|Home | About | Journals | Submit | Contact Us | Français|
This study investigated whether the synthetic peptide B2A (B2A2-K-NS) could induce in vitro chondrogenic differentiation and enhance the in vivo repair of damaged cartilage in an osteoarthritis model. In vitro, micromass cultures of murine and human stem cells with and without B2A were used as models of chondrogenic differentiation. Micromasses were evaluated for gene expression using microarray analysis and quantitative PCR; and for extracellular matrix production by Alcian blue staining for sulfated glycosaminoglycan and immunochemical detection of collagen type II. In vivo, osteoarthritis was chemically-induced in knees of adult rats by an injection of mono-iodoacetate (MIA) into the synovial space. Treatment was administered at 7- and 14 days after the MIA by injection into the synovial space of B2A or saline and terminated at 21 days, after which knee cartilage damage was determined and scored by histological analysis. In murine C3H10T1/2 micromass culture, B2A induced the expression of more than 11 genes associated with growth factors/receptors, transcription, and the extracellular matrix, including PDGF-AA. B2A also significantly increased the sulfated glycosaminoglycan and collagen of murine- and human micromass cultures. In the knee osteoarthritis model, B2A treatment enhanced cartilage repair compared to untreated knees as determined histologically by a decrease in damage indicators. These findings suggest that B2A induces stem cells chondrogenic differentiation in vitro and enhances cartilage repair in vivo. The results suggest that B2A might be useful to promote cartilage repair.
Current clinical treatment of damaged articular cartilage generally targets pain relief and not the repair of damaged cartilage. This is due, in part, to the fact that chondrocytes in articular cartilage have a limited regenerative capacity. However, therapeutic treatment of damaged cartilage by use of chondrocytes, bone marrow stem cells, and other pluripotent cells is being actively investigated(1–3). In some cases, those cells are expanded in culture prior to implant with the use of recombinant growth factors as modulators of proliferation and differentiation. Also, recombinant growth factors such as insulin-like growth factor, platelet-derived growth factor, and bone morphogenic proteins (BMP) have been used in research settings to facilitate cartilage repair(4–10). Other growth factors/cytokines may be useful in that regard as well.
Another therapeutic approach is to use bioactive peptides to improve cartilage repair(11). B2A is one such bioactive peptide. B2A is a positive BMP-2 receptor modulator(12, 13) whose design is modeled on BMP and incorporates three domains: a heparin-binding domain, a hydrophobic domain, and a receptor-targeted domain(13). B2A-type peptides bind to type I and type II receptors, and appear to have a selectivity for BMPR-Ib(13). B2A stimulation of cells increases ERK1/2 activation and, in the presence of BMP-2, augments Smads and alkaline phosphatase activities(13). Since B2A binds BMP receptors, and BMPs and BMP receptors are involved in virtually every aspect of chondrocyte formation, differentiation, and survival, it was hypothesized that B2A could play a role in augmenting chondrocyte growth, differentiation, and ultimately cartilage repair.
In this study we used two strategies to investigate the effects of B2A on chondrogenic repair. One evaluated the in vitro effects of B2A on chondrocytes or chondrocyte precursor cells. The second evaluated the in vivo reparative effects of B2A on cartilage in experimental osteoarthritis in rat knees.
Human bone marrow mesenchymal stem cells (hBMSC), and human articular chondrocytes (hNAC) were purchased from Cambrex BioScience. The murine multipotential embryonic stem cell line C3H10T1/2 was purchased from ATCC (Manassas, VA).
C3H10T1/2 cells were seeded into wells of 96-well plates in 100μl serum-free medium (DMEM:F12 supplemented with 1mM sodium pyruvate, 2mM glutamate, and 0.5mg/mL gentamicin, 5mg/mL lipid-rich bovine serum albumin, 10μg/mL insulin, 5.5μg/mL transferrin, 5.5ng/mL sodium selenite, and 2μg/mL ethanolamine) for 4 hours prior to B2A (4μg/mL) treatment. After 4 days of B2A treatment the cell proliferation was determined using CyQUANT® Cell Proliferation Assay Kit (Invitrogen, Carlsbad, CA) following the directions of the manufacturer.
High density micromass cultures were prepared in two similar methods. In the studies of gene expression, PDGF-AA secretion, and collagen type II protein expression, C3H10T1/2 micromasses were established by centrifugation of a cell suspension (3×105 cells/mL) in a conical tube at 600 rpm for 1 minute. After the micromasses formed, chondrogenic differentiation medium with or without B2A (10μg/mL) were added to the tube. Medium was changed every 3 days. To prepare samples for immunohistostaining of collagen type II, 10μL of a cell suspension containing 1×107 cells/mL was placed in polypropylene tissue culture vessels. After the micromass was formed, chondrogenic differentiation medium with without B2A (5μg/mL) was added. The chondrogenic differentiation medium was a standard basal media (DMEM/F12) containing 10% FBS (15% FBS for hBMSCs) and supplemented with 1mM sodium pyruvate, 2mM glutamate, 0.5mg/mL gentamicin, and 50μg/mL phosphate ascorbic acid. For the hNAC micromass culture, 100nM dexamethasone was also added. The micromass cultures were maintained in 37°C, 5%CO2 humidified incubator and the culture medium was changed every 3 days. It is noted that in order to investigate the intrinsic effects of B2A in chondrogenic differentiation, no other growth factor was added to the differentiation medium.
Real-time assays were performed with the RT2 Profiler™ PCR array: mouse osteogenesis (SA Biosciences, Frederick, MD, catalogue number PAMM-026A), according to the manufacturer’s directions. The starting amount of RNA used was 0.5μg. PCR was carried out on a Lightcycler 480 II (Roche Diagnostics Corporation, Indianapolis, IN) using an extended two-step cycling program. The HotStart DNA polymerase was heat activated at 95°C for 10 minutes. In the next step which was forty-five cycles of PCR, each cycle was held at 95°C for 15 seconds followed by 60°C for a minute. Total of 6 micromasses were analyzed in two independent experiments for each treatment, and the samples from each experiment were analyzed twice with the RT2 Profiler™ PCR array. Data were analyzed using the PCR Array Data Analysis Web Portal (SA Biosciences Corporation, Frederick, MD, USA). As recommended by the manufacturer, a reliable gene expression level was called when the Ct (cycle threshold) value is <30. The relative gene expression levels were determined by normalizing the genes of interested to the housekeeping genes of the B2A-treated and the control samples.
The medium was collected from the day 7 micromass culture. The samples were spun at 1000 × g for 15 minutes to remove particulates. The secretion of PDGF-AA in culture supernatants was measured by ELISA (Quantikine human/mouse PDGF-AA ELISA kit, R&D Systems; Minneapolis, MN).
After 12 days of treatment with B2A the micromasses were washed with PBS, and lysed in SDS-lysis buffer. Cell lysates were resolved by SDS-PAGE and transferred onto PVDF membranes. Blocked membranes were incubated with anti-collagen type II (Chemicon, Temecula, CA) and detected by chromogenic immunodetection. Total ERK1/2 detected by anti-MAP-kinase antibody (Sigma-Aldrich, St. Louis, MO) was used to monitor equal sample loading. The intensity of the bands was determined using ImageJ and normalized to untreated control and graphically represented as fold increase.
At day 12, micromasses were washed with PBS, fixed with 10% buffered formalin containing 0.5% cetylpyridinium chloride, embedded in paraffin, sectioned, and stained with 0.5% Alcian blue 8GX (Sigma-Aldrich, St. Louis, MO) in 3% acetic acid. To immunostain collagen type II, the sections were incubated with anti- collagen type II (Santa Cruz Biotechnology, Santa Cruz, CA) followed by incubation with Alexa Fluor 488-conjugated donkey anti-goat IgG (Invitrogen, Carlsbad, CA). The collagen expression was then revealed by fluorescent microscopy.
A rat model that mimicked aspects of osteoarthritis was used this study(14). Studies were conducted under an IACUC-approved protocol and using pathogen-free, skeletally-mature male Sprague-Dawley rats. In anesthetized rats, MIA (3 mg in 25μL saline) was injected into the synovial space through the patellar ligament with the right leg flexed at a 90° angle at the knee. At day 7- and 14 after MIA, groups of 6 rats were similarly injected with with B2A (500ng) or saline (control), and then sacrificed at 21 days. All animals gained weight during the study period, and there were no mortalities or infection/ inflammation at the injection site. After sacrifice, the knee joints were dissected, fixed, decalcified, and stained with either hematoxylin and eosin or Safranin O/Fast Green. The sections were examined and evaluated blinded using a scoring system described by Nishida et al(14) where a score of 0 indicates cartilage equivalent to normal and where 3 indicates markedly damaged. Sections stained with hematoxylin and eosin stained slides were scored based on general morphology as 0 (normal hyaline cartilage); 1 (slightly reduced,<1/4); 2 (markedly reduced, 1/4–3/4), 3 (no metachromatic stain, more than>3/4 reduction). Safranin O/Fast Green stained slides were scored for glycosaminoglycan content as 0 (normal); 1 (slightly reduced,<1/4); 2 (markedly reduced, 1/4–3/4), 3 (no metachromatic stain, more than 3/4 reduction). A total score was determined by summing scores from both stained to generate a total score where 0 was normal and 6 was markedly damaged.
Data are presented as the mean ± S.D., or mean with the lower-and upper limits of the 95% confidence interval. Comparison between means was assessed by unpaired Student’s t test or all pairwise multiple comparison procedure (Student-Newman-Keuls method) followed by Mann-Whitney rank sum test, as indicated in respective figures and tables.
The relative gene expression levels of C3H10T1/2 micromasses with and without B2A treatment were examined at 7 days using PCR arrays containing 84 genes of osteo-chondrogenic lineage (the list of genes is presented in the Supplementary Material. The genes shown up- or down-regulated with statistical significance are presented in Figure 1. Compared to the controls, 11 genes were found up-regulated in the B2A treated micromass samples and one gene was down-regulated. Genes that were up-regulated in B2A treatment micromasses can be categorized into three groups: (1) genes associated with growth factors and growth factor receptors, (2) genes associated with transcription factors and gene regulation, and (3) genes of matrix proteins. In the first group, Fgfr1 and Fgfr2 were found significantly increased in the B2A-treated samples. We also detected a moderate up-regulation of growth factors Fgf1. In the second gene group, the Smad1, Smad4, and Twist1 were up-regulated. Twist1, it should be noted, is required to promote and maintain chondrogenic differentiation in immature chondrocytes(15). Among the matrix genes of the third gene group, collagens Col11a1 and Col3a1 were up-regulated, with Col3a1 being the most pronounced. In contrast, Col5a1, a collagen that is not associated with cartilage, was noticeably down-regulated by B2A. Phex and Serpinh1, genes that regulate extracellular matrix stability, were found to be increased. The Bmp1 gene, encoding a metalloproteinase, which induces ectopic cartilage formation, was found significantly increased in the B2A-treated cells. The microarray-base expression analyses also revealed that Sox9 and Col2 were up-regulated but did not reach statistical significance. In individual quantitative real-time PCR, Sox9 and Col2 were found increased in the B2A-treated micromasses with 1.3 (p<0.05) and 1.6 (p<0.0001) fold of control, respectively.
Previous reports (12, 13) indicated that B2A-type peptides enhance BMP-2-induced osteogenic differentiation, yet by itself B2A did not initiate osteogenic differentiation in vitro. In this study we set out to investigate whether B2A possesses intrinsic activities for promoting chondrogenic differentiation and cartilage repair. B2A was evaluated for proliferation in a pluripotent cell lines, C3H10T1/2. As shown in Table 1, B2A stimulated a significant increase of proliferation compared to untreated control.
In the PCR array study, the Pdgfa gene was found up-regulated in the B2A treated micromass, although it did not reach statistical significance. To verify if the Pdgfa gene was up-regulated, we investigated whether B2A increased the production of PDGF-AA, the gene product of Pdgfa. At 7 days, B2A significantly increased the amount of PDGF-AA found in the medium of micromasses (Table 2).
It is expected that accumulation of proteins to a quantitatable levels might take a longer time compared to mRNA level detected by PCR. Therefore, expression of collagen type II protein and sulfated glycosaminoglycans were examined on day 12. The expression of collagen type II was investigated by immunoblotting as well as immunohistochemical staining. An increased expression of collagen type II was found in the B2A-treated C3H10T1/2 micromass in immunoblotting experiments (Figure 2). BMP-2, known to increase collagen type II in this system, was used as positive control and also increased the collagen type II expression. Immunohistochemical analysis of collagen type II expression was carried out in hBMSC and C3H10T1/2 cells. The control micromasses expressed low levels of collagen type II, whereas B2A treated micromasses exhibited increased cartilage specific collagen (Figure 3).
On day 12 hNAC micromasses were stained with Alcian blue to reveal sulfated glycosaminoglycans, an indication of cartilage differentiation. The control cultures displayed minimal Alcian blue staining indicating a low degree of differentiation. On the other hand, cells continuously treated with 5μg/mL B2A exhibited pronounced Alcian blue staining suggesting that the micromass culture underwent an increased production of glycosaminoglycans, indicative of the hNAC underwent re-differentiated chondrocytes after B2A treatment (Figure 4).
B2A was evaluated, in this pilot study, for its ability to increase the repair of damaged cartilage in a MIA induced osteoarthritis model. If left untreated (saline group), the MIA-induced animals developed severe damage in the tibial trochlear cartilage. Histological observation of that cartilage revealed central zones with marked reduction of basophilic nuclear staining indicative of cell death extending from the superficial layer to the boney layer and bracketed laterally by disorganized cells (Figure 5).
In addition, Safranin O staining for cartilage glycosaminoglycans was markedly reduced in the untreated knee compared to the normal contra-lateral knee. In contrast, rats that received B2A treatment exhibited significant repair of the damaged cartilage (Figure 5). There were no discernible zones of necrosis, and there were clearly discernible cells in isogenous groups across the section plane. Compared to the saline treated group, there was a noticeable improvement of histological score in the B2A treated group. However, the B2A treated group had not returned to normal by day 21. It should be pointed out, however, that while there did appear to be a substantial increase in cell density in the B2A treated group compared to the saline treated group, no effort was made in this pilot study to quantitate any difference by histomorphometric analysis.
In this study, we report that using in vitro micromass culture models of chondrogenesis, the peptide B2A was able to up-regulate a number of genes associated with chondrogenesis. That up-regulation was reflected in an increased production of sulfated glycosaminoglycans and type II collagen. In vivo, using a chemically-induced model of osteoarthritis in rat knees, B2A increased cartilage repair as monitored histologically.
The micromass culture model has been used extensively to study chondrogenic differentiation(16–20). In this model, chondrocyte precursors, including C3H10T1/2 cells, undergo processes similar to cartilage development. Under high cell density conditions and in the presence of appropriate growth factors, cells differentiate and express collagen type II and cartilage-specific proteoglycans, and eventually form tissue masses that resemble cartilage. In the present study, C3H10T1/2 cells were used in concert with PCR to evaluate gene expression following B2A stimulation.
Twist1, a member of the basic helix-loop-helix transcription factors family, was up-regulated upon B2A treatment. A high level of Twist1 is required to promote and maintain chondrogenic differentiation in immature chondrocytes(15), and thus its up-regulation would be expected. As the chondrocytes turn into mature hypertrophic chondrocytes, Twist1 level decreases. The higher level of Twist1 found in the 7-day micromass culture suggests that the cells are in transition from stem cells to premature chondrocytes.
Additionally, Fgf1, Fgfr1, and Fgfr2 were up-regulated in the B2A treated micromass cultures. B2A stimulated the production of PDGF-AA. PDGF-AA has been found to be a specific PDGF isoform associated with chondrogenic phenotype(21). It is a potent mitogenic and chemotactic factor for mesenchymal stem cells and chondrocytes, and can also increase proteoglycan production in chondrocytes. The increased Pdgfa might serve as an autocrine loop to propel the chondrogenic differentiation. Interestingly, Bmp1, a member of the BMP1/TLD metalloproteinases (review see Hopkins et al (22)), was found significantly up-regulated (~500 fold). In addition to playing critical roles in regulating the formation of various extracellular matrixes, BMP1 also regulates activity of TGF-β family members. BMP1 activates BMP2/4 by cleaving their antagonist, chordin (23, 24). In vivo, BMP1 is able to induce ectopic cartilage (25, 26). The increase of Bmp1 expression might play a role in the B2A enhancement of cartilage repair. Intriguingly, Col3a1 and Col11a1 were found increased significantly at day 7. Col3a1 and Col11a1 are detected in articular cartilage as an intrinsic part of the type II collagen fibril. They consist of 10- and 3% of the total collagen found in the cartilage matrix, respectively (review see Eyre et al (27)). The up-regulation of Col3a1 and Col11a1 by B2A might play a role in articular cartilage matrix repair and remodeling.
Productions of glycosaminoglycans and type II collagen are two hallmarks of chondrogenic differentiation. It has been reported that TGF-β, BMP-2, and GDF5, growth factors that play positive roles in chondrogenic differentiation during embryonic development and cartilage repair in adulthood, significantly increase the glycosaminoglycan and type II collagen in vitro. B2A also stimulated the production of glycosaminoglycans and type II collagen of both hBMSC and hNAC in micromass culture. These findings suggest that B2A might enhance chondrogenic differentiation and cartilage repair in vivo as well. It should be noted that in the experiments described here TGF-beta was not included in the medium as is normally done for most micromass cultures, and thus the effect of B2A was specific.
The ability of B2A in vitro to influence gene expression and cartilage-like phenotypes suggests that B2A might have a role in promoting cartilage repair. In this report, we demonstrate that B2A increased the repair in damaged cartilage in vivo. MIA inhibits glycolysis in the joint and causes the death of chondrocytes(28). MIA, as used in the present study, causes progressive cartilage degeneration and produces a pathologic manifestation similar to that found in arthritis(14). Guzman and colleagues reported that MIA causes chondrocyte degeneration at as early as day 1 post-injection and by day 5 and 7, marked loss of chondrocytes and moderate collapse of the cartilaginous matrix was observed(29). In the present study, at 7 days after MIA injection, B2A treatment was initiated (7- and 14 days). The separation of MIA injection and B2A treatment minimized the possibility that B2A could interact with MIA and compromise interpretation of results. B2A significantly improved the cartilage repair in the damaged knees. While this study indicates that B2A improves cartilage repair, additional studies focused on optimizing the B2A dose and carrier would be useful. Moreover, testing the B2A effects in models other than MIA and study of the B2A-enhanced repair kinetics will also provide further information to determine whether B2A applied in the synovial space is a viable therapeutic approach.
While it is clear that additional studies are needed to clarify the role that B2A might play in cartilage repair, the results reported here are encouraging. Repair of articular cartilage(30) is a significant clinical challenge in part because cartilage has no vascularity and low cell density(31), and chondrocytes have a low proliferation capacity. Repopulation of cartilage defects does occur, at least to some degree in humans, and perhaps more so in experimental animals where the origin of the repopulating cells appears to be primitive mesenchymal cells of the marrow and synovial fat pad(32). B2A may also play a role in the therapeutic use of MSCs as an agent to help expand cell populations and to prime the cells towards the chondrogenic differentiation prior to implantation.
In conclusion, B2A might facilitate cartilage repair via multiple aspects: 1) increasing the expression of many genes associated with chondrocyte maturation in progenitor cells, 2) stimulating chondrogenesis in stem cells as evidenced by the induction of multiple genes associated with chondrocyte differentiation, and 3) increasing the production of extracellular matrix by chondrocytes. Continued study of B2A is warranted to further evaluate this peptide in cartilage repair.
This work was supported in part by “Regulation of Stem Cell-Based Cartilage Bioengineering by Transglutaminase” from Maryland Stem Cell Research Fund (Grant #2010-MSCRFE-0156-00 to M.N.) and the NIH grants (R56DK071920 and R03AR057126) to M.N.
COMPETING INTERESTS: X. Lin, Y. Liu, and P. Zamora are employees of BioSurface Engineering, Inc. and have received stock and or stock options.