups and analyzed in further procedures. 2.7. Cell proliferation and viability assays After harvested from the bioreactor, strips of 5 mm width matrix were cut through the intact constructs crossly before the matrixes were chopped into 1 mm3 cube lets with a blade. The MTS assay kit (Promega Corporation, Madison, USA) was

used to investigate the proliferation of hBMSC growing in the scaffold as we described in our previous study [45]. The tissue matrixes were cut into slices about 1 mm thickness with blades immediately after harvested from the bioreactor. Viability of cells in matrixes was examined using a Live/Dead assay kit (Invitrogen, Carlsbad, CA) according to the procedure described by Park [46]. 2.8. Biochemical analysis After 24 h, 1 week and 2 weeks, cell-scaffold constructs (n ¼ 6) were lysed in 1 ml of lysis buffer (20 mM Tris, 300 mM NaCl, 1% Triton X-100, 1% Sodium Deoxycholate, 1 mM EDTA and 0.1% SDS) supplemented with 100 mM phenylmethylsulfonyl fluoride (PMSF, SigmaeAldrich, USA) over night at 4  C. Afterwards, protein quantification was determined with Coomassie Plus Assay Reagent (Thermo Scientific, USA) [47]. Commercial radioimmunoassay kits (Orion Diagnostica, Finland) were employed to investigate the quantification of the procollagen type I N-terminal propeptide (PINP) and procollagen type III N-terminal propeptide (PIIINP) [48,49]. The content of PINP and PIIINP was normalized to total protein content.

Fig. 2. Biopsy plugs taken from the cell-laden scaffolds for confined compression analysis (A and C). Sample of dog-bone shape taken for tensile testing (B and D).

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

1055

Fig. 3. SEM microphotographs and Sudan Black B staining of polyurethane scaffold, with 81% porosity. Macro pores varying from 80 mm to 400 mm and micro pores ranging from 0.3 mm to 9.0 mm were observed. Scale bars represent 200 mm (A), 50 mm (B), 200 mm (C) and 30 mm (D), respectively. Asterisk: Macro pore; Arrow: Micro pore.

2.9. Biomechanical analysis All specimens were soaked in expansion medium containing 10% (v/v) dimethyl sulfoxide (DMSO; SigmaeAldrich) and stored at 70  C [18]. Before biomechanical analysis, specimens were thawed, equilibrated in PBS, and tested within 1 h. Cylindrical disks of 6 mm diameter and 1 mm thickness were cut from the center of the cell-loaded matrices (Fig. 2). A ten-cycle confined compression stress relaxation test was performed to evaluate the equilibrium modulus [50]. The equilibrium modulus, representing the stiffness of the matrices, was defined as the slope of the stress-strain curve which passes through the equilibrium state of each stress relaxation cycle [51]. After mounting each disk in the confined compression chamber, the individual disks were compressed in 10 sequential steps of 50 mm (Zwick/Roell 1445, Zwick GmbH &Co KG, Ulm, Germany). Dog-bone

shape strips of 1 mm thick were cut from the surface of constructs for determination of the tensile modulus investigation [52] (Fig. 2). Samples were tested to failure at a strain rate of 0.1%/s. The tensile modulus was calculated from the stresses train curve [53]. The tensile modulus of elasticity, describing the strength of the samples in tension, was defined as the ratio of the tensile force to the cross section area of the specimen [54]. The tensile force was recorded with the resolution of 1datapoint/1 mm and the specimen cross section area is measured by a custom-built 2D laser scanner [55]. 2.10. Scanning electron microscopy (SEM) Acellular scaffolds and cell morphology and distribution were observed by Scanning electron microscope analysis. Samples were fixed with 2.5%

Fig. 4. Respective photographs of intact implant and cross-sectional view 4 h after cell seeding. Scale bars represent 10 mm (intact implant) and 5 mm (cross-section). SEM micrographs of 4 h and 24 h after culturing showed that hBMSC were able to adhere and appear to remain viable within the scaffold structure. Asterisk represents the macro pore; arrow represents the micro pore. Scale bar represent 20 mm.

1056

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

Fig. 5. Respective photographs of intact implants and cross-sectional views at 1 week and 2 weeks for the static and dynamic culture conditions. Scale bars represent 10 mm (intact implants) and 5 mm (cross sections).

glutaraldehyde (pH ¼ 7.4) for 24 h, and subsequently dehydrated in a graded ethanol series (20, 40, 60, 80 and 100%) for 10 min. Then the samples were critical-dried, coated with Au, and examined with SEM (Philips, Eindhoven, Netherlands).

with 1% bovine serum albumin at 37  C followed by primary antibody against procollagen type I and type III (Santa Cruz Biotechnology, Inc. CA) for 2 h. The sections were washed and incubated with FITC-labeled second antibody (1:100) (Santa Cruz Biotechnology, Inc. CA). The sections were counterstained with DAPI (4,6-diamidino-2-phenylindole) (Invitrogen, Camarillo, CA, USA).

2.11. Histological and immunofluorescence analysis 2.12. Statistical analysis Samples for histology were taken as complete cross sections of the middle portions of the specimens. They were embedded in OCTÔ compound and sectioned at 20 mm. HE and Masson Goldner staining were performed to observe the distribution of cells and matrix synthesis in the scaffolds. Similarly, the deposition of procollagen type I and type III was identified through immunofluorescence. Briefly, After fixed, hydrated, and permeabilized, the sections were incubated for 30 min

All values are expressed as mean values  standard deviation. Data were analyzed by one-way analysis of variance. Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS 15.0 for Windows; SPSS Inc.). A significance level of 95% with a p value of 0.05 was used in all statistical tests performed.

Fig. 6. MTS proliferation assay of cell-constructs cultured under the different conditions following 24 h, 1 week and 2 weeks after cell seeding. Data represents mean  standard deviation (n ¼ 6, *p < 0.05, #p < 0.01).

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

1057

3. Results

3.4. Biochemistry analysis

3.1. Properties of the PU scaffold

After being cultured with cells for 4 and 24 h, the scaffolds were examined by SEM. It was observed that the rounded cells became to polygon after 24 h and adhered on the wall of the pores in the scaffolds (Fig. 4). All engineered constructs in different groups retained wedge shape over the duration of culture (Fig. 5). Visual inspection of whole constructions and cross sections showed a time-depend increase in tissue homogeneity. Notably, samples of the perfusion group and the mechanical stimulation II group at 2 weeks exhibited darker color compared to the samples in static and the mechanical stimulation I groups.

After the culture for 1 week, the PINP expression level of samples collected from the different culture groups (static group, perfusion group, mechanical stimulation I and mechanical stimulation II) were 1.32-, 2.14-, 0.73-, and 2.45-fold, respectively, compared to the expression levels of the baseline (after 24 h). In addition, the on-off (mechanical stimulation II) compression stimulation resulted in the highest PINP expression among all experimental groups (p < 0.01), which was not observed in the longer-period (8 h/time) mechanical stimulation group. Unexpectedly, the PINP expression was not increased any more, even decreased slightly in the static and the perfusion groups (Fig. 8B), while the PINP content increased persistently up to 3.02-fold in the mechanical stimulation II group after 2 weeks compared to 24 h. Different phenomena were observed from the PIIINP assay results. After 1 week, the PIIINP level was elevated significantly in the perfusion and the mechanical stimulation II groups compared to static group (p < 0.05). After 2 weeks, the PIIINP level enhanced continuously in the perfusion group, which was 2.24-fold compared to the baseline. However, the PIIINP content in the scaffolds of the mechanical stimulation II group decreased slightly (Fig. 8C).

3.3. Cell proliferation and cell viability

3.5. Biomechanical analysis

The data obtained from MTS assay demonstrated a significant increase in proliferation after 1 and 2 weeks in all culture conditions except the mechanical stimulation I group. In addition, the maximum cell number was observed in perfusion group at each time point, which was significantly different from static group. There was no obvious difference between perfusion and mechanical stimulation II groups (Fig. 6). According to the result of cell viability assay, it was obvious that there were much more unviable cells in the mechanical stimulation I group. The percentage of viable cells after 24 h was about 82% in all culture groups and was altered to 58  11% (static control), 83  10% (perfusion), 52  7% (mechanical stimulation I) and 76  7% (mechanical stimulation II) after 2 weeks (Fig. 7).

Differences in the equilibrium modulus between the different culture conditions in the course of time were revealed from confined compression testing (Fig. 9A). It was observed that the equilibrium modulus in the perfusion and the mechanical stimulation II groups were improved in a time-dependent tendency. After 1 week of culture, the compressive equilibrium modulus of samples was enhanced 1.47  0.12 fold in the perfusion group and 1.52  0.15 fold in the mechanical stimulation II group, respectively, compared to the acelluar scaffolds (p < 0.05). No change was demonstrated in the mechanical stimulation I group. After 2 weeks, the equilibrium modulus of the mechanical stimulation II group increased up 1.85  0.15 fold compared to the acelluar scaffolds. The difference between the perfusion and the mechanical

In this study, the foamy PU scaffolds consisted of an interconnected network of pores with approximately 81% porosity. The SEM and Sudan Black B staining results showed that the macro pore size varied from 80 mm to 400 mm, the average of which was 226  26 mm. In addition, micro pores, the size of which ranged from 0.3 mm to 9.0 mm, were also observed (Fig. 3). 3.2. Cells adhesion and construct appearance

Fig. 7. (A) Cell viability of cell-culture in different groups was analyzed by Calcein AM/EthD-1 staining after 24 h, 1 week and 2 weeks. Scale bars represent 100 mm (red: nucleus of unviable cells, green: viable cells). (B) Graph of the cell viability of different groups. Data represents mean  standard deviation (n ¼ 6, *p < 0.05, #p < 0.01). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

1058

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

values observed after 1 week in all groups. However, the augmentations in the static, the perfusion and the mechanical stimulation II groups were completely different. The tensile modulus of perfusion group was the highest among all groups (p < 0.01). Unexpectedly, the tensile modulus of three dynamic groups decreased significantly after 2 weeks, which was not observed in static culture group. 3.6. SEM The interaction of the hBMSC with the scaffolds was further examined by SEM every week (Fig. 10). During the 2 weeks cultivation, cells were observed growing through the pores within the scaffolds and spreading uniformly and extensively in the static, perfusion and mechanical stimulation II groups. In addition, lamellipodium and interlaced fibrous network were observed, which demonstrated the cells adhered on the scaffold well. Compared to static culture condition, cell density appeared to be higher in the perfusion and mechanical stimulation II groups after the same time interval. It was further observed that the cells formed long cytoplasmic branches and interacted with each other in the perfusion and mechanical stimulation II groups. However, SEM images showed that the cell density was much lower in mechanical I group compared to the other groups, and cells clustered together instead of expanding with cytoplasmic branches into the pores of the scaffolds. 3.7. Histological and immunofluorescence analysis

Fig. 8. Biochemical assay results showing (A) total protein; (B) PINP/protein and (C) PIIINP/protein estimated in scaffolds seeded with hBMSCs and cultured under different conditions after 24 h, 1 week and 2 weeks. Data represents mean  standard deviation (n ¼ 6, *p < 0.05, #p < 0.01).

stimulation II groups was significant (p < 0.05). The equilibrium modulus of the mechanical stimulation I group increased by 18% after 2 weeks treatment, which was not different statistically from the value of static group (p > 0.05). Our finding showed that tensile modulus varied diversely with time in the different groups (Fig. 9B). It was observed that the tensile modulus was improved after culture with the maximum

Sections were subsequently stained with H&E, Masson Goldner staining to determine the distribution of the cells and the deposition of the extracellular matrix within the scaffolds (Figs. 11 and 12). The histology assay was coincidence with the results observed by SEM. Samples taken after 2 weeks were stained deeply in all groups except the mechanical stimulation I group where sparse cells and little of extracellular matrix deposition were observed. High density of cells was observed in the perfusion group and the mechanical stimulation II group, which was lower in the static group at the same time point. In addition, increased ECM production was observed in the mechanical stimulation II group. Immunofluorescence staining revealed the procollagen I and III within the scaffolds under different culture conditions, which was coincident with the radioimmunoassay results. Immunofluorescence characterization of the engineered constructs is summarized and representative fields are presented (Table 3 and Fig. 13). After 1 week, staining for type I procollagen was slight in the static group, whereas focal and extensive staining was observed in the perfusion and mechanical stimulation II groups. The deposition of type III procollagen was similar with type I procollagen. After 2 weeks, focal staining was observed for both type I and type III procollagens in static group. However, a markedly different pattern was observed in the dynamic culture groups: the matrix in the mechanical stimulation II group was stained strongly for type I procollagen but weakly for type III procollagen, which was reversed in perfusion group after 2 weeks. The staining in mechanical stimulation I group appeared to be the slightest among all experimental groups through the whole study intervals. 4. Discussion A porous scaffold without cytotoxicity to repair the damaged meniscus is of a major interest in regenerative medicine. Since meniscus is a dynamic tissue, the architecture of the implant should adapt to the mechanical requirements in vivo. Perfusion and mechanical stimulations are essential factors employed in tissue

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

1059

Fig. 9. Biomechanical assay results showing compressive equilibrium modulus (A) and tensile modulus (B) of the polyurethane scaffolds seeded with hBMSC and cultured under different conditions after 24 h, 1 week and 2 weeks. Data represents mean  standard deviation (n ¼ 6, *p < 0.05, #p < 0.01).

engineering research. Therefore, we used a porous meniscal scaffold made from an aliphatic polyester-polyurethane, 1, 4butanediisocyanate (BDI)-derived, which provides a 3-D culture condition for hBMSC in this study. In addition, a bioreactor was utilized to produce perfusion and mechanical stimulations in order to mimic the physiological conditions in the joint. The effects of perfusion and cyclic compression on hBMSC growing in the 3-D constructs were investigated. Perfusion was demonstrated to be the most effective stimulus for the cell proliferation among all experimental groups. The on-off dynamic compression (10% strain, 0.5 Hz, 4 times/day, 2 h/time with 4 h of rest thereafter) significantly increased extracellular matrix formation within the constructs, and yielded samples with a higher equilibrium modulus compared to the static culture condition, which was not observed in the long-term dynamic compression group (10% strain, 0.5 Hz, 1 time/day, 8 h/time).

The meniscus scaffolds investigated in our study was made from polyurethanes that were based on 1, 4-butanediisocyanate (BDI). We hypothesized our meniscus scaffold was non-cytotoxic, which was confirmed by MTS and Live/Dead assays (Figs. 6 and 7). According to the SEM assay, the porous polymer scaffold consisted of a considerable interconnected network of macro pores that presented a size between 80 and 400 mm, the average of which was 226  26 mm (Fig. 3). This pore size was supposed to be ideal for cells growth inside [56]. In addition, micro pores ranging from 0.3 to 9.0 mm were also observed (Fig. 3). It was reported that 3-D structure is essential for cell proliferation and the synthesis of cartilage-specific matrix proteins [57,58]. The macro pores in 3-D scaffolds is essential for enhancing cell migration [57]. On the other side, the micro pores with diameter less than 10 mm was pronounced to be helpful for nutrient transportation and fibrovascular colonization [59]. The acellular polyurethane scaffold with

1060

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

Fig. 10. Scanning electron micrographs showed hBMSC proliferation on polyurethane meniscus scaffolds under Static (a), Perfusion (b), Mechanical stimulation I (c) and Mechanical stimulation II (d) culture conditions after 1 week (A) and 2 weeks (B). Scale bar represents 20 mm.

the similar pore dimensions was investigated in the study of Verdonk, and capillaries within the scaffold were found 12 months after implantation [60]. In this study, a cyclic compression with axial displacement about 800 mm was imposed on the constructs. The compressive strain experienced under this loading condition was 10%, which mimics the reported 10% compressive strain for a native meniscus

in vivo [61]. It was demonstrated that 10 ml/min perfusion rate benefited the hBMSC proliferation in our previous study [45]. From the MTS and Live/Dead assays, perfusion was demonstrated to be the most effective factor for the cell proliferation as expected. Cartmell et al. indicated that high continuous perfusion rate (more than 1 ml/min) led to increased levels of cell death due to the high flow-induced shear stresses [62]. The volume of our bioreactor was

Fig. 11. HE staining showed hBMSC growth and ECM deposition on polyurethane meniscus scaffolds under the different culture conditions after 24 h, 1 week and 2 weeks. Scale bar represents 300 mm.

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

1061

Fig. 12. Masson goldner staining showed hBMSC growth and collagen-like tissue deposition on polyurethane meniscus scaffolds under the different culture conditions after 24 h, 1 week and 2 weeks. Scale bar represents 300 mm.

much bigger than that in Sarah et al., so the flow-induced shear stresses was not so deadly that the cell viability was decreased. We found that the on-off and long-term dynamic compressions had completely different influences on the cell proliferation and viability for the first time, even though the strain, frequency and total stimulation period were coincidence. It was demonstrated that the 0.5 Hz and 10% strain cyclic compression maintained the viability of hBMSC and enhanced its proliferation [63]. However, it seems like the model of stimulation period should be considered, because the continuous stimulation of 8 h/d had a deadly effect on the cell viability and soundly counteracted the positive contribution of perfusion in our study. The extracellular matrix of the meniscus is composed mainly of collagen, which is predominantly type I, with smaller quantities of types II and III [64]. In the present study, PINP and PIIINP contents were detected since there is no PIIP radioimmunoassay kit available by now. It was found that both PINP and PIIINP contents increased significantly in the perfusion and mechanical stimulation II groups (Fig. 8). After 1 week, the increase in PINP along with PIIINP

synthesis reflects fibro-cartilaginous differentiation might have occurred due to the physiological stimulations. In comparison, continuous mechanical stimulation for 8 h each day had no obvious contribution to the differentiation of cells. On-off mechanical stimulation increased the ECM formation and synthesis of PINP and PIIINP compared to the constructs cultured under free-swelling condition (Figs. 8, 11 and 12). This was possibly due to the fact that the compression imposed on the scaffolds decreased the intercellular spaces, which in turn could have increased the subcellular communication compared to the static condition. However, it was surprising that PIIINP content was decreased slightly in the mechanical stimulation II group after 2 weeks, instead of showing a persistent increase which was observed in the perfusion group. Ballyns et al. reported prolonged dynamic compression decreased GAG content in constructs, and enhanced the loss of ECM to the culture media [65]. We conjecture that this phenomenon occurred in our experiment also. Possibly, the speed of PIIINP synthesis was lower than the speed of loss after 2 weeks in the mechanical stimulation II group as well. On the other hand, it

Table 3 Immunophenotypic characterization of engineered menisci. Static

Procollagen I Procollagen III

Perfusion

MS I

MS II

1 week

2 weeks

1 week

2 weeks

1 week

2 weeks

1 week

2 weeks

e e

 

 þ

þ þþ

e e

e e

þ 

þþ þ

Expression of procollagen types I and III was assessed in tissue constructs cultured statically, under perfusion condition and mechanical stimulation (MS) I/II conditions. Scores are based on the percentage of positive tissue as: slight (<25% area of visual field positive) staining (); focal (<50% area of visual field positive) staining (); extensive (>50% area of visual field positive) weak staining (þ); extensive (>50% area of visual field positive) strong staining (þþ).

1062

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

Fig. 13. Immunohistochemistry of polyurethane-based engineered tissues. Representative immunohistochemical stain for procollagen type I (A, B, C, D) and procollagen type III (E, F, G, H) of the tissues cultured for 2 weeks in static (A, E), perfusion (B, F), mechanical stimulation I (C, G) and mechanical stimulation II (D, H) groups. Asterisk: scaffold; White arrow: nucleus; Triangle: Extracellular Matrix. Scale bar represents 50 mm.

was reported that cells cultured under perfusion condition expose to a fluid-induced shear, which is different from direct compression stimulation [66,67]. Although slight intension, that kind of shear force might activate some signal transduction pathway, which is essential for the fibro-cartilaginous differentiation of hBMSC [68]. Hereby, we concluded that the shear stress causes the persistent PIIINP synthesis in the perfusion group. In addition, it was possible that the contribution of shear stress could be attenuated by direct compression, which might be the secondary reason for the decrease of PIIINP in the mechanical stimulation II group after 2 weeks. It was found that the cell proliferation and PINP/PIIINP synthesis varied in different groups from the histological observation (Figs. 11e13). The biomechanical properties of scaffolds could be enhanced notably owning to the production of ECM [69]. Therefore, we employed a quantitative method that could identify alterations of mechanical properties in the presence of different cultural conditions. The equilibrium modulus and tensile modulus of the constructs was evaluated following static or dynamic culture. The results shown the equilibrium modulus of the scaffolds seeded with hBMSC increased compared to the acellular ones. The equilibrium modulus was enhanced with the timedepended accumulation of PINP, which was more obviously in dynamic culture systems (Fig. 9A). After 2 weeks, the equilibrium modulus of the scaffolds in mechanical stimulation II group was obviously higher than perfusion group (p < 0.05). Allowing for the higher accumulation of PINP in the mechanical stimulation II group, we presumed that PINP contributed to the equilibrium modulus. However, the equilibrium modulus in the mechanical stimulation I groups increased unexpectedly after 2 weeks even there was almost no ECM formation in the scaffolds. The possible reason was the long-term compression might squeeze the scaffolds slightly, which lead to the change of biomechanical properties. The tensile modulus evaluation shown there was the most significant increase in the perfusion group among all groups after 1 week (Fig. 9B). Similar tendency was observed in the PIIINP RIA assay result. Even though it was reported that type I collagen contributes to the tensile property of tissues [70], type III collagen was demonstrated to be crucial for the function of type I collagen [71]. However, the augmented effect of dynamic culture on the tensile property of the cell-laden constructs disappeared after

2 weeks. Possibility, the irrigating medium and the compression stimulation accelerated the degradation of the polyurethane scaffold, which vitiated the potency of the accumulative PINP/ PIIINP in the scaffolds. 5. Conclusion In our present study, the polyurethane-based meniscus scaffold was demonstrated to be non-cytotoxic for the growing of hBMSC seeded inside. The perfusion and mechanical stimulations benefited the fibro-cartilaginous differentiation of hBMSC separately. The hBMSC grown in the scaffolds cultured in the 10 ml/min rate perfusion group exhibited the highest degree of proliferation and PIIINP synthesis among all groups. The on-off dynamic compression (10% strain, 0.5 Hz, 4 times/day, 2 h/time) contributed to the PINP synthesis and the enhancement of equilibrium modulus time-dependently. In future, in vivo studies will be performed to investigate the characteristics of the constructs after the action of the physical stimulations in vitro. Grant support This work was finically supported by “Deutsche Arthrosehilfe”, the German Society of Orthopaedic Traumatologic Sports Medicine (GOTS) and AO Foundation’s Research Fund (AO 05-H74). Chaoxu Liu received financial support from China Scholarship Council (CSC). Conflict of interest The authors have no financial disclosures or conflicts of interest with the research presented here. Acknowledgments We gratefully thank J. Viering and H. Schumann (Institutional Central Research Laboratory) for the construction of the bioreactor system. We appreciate Gudrun Brandes and Gerhard Preiss from Cell Biology Laboratory MHH for their profound technical assistance. We also acknowledge the support of Michael Shin from Orteq Inc. for this study.

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

References [1] Cook JL. The current status of treatment for large meniscal defects. Clin Orthop Relat Res 2005;435:88e95. [2] McDermott ID, Amis AA. The consequences of meniscectomy. J Bone Jt Surg Br 2006;88:1549e56. [3] Matava MJ. Meniscal allograft transplantation: a systematic review. Clin Orthop Relat Res 2007;455:142e57. [4] Fauno P, Nielsen AB. Arthroscopic partial meniscectomy: a long-term followup. Arthroscopy 1992;8:345e9. [5] Hershman EB, Nisonson B. Arthroscopic meniscectomy: a follow-up report. Am J Sports Med 1983;11:253e7. [6] Wroble RR, Henderson RC, Campion ER, el-Khoury GY, Albright JP. Meniscectomy in children and adolescents. A long-term follow-up study. Clin Orthop Relat Res 1992;279:180e9. [7] Aagaard H, Verdonk R. Function of the normal meniscus and consequences of meniscal resection. Scand J Med Sci Sports 1999;9:134e40. [8] Reigel CA, Mulhollan JS, Morgan CD. Arthroscopic all-inside meniscus repair. Clin Sports Med 1996;15:483e98. [9] Arnoczky SP, Warren RF. Microvasculature of the human meniscus. Am J Sports Med 1982;10:90e5. [10] Kohn D, Rudert M, Wirth CJ, Plitz W, Reiss G, Maschek H. Medial meniscus replacement by a fat pad autograft. An experimental study in sheep. Int Orthop 1997;21:232e8. [11] Lecumberri JJ, Diaz de Rada MO, Santamaria M. Study of meniscal regeneration following total meniscectomy and the replacement of meniscus by different autologous plasties. An Sist Sanit Navar 1997;20:165e74. [12] Cisa J, Basora J, Madarnas P, Ghibely A, Navarro-Quilis A. Meniscal repair by synovial flap transfer. Healing of the avascular zone in rabbits. Acta Orthop Scand 1995;66:38e40. [13] Wirth CJ, Peters G, Milachowski KA, Weismeier KG, Kohn D. Long-term results of meniscal allograft transplantation. Am J Sports Med 2002;30:174e81. [14] Kohn D. Autograft meniscus replacement: experimental and clinical results. Knee Surg Sports Traumatol Arthrosc 1993;1:123e5. [15] Gloria A, De Santis R, Ambrosio L. Polymer-based composite scaffolds for tissue engineering. J Appl Biomater Biomech 2010;8:57e67. [16] Gunja NJ, Huey DJ, James RA, Athanasiou KA. Effects of agarose mould compliance and surface roughness on self-assembled meniscus-shaped constructs. J Tissue Eng Regen Med 2009;3:521e30. [17] Kon E, Chiari C, Marcacci M, Delcogliano M, Salter DM, Martin I, et al. Tissue engineering for total meniscal substitution: animal study in sheep model. Tissue Eng Part A 2008;14:1067e80. [18] Pabbruwe MB, Kafienah W, Tarlton JF, Mistry S, Fox DJ, Hollander AP. Repair of meniscal cartilage white zone tears using a stem cell/collagen-scaffold implant. Biomaterials 2010;31:2583e91. [19] Weinand C, Xu JW, Peretti GM, Bonassar LJ, Gill TJ. Conditions affecting cell seeding onto three-dimensional scaffolds for cellular-based biodegradable implants. J Biomed Mater Res B Appl Biomater 2009;91:80e7. [20] Kobayashi M. A study of polyvinyl alcohol-hydrogel (PVA-H) artificial meniscus in vivo. Biomed Mater Eng 2004;14:505e15. [21] Tan Y, Zhang Y, Pei M. Meniscus reconstruction through coculturing meniscus cells with synovium-derived stem cells on small intestine submucosa e a pilot study to engineer meniscus tissue constructs. Tissue Eng Part A 2010;16: 67e79. [22] Gunja NJ, Athanasiou KA. Additive and synergistic effects of bFGF and hypoxia oin leporine meniscus cell-seeded PLLA scaffolds. J Tissue Eng Regen Med 2010;4:115e22. [23] Tan GK, Dinnes DL, Butler LN, Cooper-White JJ. Interactions between meniscal cells and a self assembled biomimetic surface composed of hyaluronic acid, chitosan and meniscal extracellular matrix molecules. Biomaterials 2010;31: 6104e18. [24] Tan GK, Cooper-White JJ. Interactions of meniscal cells with extracellular matrix molecules: towards the generation of tissue engineered menisci. Cell Adh Migr 2011;5:220e6. [25] Buma P, Ramrattan NN, van Tienen TG, Veth RP. Tissue engineering of the meniscus. Biomaterials 2004;25:1523e32. [26] Naito H, Dohi Y, Zimmermann WH, Tojo T, Takasawa S, Eschenhagen T, et al. The effect of mesenchymal stem cell osteoblastic differentiation on the mechanical properties of engineered bone-like tissue. Tissue Eng Part A 2011; 17:2321e9. [27] Jin G, Prabhakaran MP, Ramakrishna S. Stem cell differentiation to epidermal lineages on electrospun nanofibrous substrates for skin tissue engineering. Acta Biomater 2011;7:3113e22. [28] Marquass B, Schulz R, Hepp P, Zscharnack M, Aigner T, Schmidt S, et al. Matrix-associated implantation of predifferentiated mesenchymal stem cells versus articular chondrocytes: in vivo results of cartilage repair after 1 year. Am J Sports Med 2011;39:1401e12. [29] Yang H, Wu J, Liu J, Ebraheim M, Castillo S, Liu X, et al. Transplanted mesenchymal stem cells with pure fibrinous gelatin-transforming growth factor-beta1 decrease rabbit intervertebral disc degeneration. Spine J 2010; 10:802e10. [30] Qiu Y, Lim JJ, Scott Jr L, Adams RC, Bui HT, Temenoff JS. PEG-based hydrogels with tunable degradation characteristics to control delivery of marrow stromal cells for tendon overuse injuries. Acta Biomater 2011;7:959e66.

1063

[31] Arnoczky SP. Building a meniscus. Biologic considerations. Clin Orthop Relat Res 1999;367:S244e53. [32] Mauck RL, Nicoll SB, Seyhan SL, Ateshian GA, Hung CT. Synergistic action of growth factors and dynamic loading for articular cartilage tissue engineering. Tissue Eng 2003;9:597e611. [33] Mauck RL, Seyhan SL, Ateshian GA, Hung CT. Influence of seeding density and dynamic deformational loading on the developing structure/function relationships of chondrocyte-seeded agarose hydrogels. Ann Biomed Eng 2002; 30:1046e56. [34] Mauck RL, Soltz MA, Wang CC, Wong DD, Chao PH, Valhmu WB, et al. Functional tissue engineering of articular cartilage through dynamic loading of chondrocyte-seeded agarose gels. J Biomech Eng 2000;122:252e60. [35] Waldman SD, Spiteri CG, Grynpas MD, Pilliar RM, Kandel RA. Long-term intermittent shear deformation improves the quality of cartilaginous tissue formed in vitro. J Orthop Res 2003;21:590e6. [36] Discher DE, Mooney DJ, Zandstra PW. Growth factors, matrices, and forces combine and control stem cells. Science 2009;324:1673e7. [37] Guilak F, Cohen DM, Estes BT, Gimble JM, Liedtke W, Chen CS. Control of stem cell fate by physical interactions with the extracellular matrix. Cell Stem Cell 2009;5:17e26. [38] Deng D, Liu W, Xu F, Yang Y, Zhou G, Zhang WJ, et al. Engineering human neotendon tissue in vitro with human dermal fibroblasts under static mechanical strain. Biomaterials 2009;30:6724e30. [39] Jin M, Emkey GR, Siparsky P, Trippel SB, Grodzinsky AJ. Combined effects of dynamic tissue shear deformation and insulin-like growth factor I on chondrocyte biosynthesis in cartilage explants. Arch Biochem Biophys 2003;414: 223e31. [40] van Tienen TG, Heijkants RG, Buma P, de Groot JH, Pennings AJ, Veth RP. Tissue ingrowth and degradation of two biodegradable porous polymers with different porosities and pore sizes. Biomaterials 2002;23:1731e8. [41] Spaans CJ, De Groot JH, Belgraver VW, Pennings AJ. A new biomedical polyurethane with a high modulus based on 1,4-butanediisocyanate and epsiloncaprolactone. J Mater Sci Mater Med 1998;9:675e8. [42] Heijkants RG, van Calck RV, van Tienen TG, de Groot JH, Buma P, Pennings AJ, et al. Uncatalyzed synthesis, thermal and mechanical properties of polyurethanes based on poly(epsilon-caprolactone) and 1,4-butane diisocyanate with uniform hard segment. Biomaterials 2005;26:4219e28. [43] Soldani G, Losi P, Milioni C, Raffi A. Light microscopy evaluation of polyurethane vascular grafts porosity by Sudan Black B staining. J Microsc 2002; 206:139e45. [44] Jagodzinski M, Drescher M, Zeichen J, Hankemeier S, Krettek C, Bosch U, et al. Effects of cyclic longitudinal mechanical strain and dexamethasone on osteogenic differentiation of human bone marrow stromal cells. Eur Cell Mater 2004;7:35e41. [45] Jagodzinski M, Breitbart A, Wehmeier M, Hesse E, Haasper C, Krettek C, et al. Influence of perfusion and cyclic compression on proliferation and differentiation of bone marrow stromal cells in 3-dimensional culture. J Biomech 2008;41:1885e91. [46] Park SH, Choi BH, Park SR, Min BH. Chondrogenesis of rabbit mesenchymal stem cells in fibrin/hyaluronan composite scaffold in vitro. Tissue Eng Part A 2011;17:1277e86. [47] Merceron C, Vinatier C, Portron S, Masson M, Amiaud J, Guigand L, et al. Differential effects of hypoxia on osteochondrogenic potential of human adipose-derived stem cells. Am J Physiol Cell Physiol 2010;298:C355e64. [48] Eastell R, Rogers A, Ni X, Krege JH. Effects of raloxifene and alendronate on bone turnover as assessed by procollagen type I N-terminal propeptide. Osteoporos Int 2011;22:1927e34. [49] Chalmers RJ, Kirby B, Smith A, Burrows P, Little R, Horan M, et al. Replacement of routine liver biopsy by procollagen III aminopeptide for monitoring patients with psoriasis receiving long-term methotrexate: a multicentre audit and health economic analysis. Br J Dermatol 2005;152:444e50. [50] Chang SC, Rowley JA, Tobias G, Genes NG, Roy AK, Mooney DJ, et al. Injection molding of chondrocyte/alginate constructs in the shape of facial implants. J Biomed Mater Res 2001;55:503e11. [51] Kelly DJ, Crawford A, Dickinson SC, Sims TJ, Mundy J, Hollander AP, et al. Biochemical markers of the mechanical quality of engineered hyaline cartilage. J Mater Sci Mater Med 2007;18:273e81. [52] Ballyns JJ, Wright TM, Bonassar LJ. Effect of media mixing on ECM assembly and mechanical properties of anatomically-shaped tissue engineered meniscus. Biomaterials 2010;31:6756e63. [53] Drury JL, Dennis RG, Mooney DJ. The tensile properties of alginate hydrogels. Biomaterials 2004;25:3187e99. [54] Willett JL. Mechanical properties of LDPE/granular starch composites. J Appl Polym Sci 1994;54:1685e95. [55] Haasper C, Colditz M, Budde S, Hesse E, Tschernig T, Frink M, et al. Perfusion and cyclic compression of mesenchymal cell-loaded and clinically applicable osteochondral grafts. Knee Surg Sports Traumatol Arthrosc 2009;17:1384e92. [56] del Real RP, Wolke JG, Vallet-Regi M, Jansen JA. A new method to produce macropores in calcium phosphate cements. Biomaterials 2002;23:3673e80. [57] El-Ayoubi R, DeGrandpre C, DiRaddo R, Yousefi AM, Lavigne P. Design and dynamic culture of 3D-scaffolds for cartilage tissue engineering. J Biomater Appl 2011;25:429e44. [58] Liu H, Lin J, Roy K. Effect of 3D scaffold and dynamic culture condition on the global gene expression profile of mouse embryonic stem cells. Biomaterials 2006;27:5978e89.

1064

C. Liu et al. / Biomaterials 33 (2012) 1052e1064

[59] Wang H, Li Y, Zuo Y, Li J, Ma S, Cheng L. Biocompatibility and osteogenesis of biomimetic nano-hydroxyapatite/polyamide composite scaffolds for bone tissue engineering. Biomaterials 2007;28:3338e48. [60] Verdonk R, Verdonk P, Huysse W, Forsyth R, Heinrichs EL. Tissue ingrowth after implantation of a novel, biodegradable polyurethane scaffold for treatment of partial meniscal lesions. Am J Sports Med 2011; 39:774e82. [61] Zielinska B, Killian M, Kadmiel M, Nelsen M, Haut Donahue TL. Meniscal tissue explants response depends on level of dynamic compressive strain. Osteoarthritis Cartilage 2009;17:754e60. [62] Cartmell SH, Porter BD, Garcia AJ, Guldberg RE. Effects of medium perfusion rate on cell-seeded three-dimensional bone constructs in vitro. Tissue Eng 2003;9:1197e203. [63] Pelaez D, Huang CY, Cheung HS. Cyclic compression maintains viability and induces chondrogenesis of human mesenchymal stem cells in fibrin gel scaffolds. Stem Cells Dev 2009;18:93e102. [64] McDevitt CA, Webber RJ. The ultrastructure and biochemistry of meniscal cartilage. Clin Orthop Relat Res 1990;252:8e18. [65] Ballyns JJ, Bonassar LJ. Dynamic compressive loading of image-guided tissue engineered meniscal constructs. J Biomech 2011;44:509e16.

[66] Nerurkar NL, Sen S, Baker BM, Elliott DM, Mauck RL. Dynamic culture enhances stem cell infiltration and modulates extracellular matrix production on aligned electrospun nanofibrous scaffolds. Acta Biomater 2011;7:485e91. [67] Jungreuthmayer C, Donahue SW, Jaasma MJ, Al-Munajjed AA, Zanghellini J, Kelly DJ, et al. A comparative study of shear stresses in collagenglycosaminoglycan and calcium phosphate scaffolds in bone tissueengineering bioreactors. Tissue Eng Part A 2009;15:1141e9. [68] Shin SJ, Fermor B, Weinberg JB, Pisetsky DS, Guilak F. Regulation of matrix turnover in meniscal explants: role of mechanical stress, interleukin-1, and nitric oxide. J Appl Physiol 2003;95:308e13. [69] Appelman TP, Mizrahi J, Elisseeff JH, Seliktar D. The differential effect of scaffold composition and architecture on chondrocyte response to mechanical stimulation. Biomaterials 2009;30:518e25. [70] Maffulli N, Ewen SW, Waterston SW, Reaper J, Barrass V. Tenocytes from ruptured and tendinopathic achilles tendons produce greater quantities of type III collagen than tenocytes from normal achilles tendons. An in vitro model of human tendon healing. Am J Sports Med 2000;28:499e505. [71] Liu X, Wu H, Byrne M, Krane S, Jaenisch R. Type III collagen is crucial for collagen I fibrillogenesis and for normal cardiovascular development. Proc Natl Acad Sci U S A 1997;94:1852e6.