Acta Biomaterialia 10 (2014) 1333–1340

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

Differential effect of hypoxia on human mesenchymal stem cell chondrogenesis and hypertrophy in hyaluronic acid hydrogels Meiling Zhu a,b,1, Qian Feng a,b,1, Liming Bian a,b,c,⇑ a

Division of Biomedical Engineering, The Chinese University of Hong Kong, Hong Kong Department of Mechanical and Automation Engineering, The Chinese University of Hong Kong, Hong Kong c Shun Hing Institute of Advanced Engineering, The Chinese University of Hong Kong, Hong Kong b

a r t i c l e

i n f o

Article history: Received 14 August 2013 Received in revised form 25 November 2013 Accepted 9 December 2013 Available online 14 December 2013 Keywords: Hyaluronic acid Hydrogel Cartilage tissue engineering Mesenchymal stem cells Chondrogenesis

a b s t r a c t Photocrosslinked hyaluronic acid (HA) hydrogels provide a conducive 3-D environment that supports the chondrogenesis of human mesenchymal stem cells (hMSCs). The HA macromer concentration in the hydrogels has a significant impact on the chondrogenesis of the encapsulated MSCs due to changes in the physical properties of the hydrogels. Meanwhile, hypoxia has been shown to promote MSC chondrogenesis and suppress subsequent hypertrophy. This study investigates the combinatorial effect of tuning HA macromer concentration (1.5–5% w/v) and hypoxia on MSC chondrogenesis and hypertrophy. To decouple the effect of HA concentration from that of crosslinking density, the HA hydrogel crosslinking density was adjusted by varying the extent of the reaction through the light exposure time while keeping the HA concentration constant (5% w/v at 5 or 15 min). It was found that hypoxia had no significant effect on the chondrogenesis and cartilaginous matrix synthesis of hMSCs under all hydrogel conditions. In contrast, the hypoxia-mediated positive or negative regulation of hMSC hypertrophy in HA hydrogels is dependent on the HA concentration but independent of the crosslinking density. Specifically, hypoxia significantly suppressed hMSC hypertrophy and neocartilage calcification in low HA concentration hydrogels, whereas hypoxia substantially enhanced hMSC hypertrophy, leading to elevated tissue calcification in high HA concentration hydrogels irrespective of their crosslinking density. In addition, at a constant high HA concentration, increasing hydrogel crosslinking density promoted hMSC hypertrophy and matrix calcification. To conclude, the findings from this study demonstrate that the effect of hypoxia on hMSC chondrogenesis and hypertrophy is differentially influenced by the encapsulating HA hydrogel properties. Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction Human mesenchymal stem cells (hMSCs) are an ideal cell source for regenerative medicine, especially for cartilage repair, because of major beneficial features, such as easy availability and multipotency. However, inadequate cartilaginous matrix production by chondrogenically induced MSCs and the unstable chondrogenic phenotype of hMSCs following the initial induction are considered to be the two major hurdles to the successful application of hMSCs in cartilage repair and regeneration [1,2]. Photocrosslinked hyaluronic acid hydrogels have been proven to be an effective biomaterial that supports the chondrogenesis of hMSCs [3,4]. However, in addition to their ability to undergo chondrogen⇑ Corresponding author at: Department of Mechanical and Automation Engineering, Room 213, William M.W. Mong Engineering Building, The Chinese University of Hong Kong, Shatin, Hong Kong. Tel.: +852 39438342; fax: +852 26036002. E-mail address: [email protected] (L. Bian). 1 These authors contributed equally.

esis, hMSCs also exhibit the tendency to differentiate towards a hypertrophic phenotype after the initial chondrogenic induction, similar to that observed in the terminal differentiation of hypertrophic chondrocytes during endochondral ossification, leading to extensive calcification of the neocartilage matrix after ectopic transplantation in subcutaneous mouse models [2,5]. It is known that the components and structure of the extracellular and pericellular matrices play an important role in the regulation of chondrocyte hypertrophy and matrix calcification [6–8]. Chondrocytes entering terminal differentiation also substantially remodel their surrounding cartilage matrix to produce a template that facilitates calcification [9]. Therefore, the physical properties of the hydrogel scaffold, such as macromer concentration and crosslinking density, which controls the quality and distribution of the newly formed cartilage matrix, may influence the hypertrophic differentiation of chondrogenically induced MSCs and consequently calcification of the neocartilage matrix. Previously, we showed that changing the crosslinking density of HA hydrogels, by varying the HA macromer concentration or changing the ultra-

1742-7061/$ - see front matter Ó 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.actbio.2013.12.015

1334

M. Zhu et al. / Acta Biomaterialia 10 (2014) 1333–1340

violet light (UV) exposure time, influences neocartilage formation, hypertrophy and neocartilage calcification by encapsulated chondrocytes and MSCs [10–12]. Meanwhile, hypoxia, or low oxygen tension, which mimics the physiological avascular microenvironment of articular cartilage, has been shown to significantly influence cartilaginous matrix production by chondrocytes and the chondrogenesis of MSCs via factors such as hypoxia-inducible transcription factor (HIF-1a) [13–15]. HIF-1a plays a critical role in chondrogenesis and cartilage development during skeletogenesis by regulating Sox9, a chondrogenic marker gene required for the initiation of chondrogenesis [16,17]. Recent studies have also indicated that hypoxia suppresses hypertrophy differentiation of chondrocytes and multipotent stromal cells [15,18,19]. We have previously shown that, after chondrogenic induction (either with or without subsequent hypertrophy induction), hMSC hypertrophy and the resulting neocartilage calcification can be mitigated by co-culture of chondrocytes with hMSCs, as well as by mechanical loading [20,21]. However, further understanding of the regulation of hMSCs hypertrophy and tissue calcification is needed to ensure the successful clinical application of hMSCs for cartilage repair. With these issues in mind, we hypothesize that the hypoxia condition will regulate the chondrogenic and hypertrophic differentiation of hMSCs encapsulated in HA hydrogels. As shown previously by our group, the HA hydrogel macromer concentration and crosslinking density influence the initial chondrogenesis, neocartilage formation and subsequent hypertrophy and matrix calcification by encapsulated MSCs [12]. Hence, the second hypothesis is that there will be an interactional effect between hypoxia and the hydrogel macromer concentration or crosslinking density on hMSC hypertrophy and cartilage calcification. Therefore, the objective of this study was to investigate the effect of hypoxia on hMSC chondrogenesis and hypertrophy in HA hydrogels of varying HA macromer concentration and crosslinking density. To facilitate the investigation, the hypertrophy of hMSCs was studied using an in vitro hypertrophy model established by a previous study [22]. 2. Material and methods

described in a previous study [12]. Briefly, HA hydrogel disks were incubated in a fluorescein-labeled dextran solution (10 kDa, 10 mg ml 1; Molecular Probes). At selected time points, the distribution of fluorescent intensity across the hydrogel cross-section was imaged. The average fluorescence intensity was fitted to a finite element simulation of diffusion to simulate dextran diffusion. An estimated value of the effective diffusivity was derived by fitting the simulation curve to the experimental findings using the least squares method.

2.3. Sample preparation and in vitro culture Human MSCs (Lonza) were expanded to passage 3 under the condition of normoxia (21% oxygen tension) in growth medium consisting of a-minimum essential medium with 16.7% fetal bovine serum and 1% penicillin/streptomycin, to 20 million hMSCs ml 1. The hMSCs were then photoencapsulated with UV light (wavelength 360 nm; intensity 1.2 mW cm 2) in three different formulations of MeHA hydrogel disks of identical size (5 mm diameter, 2.5 mm thickness). The three formulations (x%–ym, where x is the MeHA concentration (w/v) and y is the UV exposure time (min)) were: low HA concentration and low crosslinking density hydrogels (1.5%–15m), high HA concentration and high crosslinking density hydrogels (1.5%–15m), and high HA concentration and low crosslinking density hydrogels (5%–5m) (Fig. 1A and B). The high MeHA concentration in the 5% (w/v) precursor solution (hence more crosslinkable methacrylate groups in the precursor solution) allows formation of hydrogels with differential degrees of crosslinking by varying the UV exposure time. The constructs formed were cultured in chondrogenic medium (Dulbecco-s modified Eagle’s medium, 1% ITS + Premix, 50 lg ml 1 L-proline, 0.1 lM dexamethasone, 0.9 mM sodium pyruvate, 50 lg ml 1 ascorbate, antibiotics) supplemented with transforming growth factor-b3 (TGF-b3, 10 ng ml 1), which was changed three times per week [24]. Normoxia groups were cultured at 21% atmospheric oxygen level and 5% carbon dioxide. The hypoxia groups were incubated in a hypoxia chamber supplied with 1% oxygen and 5% carbon dioxide. To induce hypertrophy, constructs were first

2.1. Macromer synthesis Methacrylated HA (MeHA) was synthesized as previously reported [23]. Briefly, methacrylic anhydride (94%, FW 154.17, Sigma) was added to a solution of 1 wt.% HA (sodium hyaluronate powder, research grade, MW 74 kDa, Lifecore) in deionized water, adjusted to pH 8 with 5 N NaOH, and reacted on ice for 24 h. The macromer solution was purified via dialysis (MW cutoff 6–8 k) against deionized water for a minimum of 48 h with repeated changes of water. The final product was obtained by lyophilization and stored at 20 °C in powder form prior to use. The final macromer products were confirmed by 1H NMR to have a methacrylation level of 29%. Lyophilized macromers were dissolved in phosphate-buffered saline containing 0.05 wt.% of the photoinitiator 2-methyl-1-[4-(hydroxyethoxy)phenyl]-2-methyl-1-propanone (I2959, Ciba) to allow for UV-mediated polymerization. 2.2. Characterization of HA hydrogels The Young’s moduli of acellular HA hydrogels were evaluated under unconfined compression with a mechanical tester (TA Instruments) at a strain rate of 10% min 1 up to a compressive strain of 20%, and moduli were determined by fitting the stress vs. strain curve and calculating the slope. The effective diffusivity of macromolecules in acellular HA hydrogels was determined as

Fig. 1. Fabrication of MeHA hydrogels with varying crosslinking density by changing either the macromer concentration or the exposure time (A). 1.5%–15m: 1.5% MeHA solution crosslinked by 15 min of UV exposure; 5%–5/15m: 5% MeHA solution crosslinked by 5 or 15 min of UV exposure (B). Timeline, culture condition and media supplements of all groups. TGF-b3: transforming growth factor, T3: triiodothyronine, b-gly: b-glycerophosphate (C).

1335

M. Zhu et al. / Acta Biomaterialia 10 (2014) 1333–1340

cultured in chondrogenic medium for 2 weeks. The medium was then switched to hypertrophic induction medium (1 nM dexamethasone, 1 nM triiodothyronine (T3) and 10 mM b-glycerophosphate (b-gly)) from day 15 to day 28 of the culture [22] (Fig. 1C). Cell viability was assessed using a LIVE/DEAD assay kit (Molecular Probes), in which live cells are stained green with calcein-AM and dead cells are stained red with ethidium homodimer. Confocal microscopy images of the stained sample (scanned layer thickness 10 lm) were acquired within a depth of 150 lm from a cut surface along the axial axis of the hydrogel disk.

2.4. Gene expression analysis For gene expression analysis, samples were homogenized in Trizol Reagent (Invitrogen) with a tissue grinder, RNA was extracted according to the manufacturer’s instructions and the RNA concentration was determined using an ND-1000 spectrophotometer (Nanodrop Technologies). One microgram of RNA from each sample was reverse transcribed into cDNA using reverse transcriptase (Superscript II, Invitrogen) and oligoDT (Invitrogen). Polymerase chain reaction (PCR) was performed on an Applied Biosystems 7300 Real-Time PCR system using Taqman primers and probes specific for GAPDH (a housekeeping gene) and other genes of interest. Sequences of the primers and probes used are listed in Table 1. The relative gene expression was calculated using the DDCT method, where the fold difference was calculated using the expression 2DDCT. Each sample was internally normalized to GAPDH and each group was normalized to the expression levels of MSCs at the time of encapsulation (i.e. after expansion and before differentiation). Relative expression levels greater than 1 represent up-regulation with culture, while relative expression levels less than 1 represent down-regulation of that gene compared to that of initially encapsulated MSCs.

2.5. Biochemical analysis For biochemical analysis, half of each construct was weighed wet, lyophilized, reweighed dry and digested in 0.5 mg ml 1 Proteinase-K (Fisher Scientific) at 56 °C for 16 h. The PicoGreen assay (Invitrogen, Molecular Probes) was used to quantify the DNA content of the constructs, with lLambda phage DNA (0–1 mg ml 1) as the standard [25]. For each sample, the masses of both the entire gel and the half gel used for the DNA assay were measured. The total amount of DNA per sample was calculated by scaling the amount of DNA detected in the half gel by a weight ratio (total weight/half weight). The GAG content was measured using the dimethylmethylene blue (Sigma Chemicals) dye-binding assay, with shark chondroitin sulfate (0–50 mg ml 1) as the standard [26]. The overall collagen content was assessed by measuring the orthohydroxyproline (OHP) content via dimethylaminobenzaldehyde and chloramine T assay. The collagen content was calculated by assuming a 1:7.5 OHP-to-collagen mass ratio [27]. The collagen and GAG contents were normalized to the disk wet weight. Separate samples were digested in 1 N HCl for 3 days. The calcium con-

tent was determined by analyzing the diluted supernatant from the digestion using a calcium quantification kit (Biovision). 2.6. Histological analysis The remaining halves of the constructs were fixed in 4% formalin for 24 h, embedded in paraffin and processed using standard histological procedures. The histological sections (8 lm thick) were stained for targets of interest using the Vectastain ABC kit and the DAB Substrate kit for peroxidase (Vector Labs). Briefly, sections were predigested in 0.5 mg ml 1 hyaluronidase for 30 min at 37 °C and incubated in 0.5 N acetic acid for 4 h at 4 °C to swell the samples prior to overnight incubation with primary antibodies at dilutions of 1:100, 1:200 and 1:3 for chondroitin sulfate (mouse monoclonal anti-chondroitin sulfate, Sigma) and type I (mouse monoclonal anti-collagen type 1, Sigma) and type II collagen antibodies (mouse monoclonal anti-collagen type II, Developmental Studies Hybridoma Bank), respectively. Non-immune controls underwent the same procedure without primary antibody incubation. 2.7. Statistical analysis All data are presented as mean ± standard deviation. Statistica (Statsoft, Tulsa, OK) was used to perform statistical analyses using two-way analysis of variance, followed by Tukey’s HSD post hoc test to allow comparison between groups (n = 4 samples per group), with culture duration and experimental group as independent factors. 3. Results 3.1. Encapsulation of hMSCs in HA hydrogels and cell viability The Young’s modulus of the hydrogels increased significantly with increasing crosslinking density (low crosslinking density: 1.5%–15m @ 4.2 kPa, 5%–5m @ 5.2 kPa vs. high crosslinking density: 5%–15m @ 51.2 kPa). Meanwhile, the macromolecule diffusivity in the hydrogels decreased significantly (low crosslinking density: 1.5%–15m @ 5.7  10 5 mm2 s 1, 5%–5m @ 6.0  10 5mm2 s 1 vs. high crosslinking density: 5%–15m @ 3.5  10 5 mm2 s 1) with increasing crosslinking density as reported previously [12]. Viability staining indicated that the majority (>90%) of the encapsulated cells remained viable in all groups after 14 days of culture before switching to the hypertrophy induction medium (Fig. 2). The viability staining on day 28 of the culture was not significantly different from that on day 14 (data not shown). 3.2. Gene expression of chondrogenic and hypertrophic markers On day 28 of the culture, both type II collagen and aggrecan expression was significantly up-regulated in all groups compared to the undifferentiated hMSCs, while type I collagen expression remained at the baseline level. Aggrecan expression was higher in the low crosslinking density groups (1.5%–15m, 5%–5m) compared

Table 1 Sequences of primers and probes used for real-time PCR. Gene

Forward primer

Reverse primer

Probe

GAPDH COL I COL II Aggrecan

AGGGCTGCTTTTAACTCTGGTAAA AGGACAAGAGGCATGTCTGGTT GGCAATAGCAGGTTCACGTACA TCGAGGACAGCGAGGCC

GAATTTGCCATGGGTGGAAT GGACATCAGGCGCAGGAA CGATAACAGTCTTGCCCCACTT TCGAGGGTGTAGCGTGTAGAGA

CCTCAACTACATGGTTTAC TTCCAGTTCGAGTATGGC CTGCACGAAACATAC ATGGAACACGATGCCTTTCACCACGA

The sequences related to gene type X collagen, ALP and MMP13 are proprietary to Applied Biosystems Inc. and are not disclosed.

1336

M. Zhu et al. / Acta Biomaterialia 10 (2014) 1333–1340

to the high crosslinking density group (5%–15m) (Fig. 3). However, there was no significant difference in the expression level of type II collagen and aggrecan between the normoxia and hypoxia groups of all hydrogel formulations (Fig. 3). Type X collagen, MMP13 and alkaline phosphatase (ALP) expression was significantly up-regulated in all groups on day 28 compared to the undifferentiated hMSCs (Fig. 3). When encapsulated in the low HA concentration hydrogels (1.5%–15m), hMSCs cultured under hypoxia expressed significantly lower levels of hypertrophic marker mRNAs (type X collagen, MMP13 and ALP) compared to the normoxia condition (hypoxia 1.5%–15m < normoxia 1.5%–15m) (Fig. 3). In contrast, hypoxia notably promoted the gene expression of type X collagen, MMP13 and ALP in hMSCs encapsulated in both groups of high HA concentration hydrogels (5%–5m and 5%–15m) compared to their normoxia counterpart regardless of crosslinking density (hypoxia 5%–5m > normoxia 5%–5m; hypoxia 5%–15m > normoxia 5%–15m) (Fig. 3). The collagen X expression increased in high HA concentration hydrogels with increasing UV exposure time (5%–5m < 5%–15m) regardless of the oxygen tension (Fig. 3).

3.3. Cartilaginous matrix synthesis After 28 days of in vitro culture, the 5%–5m group exhibited the largest volumetric swelling, while the swellings in the 1.5%–15m and 5%–15m groups were similar (Fig. 4A). The quantification of cartilage-specific matrix components showed that the hMSCseeded constructs with low HA concentration and low crosslinking

density (1.5%–15m) possessed higher glycosaminoglycan (GAG) and collagen contents compared to the constructs of high crosslinking density (5%–15m) (Fig. 4B and C). The DNA contents of all groups were similar on day 28 (Fig. 4D). There was no significant difference in the construct volume, GAG, collagen and DNA contents between the normoxia and hypoxia groups of the same hydrogel formulation (Fig. 4). Immunohistochemistry staining against chondroitin sulfate and type II collagen revealed that cartilage matrix was more evenly deposited in the intercellular space of the low crosslinking density constructs (1.5%–15m and 5%–5m). In contrast, cartilage matrix was mostly restricted to the pericellular area in constructs of high crosslinking density (5%–15m) (Fig. 5). Staining against type I collagen was minimal (Fig. 5). Again, no major difference in cartilage matrix staining was observed between the normoxia and hypoxia groups of the same hydrogel formulation (Fig. 5).

3.4. Neocartilage calcification Consistent with the expression of hypertrophic mark genes, hypoxic culture reduced the calcium content in the hMSC-seeded hydrogel constructs fabricated with low HA macromer concentration (1.5%–15m) compared to the normoxic culture. In contrast, in both of the high HA macromer concentration groups (5%–15m and 5%–5m) the calcium content increased under the hypoxia condition (Fig. 6A) relative to the normoxic culture. Von Kossa staining also confirmed this opposing effect of hypoxic culture on the calci-

Fig. 2. Viability of the hMSCs encapsulated in various hydrogels conditions after 14 and 28 days of culture. Red: dead cells; green: living cells; scale bar = 100 lm. Images were acquired from a layer of around 10 lm thickness within a depth of around 150 lm from a cut surface along the axial axis the hydrogel disk.

M. Zhu et al. / Acta Biomaterialia 10 (2014) 1333–1340

1337

Fig. 3. Gene expression (fold change, normalized to GAPDH and to monolayer cells prior to encapsulation) of selected chondrogenic and hypertrophic markers in MSC-laden HA hydrogel constructs after 28 days of in vitro culture. ⁄p < 0.05 vs. the normoxia group of the same hydrogel property; +p < 0.05 vs. all other groups under the same oxygen tension (n = 4); p < 0.05 vs. the 5%–5m group.

Fig. 4. Construct volume normalized to the day 0 value (A), GAG and total collagen content normalized by sample wet weight (B and C) and DNA content per sample (D) of hMSC-laden HA hydrogel constructs after 28 days of in vitro culture. +p < 0.05 vs. all other groups under the same oxygen tension (n = 4); p < 0.05 vs. the 5%–5m group.

fication of hydrogels containing different concentrations of HA macromer (Fig. 6B). Compared to the normoxic culture, Von Kossa staining was considerably reduced under hypoxic culture in the 1.5% HA constructs compared to the normoxic culture (hypoxia 1.5%–15m < normoxia 1.5%–15m), whereas more staining was ob-

served in the 5% HA constructs cultured under hypoxia compared to normoxia regardless of the UV exposure time (hypoxia 5%–5m > normoxia 5%–5m; hypoxia 5%–15m > normoxia 5%–15m) (Fig. 6B). However, longer UV exposure time increased tissue calcification in the high HA concentration hydrogels under

1338

M. Zhu et al. / Acta Biomaterialia 10 (2014) 1333–1340

Fig. 5. Immunohistochemical staining for chondroitin sulfate (CS), type II collagen (Col 2), type I collagen (col1) and negative control (negative) of MSC-laden HA hydrogel constructs after 28 days of in vitro culture. Scale bar = 50 lm.

both the normoxia and hypoxia conditions (normoxia 5%– 5m < normoxia 5%–15m; hypoxia 5%–5m < hypoxia 5%–15m) (Fig. 6A and B).

4. Discussion Our previous study indicated that changing the macromer concentration and crosslinking density of the HA hydrogels not only alters the early chondrogenesis of the encapsulated hMSCs but also impacts the subsequent hMSCs hypertrophy and neocartilage calcification [12]. The crosslinking density was tuned by adjusting one of the two parameters (i.e. HA macromer concentration or UV exposure time), while keeping the other constant during photocrosslinking. However, few prior studies have evaluated the interaction between the hydrogel macromer concentration, the crosslinking density and hypoxia on hMSC chondrogenesis and the subsequent hypertrophy. In this study, we demonstrate that hypoxic culture differentially regulates the hypertrophic differenti-

ation of chondrogenically induced hMSCs encapsulated in HA hydrogels, depending on the HA macromer concentration and the crosslinking density. To be more specific, hypoxia inhibited the hypertrophy of hMSCs encapsulated in the low HA concentration hydrogels but enhanced the hypertrophy of hMSCs encapsulated in the high HA concentration hydrogels regardless of the crosslinking density (Figs. 3 and 6). Furthermore, in the high HA concentration hydrogels, increasing the hydrogel crosslinking density increased the hypertrophic calcification of neocartilage under both the normoxia and hypoxia conditions (Fig. 6). In this study, an in vitro hypertrophy model was used to evaluate hMSC hypertrophy after chondrogenic induction [22]. This model allows us to examine the influence of HA hydrogel properties and hypoxia on hMSC hypertrophy in a defined in vitro setting while avoiding the systemic complexity and considerable expense of in vivo studies. Switching to the hypertrophy induction medium after 2 weeks of chondrogenic induction not only expedites the hypertrophic differentiation of hMSCs by eliminating the hypertrophy-suppressing TGF-b3 but also provides the phosphate donors

Fig. 6. Calcium content of hMSC-seeded HA hydrogel constructs after 28 days of in vitro culture (A). Von Kossa staining of HA hydrogel constructs after 28 days of in vitro culture (converted to grayscale image for better visualization) (B). Scale bar = 400 lm. ⁄p < 0.05 vs. the normoxia group of the same hydrogel property; +p < 0.05 vs. all other groups under the same oxygen tension (n = 4).

M. Zhu et al. / Acta Biomaterialia 10 (2014) 1333–1340

necessary for mineralization. Therefore, this in vitro model has been used to efficiently and effectively screen and study factors such as the biomaterial properties, culture conditions and soluble agents that regulate hMSC hypertrophy [22,28]. Hypoxia is generally believed to promote chondrogenesis of MSCs and increase cartilaginous matrix synthesis [13,29]. Furthermore, low oxygen tension has also been shown to suppress hypertrophy of chondrogenically induced MSCs in pellet or 3-D hydrogel cultures [15,19,29]. However, recent studies have also reported increased expression of hypertrophic markers and a similar amount of cartilaginous matrix synthesis by MSCs after long-term hypoxic culture compared to normoxic culture [30,31]. These conflicting reports indicate the complexity of the mechanisms underlying the effect of hypoxia on chondrogenesis and hypertrophy. A recent study reported that hypoxia regulates chondrocyte hypertrophy both negatively and positively, via different pathways [18]. For instance, hypoxia inhibits the synthesis of type X collagen, a hypertrophic marker, by activating HDAC4 or suppressing Smad signaling. Meanwhile, hypoxia promotes type X collagen production by up-regulating p38 MAPK signaling. In this study, after 28 days of culture the expression levels of chondrogenic genes (type II collagen and aggrecan) and the amount of cartilaginous matrix synthesized were not significantly different between the hypoxic and normoxic cultures irrespective of the HA macromer concentration or crosslinking density. One likely explanation for this finding is that TGF-b3 supplemented in the medium is such a potent inducer of chondrogenesis that it masks the effect of hypoxia over the 28 days of long-term culture [30]. In comparison, hypoxia had a significant impact on the hypertrophic differentiation of the chondrogenically induced hMSCs. On one hand, in the low HA concentration (1.5% w/v) hydrogels, hypoxia suppressed the gene expression of hypertrophic markers (MMP13, type X collagen and ALP) of hMSCs, leading to reduced tissue calcification. On the other hand, hypoxia promoted the expression of the hypertrophic marker mRNAs and enhanced neocartilage calcification in the high HA concentration (5% w/v) hydrogels of both low (short UV exposure of 5 min, 5%–5m) and high (long UV exposure of 15 min, 5%–5m) crosslinking density. In addition, under normoxic culture, the two low crosslinking density groups (1.5%–15m and 5%–5m) showed similar levels of matrix calcification despite the difference in HA concentration (Fig. 6). This may indicate that the higher HA concentration may have augmented the positive regulatory pathways of hypoxia in promoting hMSC hypertrophy in the 5%–5m group. Another likely explanation for the differential effect of hypoxia is the different amounts of cartilaginous matrix deposited in the hydrogels. The low HA concentration hydrogels (1.5%–15m group) accumulated more GAG and collagen than the high HA concentration hydrogels (5%–5m or 15m group) (Figs. 4 and 5). During hypertrophic calcification, matrix vesicles secreted by hypertrophic chondrocytes accumulate and mineralize the surrounding cartilage matrix and display no greater capacity for calcification than vesicles isolated from normal cartilage [32]. These findings strongly suggest that the milieu of the vesicles (the surrounding extracellular matrix) strongly influences their ability to mineralize [8,32]. The two major extracellular matrix molecules in healthy articular cartilage – proteoglycans and type II collagen – are known to promote and stabilize the chondrogenic phenotype and thereby suppress matrix calcification [8,33,34]. Furthermore, type II collagen also binds to receptors on the surface of the matrix vesicles secreted by hypertrophic chondrocytes and suppresses mineralization [34]. Proteoglycans have been shown to bind and stabilize various growth factors via the anionic domains in their glycosaminoglycan side chains, prolonging the bioactivity of the growth factors [35,36]. Some of these growth factors, such as

1339

TGF-bs, are inhibitors of hypertrophic differentiation [37]. Lastly, recent evidence has suggested that HIF-1a is involved in the molecular assembly and organization of cartilage extracellular matrices, such as collagens, which in turn influences chondrogenesis and chondrocyte behaviors [38]. Therefore, the higher cartilage matrix content in the low HA concentration hydrogels may help suppress hMSC hypertrophy. The high HA concentration and high crosslinking density group (5%–15m) showed increased collagen type X gene expression and matrix calcification compared to the high HA concentration and low crosslinking density group (5%–5m) under bother normoxic and hypoxic culture (Figs. 3 and 6). Meanwhile, the cartilaginous matrix in the low crosslinking density group (5%–5m) was higher than that of the high crosslinking density group (5%–15m) (Fig. 4). This indicates that, in hydrogels of the same HA concentration, a high cartilaginous matrix content may help reduce hMSC hypertrophy and cartilage calcification. Most of the staining against calcification was observed in the peripheral area of the hydrogel constructs (Fig. 6B). However, such significant regional variation was not observed in type II collagen or chondroitin sulfate staining. The exact mechanism underlying this phenomenon remains unclear. Decreasing oxygen tension towards the interior of the hydrogels may have contributed to this spatial variation in calcification. However, it was noted that the top and bottom peripheries of certain hydrogels showed less calcification compared to the lateral peripheries despite similar oxygen tension in these regions. In addition, since the cell viability was similar in the interior region of the hydrogels compared to that of the periphery at the end of the 28 day culture (data not shown), the uneven distribution of calcification is unlikely to be due to cell viability. Our previous studies showed that hMSC-seeded HA hydrogels became fully calcified across the entire volume after in vivo implantation of similar duration [5,12]. Hence, the limited supply of calcium ions in in vitro culture may be a possible explanation for the preferential calcification in the hydrogel periphery. 5. Conclusions This study decouples the biological activity of hyaluronic acid from the physical properties of HA hydrogel by varying the crosslinking density independently of the HA macromer concentration. Our findings demonstrate that the hypoxia-mediated positive or negative regulation of the hypertrophic differentiation of hMSC in HA hydrogels after chondrogenic induction is dependent on HA concentration but independent of crosslinking density. Knowledge obtained from this study will be important in the design and optimization of hydrogels and tissue culture protocols that maximize hMSC chondrogenesis and neocartilage development, while limiting their propensity for hypertrophic calcification upon implantation. However, additional work is required to elucidate the precise mechanisms underlying the differential effect of hypoxia on the hypertrophic differentiation of hMSCs. 6. Disclosure There is no conflict of interests. Acknowledgements This work is supported by a direct Grant from the Faculty of Engineering in the Chinese University of Hong Kong. It is also supported by project BME-8115043 of the Shun Hing Institute of Advanced Engineering, the Chinese University of Hong Kong. This research project was made possible by equipment/resources donated by Lui Che Woo Foundation.

1340

M. Zhu et al. / Acta Biomaterialia 10 (2014) 1333–1340

Appendix Figures. with essential colour discrimination Certain figures in this article, particularly Figs. 1, 2, and 5, are difficult to interpret in black and white. The full colour images can be found in the on-line version, at doi:http://dx.doi.org/ 10.1016/j.actbio.2013.12.015. References [1] Erickson IE, Huang AH, Chung C, Li RT, Burdick JA, Mauck RL. Differential maturation and structure–function relationships in mesenchymal stem celland chondrocyte-seeded hydrogels. Tissue Eng Part A 2009;15:1041–52. [2] Pelttari K, Winter A, Steck E, Goetzke K, Hennig T, Ochs BG, et al. Premature induction of hypertrophy during in vitro chondrogenesis of human mesenchymal stem cells correlates with calcification and vascular invasion after ectopic transplantation in SCID mice. Arthritis Rheum 2006;54:3254–66. [3] Chung C, Beecham M, Mauck RL, Burdick JA. The influence of degradation characteristics of hyaluronic acid hydrogels on in vitro neocartilage formation by mesenchymal stem cells. Biomaterials 2009;30:4287–96. [4] Bian L, Guvendiren M, Mauck RL, Burdick JA. Hydrogels that mimic developmentally relevant matrix and N-cadherin interactions enhance MSC chondrogenesis. Proc Natl Acad Sci USA 2013;110:10117–22. [5] Bian L, Zhai DY, Tous E, Rai R, Mauck RL, Burdick JA. Enhanced MSC chondrogenesis following delivery of TGF-beta3 from alginate microspheres within hyaluronic acid hydrogels in vitro and in vivo. Biomaterials 2011;32:6425–34. [6] Boskey AL. The role of extracellular matrix components in dentin mineralization. Crit Rev Oral Biol Med 1991;2:369–87. [7] Anderson HC. Molecular biology of matrix vesicles. Clin Orthop Relat Res 1995:266–80. [8] Jubeck B, Gohr C, Fahey M, Muth E, Matthews M, Mattson E, et al. Promotion of articular cartilage matrix vesicle mineralization by type I collagen. Arthritis Rheum 2008;58:2809–17. [9] Ortega N, Behonick DJ, Werb Z. Matrix remodeling during endochondral ossification. Trends Cell Biol 2004;14:86–93. [10] Chung C, Mesa J, Randolph MA, Yaremchuk M, Burdick JA. Influence of gel properties on neocartilage formation by auricular chondrocytes photoencapsulated in hyaluronic acid networks. J Biomed Mater Res A 2006;77:518–25. [11] Erickson IE, Huang AH, Sengupta S, Kestle S, Burdick JA, Mauck RL. Macromer density influences mesenchymal stem cell chondrogenesis and maturation in photocrosslinked hyaluronic acid hydrogels. Osteoarthritis Cartilage 2009;17:1639–48. [12] Bian L, Hou C, Tous E, Rai R, Mauck RL, Burdick JA. The influence of hyaluronic acid hydrogel crosslinking density and macromolecular diffusivity on human MSC chondrogenesis and hypertrophy. Biomaterials 2012;34:413–21. [13] Duval E, Bauge C, Andriamanalijaona R, Benateau H, Leclercq S, Dutoit S, et al. Molecular mechanism of hypoxia-induced chondrogenesis and its application in in vivo cartilage tissue engineering. Biomaterials 2012;33:6042–51. [14] Schrobback K, Klein TJ, Crawford R, Upton Z, Malda J, Leavesley DI. Effects of oxygen and culture system on in vitro propagation and redifferentiation of osteoarthritic human articular chondrocytes. Cell Tissue Res 2011;347:649–63. [15] Gawlitta D, van Rijen MH, Schrijver EJ, Alblas J, Dhert WJ. Hypoxia impedes hypertrophic chondrogenesis of human multipotent stromal cells. Tissue Eng Part A 2012;18:1957–66. [16] Amarilio R, Viukov SV, Sharir A, Eshkar-Oren I, Johnson RS, Zelzer E. HIF1alpha regulation of Sox9 is necessary to maintain differentiation of hypoxic prechondrogenic cells during early skeletogenesis. Development 2007;134:3917–28. [17] Provot S, Zinyk D, Gunes Y, Kathri R, Le Q, Kronenberg HM, et al. Hif-1alpha regulates differentiation of limb bud mesenchyme and joint development. J Cell Biol 2007;177:451–64.

[18] Hirao M, Tamai N, Tsumaki N, Yoshikawa H, Myoui A. Oxygen tension regulates chondrocyte differentiation and function during endochondral ossification. J Biol Chem 2006;281:31079–92. [19] Leijten JC, Moreira Teixeira LS, Landman EB, van Blitterswijk CA, Karperien M. Hypoxia inhibits hypertrophic differentiation and endochondral ossification in explanted tibiae. PLoS One 2012;7:e49896. [20] Bian L, Zhai DY, Mauck RL, Burdick JA. Coculture of human mesenchymal stem cells and articular chondrocytes reduces hypertrophy and enhances functional properties of engineered cartilage. Tissue Eng Part A 2011;17:1137–45. [21] Bian L, Zhai DY, Zhang EC, Mauck RL, Burdick JA. Dynamic compressive loading enhances cartilage matrix synthesis and distribution and suppresses hypertrophy in hMSC-laden hyaluronic acid hydrogels. Tissue Eng Part A 2012;18:715–24. [22] Mueller MB, Tuan RS. Functional characterization of hypertrophy in chondrogenesis of human mesenchymal stem cells. Arthritis Rheum 2008;58:1377–88. [23] Smeds KA, Pfister-Serres A, Miki D, Dastgheib K, Inoue M, Hatchell DL, et al. Photocrosslinkable polysaccharides for in situ hydrogel formation. J Biomed Mater Res 2001;54:115–21. [24] Byers BA, Mauck RL, Chiang IE, Tuan RS. Transient exposure to transforming growth factor beta 3 under serum-free conditions enhances the biomechanical and biochemical maturation of tissue-engineered cartilage. Tissue Eng Part A 2008;14:1821–34. [25] McGowan KB, Kurtis MS, Lottman LM, Watson D, Sah RL. Biochemical quantification of DNA in human articular and septal cartilage using PicoGreen and Hoechst 33258. Osteoarthritis Cartilage 2002;10:580–7. [26] Farndale RW, Buttle DJ, Barrett AJ. Improved quantitation and discrimination of sulphated glycosaminoglycans by use of dimethylmethylene blue. Biochim Biophys Acta 1986;883:173–7. [27] Hollander AP, Heathfield TF, Webber C, Iwata Y, Bourne R, Rorabeck C, et al. Increased damage to type II collagen in osteoarthritic articular cartilage detected by a new immunoassay. J Clin Invest 1994;93:1722–32. [28] Scotti C, Tonnarelli B, Papadimitropoulos A, Scherberich A, Schaeren S, Schauerte A, et al. Recapitulation of endochondral bone formation using human adult mesenchymal stem cells as a paradigm for developmental engineering. Proc Natl Acad Sci USA 2010;107:7251–6. [29] Sheehy EJ, Buckley CT, Kelly DJ. Oxygen tension regulates the osteogenic, chondrogenic and endochondral phenotype of bone marrow derived mesenchymal stem cells. Biochem Biophys Res Commun 2012;417:305–10. [30] Meretoja VV, Dahlin RL, Wright S, Kasper FK, Mikos AG. The effect of hypoxia on the chondrogenic differentiation of co-cultured articular chondrocytes and mesenchymal stem cells in scaffolds. Biomaterials 2013;34:4266–73. [31] Ghone NV, Grayson WL. Recapitulation of mesenchymal condensation enhances in vitro chondrogenesis of human mesenchymal stem cells. J Cell Physiol 2012;227:3701–8. [32] Derfus BA, Kurtin SM, Camacho NP, Kurup I, Ryan LM. Comparison of matrix vesicles derived from normal and osteoarthritic human articular cartilage. Connect Tissue Res 1996;35:337–42. [33] Chen CC, Boskey AL, Rosenberg LC. The inhibitory effect of cartilage proteoglycans on hydroxyapatite growth. Calcif Tissue Int 1984;36:285–90. [34] Jubeck B, Muth E, Gohr CM, Rosenthal AK. Type II collagen levels correlate with mineralization by articular cartilage vesicles. Arthritis Rheum 2009;60:2741–6. [35] Hubbell JA. Materials as morphogenetic guides in tissue engineering. Curr Opin Biotechnol 2003;14:551–8. [36] Hynes RO. The extracellular matrix: not just pretty fibrils. Science 2009;326:1216–9. [37] Mello MA, Tuan RS. Effects of TGF-beta1 and triiodothyronine on cartilage maturation: in vitro analysis using long-term high-density micromass cultures of chick embryonic limb mesenchymal cells. J Orthop Res 2006;24:2095–105. [38] Hirsila M, Koivunen P, Gunzler V, Kivirikko KI, Myllyharju J. Characterization of the human prolyl 4-hydroxylases that modify the hypoxia-inducible factor. J Biol Chem 2003;278:30772–80.