Abstract
Damaged hyaline cartilage shows a limited capacity for innate repair. Potential sources of cells to augment the clinical repair of cartilage defects include autologous chondrocytes and mesenchymal stem cells. We have reported that culture of human bone marrow mesenchymal stem cells with specific growth and differentiation factors as shallow multilayers on Transwell permeable membranes provided ideal conditions for chondrogenesis. Rigid translucent cartilaginous disks formed and expressed cartilage-specific structural proteins aggrecan and type II collagen. We report here the analysis of the collagen network assembled in these cartilage constructs and identify key features of the network as it became mature during 28 days of culture. The type II collagen was co-polymerized with types XI and IX collagens in a fibrillar network stabilized by hydroxylysyl pyridinoline cross-links as in epiphyseal and hyaline cartilages. Tandem ion-trap mass-spectrometry identified 3-hydroxylation of Proline 986 and Proline 944 of the α1(II) chains, a post-translational feature of human epiphyseal cartilage type II collagen. The formation of a type II collagen based hydroxylysyl pyridinoline cross-linked network typical of cartilage in 28 days shows that the Transwell system not only produces, secretes and assembles cartilage collagens, but also provides all the extracellular mechanisms to modify and generate covalent cross-links that determine a robust collagen network. This organized assembly explains the stiff, flexible nature of the cartilage constructs developed from hMSCs in this culture system.
Keywords: Bone marrow stem cell, Cartilage, Type II collagen, Extracellular matrix, Pyridinoline cross-link, 3-Hydroxyproline
Introduction
Chondrocytes synthesize and deposit three tissue-specific collagen molecules, types II, IX and XI that copolymerize to form the nascent fibrillar framework of hyaline cartilage matrix (Fig. 1). The three collagen types become cross-linked by covalent bonds to form heteropolymeric fibrils [1]. Trivalent hydroxylysyl pyridinoline (HP) cross-links between amino-(N) or carboxy-(C) telopeptides and helical sites are the most prevalent intermolecular bonds stabilizing the mature cartilage collagen [2]. These cross-links together with their divalent precursors bond type II to type II and type II to type XI in the interior and type II to type IX collagen molecules on the surface of the hybrid polymer [1,3]. The type XI collagen molecules are cross-linked head-to-tail to each other exclusively by divalent ketoamine cross-links and laterally to C-telopeptides of type II collagen [4]. This cross-linked collagen framework counterbalances the swelling pressure imparted by the osmotic properties of aggrecan and enables epiphyseal and articular cartilage to withstand mechanical stresses.
Fig. 1.
Model of the architecture of the type II/IX/XI collagen heterofibril characteristic of native articular cartilage. Collagen type II (blue) is the predominant molecule in fibrillar articular cartilage, and it forms a covalent crosslinking network with collagen types XI (yellow) and IX (red). Type XI collagen forms a filamentous template within the type II collagen fibril. Type IX collagen decorates the surface of the fibril.
The post-translational quality of type II collagen molecules with respect to prolyl 3-hydroxylation at specific sites has been shown to vary between tissue types [5]. Mutations in prolyl 3-hydroxylase isoenzymes, associated helper proteins that affect proline 3-hydroxylation [6–14] and mutations that alter the primary structure of the α1(II) collagen chains can disrupt fibrillogenesis and result in a disorganized cartilage extracellular matrix during development and growth [15–19]. Pro-986 has been shown to be highly 3-hydroxylated in α1(IIA/B) chains in mouse cartilage [20] and in the Swarm rat chondrosarcoma (RCS-LTC) cell line pN-α1(IIB) chains [21]. Pro-944 and Pro-707 are also 3-hydroxylated and positioned within 3 residues of the collagen D-period molecular stagger (234 amino acids) suggesting a supplementary role in collagen fibril assembly [5,21,22]. The variation in the degree of 3-hydroxylation at the Pro-944 in α1(II) collagen is highly tissue-specific, ranging from 10% in bovine articular cartilage, 40% in the nucleus pulposus of the intervertebral disk to 90% in the vitreous of the eye.
A reliable, reproducible method to extract donor bone marrow from humans is well established making bone marrow stem cells an attractive first choice for regenerative medicine [23]. The potential for bone marrow mesenchymal cells to renew, repair or develop into tissues is now being actively pursued. Human bone marrow stem cells (hMSCs) have been shown to have chondrogenic potential and strategies to regenerate cartilage are being explored [24–29]. Under suitable culture conditions hMSCs produce a cartilage-like tissue with a matrix based on type II collagen and aggrecan [30,31]. Transwell culture of hMSCs on flexible porous membranous supports has been shown to improve chondrogenic matrix properties. [31,32]. However, to what extent the fibrillar matrix is well developed, in terms of the post-translational biochemistry, assembly and cross-linking of type II collagen and its co-assembly with (types IX and XI collagens), has not been characterized. We have developed methods to screen for normal collagen co-polymeric assembly in order to monitor the quality of the matrix deposited by chondrocytes in culture and in cartilages affected by heritable or acquired disease states [20,33,34]. This study investigated the ability of human bone marrow stem cells undergoing chondrogenesis to deposit collagen in the neo-matrix and to assemble a type II collagen-based cross-linked network characteristic of cartilage.
Results
Human MSCs were initially seeded as a shallow 8–9 cell deep multilayer on the Transwell membrane. By day 3 the cell layer had a translucent appearance but little structure and when unfixed it collapsed into a spheroidal mass following removal from the membrane. In contrast, by day 7 a thin but coherent tissue layer formed, which remained intact on removal from the membrane and by day 14 the disk-like tissue had a glistening, firm cartilage-like appearance, which by day 28 was a firm, rigid disk having a cartilage-like consistency (Fig. 2A). This was as previously reported [31] and is shown in the example histology of fixed sections (Fig. 2B). The pattern of chondrogenesis was very uniform in these constructs and expression of genes encoding cartilage matrix proteins, collagens, and proteoglycans was greatly increased particularly in the first 7 days of culture. [31]. To characterize the collagen deposited during the initial phase of chondrogenesis, the collagen network laid down in the matrix by day 3 and day 7 was depolymerized using pepsin. The pepsin resistant collagen chains in these extracts were resolved by SDS-PAGE and identified by mass spectrometry of in-gel trypsin digests (Fig. 2C). At day 3 only the α1(I) chain was detected in Band 1, which was most likely a product of undifferentiated hMSC. However, by day 7 Band 1 was primarily the α1(II) chain with less α1(I) chain detected. At day 7 Band 2 was identified as an α1(XI) chain, consistent with type XI collagen being co-expressed with type II and forming the template for type II collagen fibrils. Abundant type II collagen and also type XI collagen were therefore deposited in the matrix by the end of the first 7 days in culture.
Fig. 2.
A. Scaffold-free hMSC cartilage constructs. Cartilage constructs expanded and differentiated into a firm tissue by day 14. A rigid, glassy, smooth cartilage-like disk developed by day 28. Left to right: Days 3, 7,14, and 28. B: Histological distribution of sulfated polyanion in sections of disk cultures. Staining with Safranin O/Fast Green at 1, 7, 14, and 28 days of culture. Modified from [31]. C. SDS-PAGE showing collagen chains identified by in-gel trypsin digestion and tandem mass spectrometry. Representative sequences of tryptic peptides from SDS-PAGE collagen bands 1 and 2 seen in gel are shown in table. P* = 3-hydroxyproline; P# = 4-hydroxyproline. Collagen types II and XI are synthesized and deposited in the extracellular matrix as early as day 7 in culture. Type I collagen is also present at day 3, potentially from undifferentiated hMSCs prior to chondrogenesis.
The collagen content of representative cartilage constructs (Table 1) increased continuously from day 7 to day 28. The collagen content by day 28 (1% of wet weight) was about a third that of 15-week human fetal epiphyseal cartilage (3% of wet weight). Mature trivalent hydroxylysyl-pyridinoline (HP) cross-links were minimal at 7 days, but had increased by 14 days to the level in 15-week fetal human cartilage and remained constant to 28 days. Lysyl-pyridinoline (LP) was too low to quantify in the constructs, or in fetal cartilage. It is very likely that this mature collagen network could be conducive to the rigidity of the cultured cartilage constructs.
Table 1.
Collagen and pyridinoline cross-link content of hMSC cartilage constructs.
Days in culture | Wet wt. tissue analyzed (mg) | μg hydroxyproline | % collagen (wet wt.) | Pyridinoline moles/ mol collagen | |
---|---|---|---|---|---|
| |||||
HP | LP | ||||
7 | 3.45 | 1.95 | 0.42 | 0.05 | – |
14 | 9.4 | 6.7 | 0.54 | 0.23 | – |
28 | 9.7 | 11.57 | 0.90 | 0.22 | – |
Human fetal cartilage control (15 weeks) | 13 | 3.98 | 2.98 | 0.23 | – |
To assess more details of the homo- and hetero-polymer collagen cross-links formed during the 28 days, pepsin resistant collagen fragments from the matrix at days 7, 14 and 28 were screened using specific antibodies after separation by SDS-PAGE. To examine if a heteropolymer of type IX collagen to type II collagen had formed in the constructs, the blot was probed with monoclonal antibody 2B4 that identifies the C-terminal sequence, KGPDP, of the α1(IX) chain, which in cartilage is cross-linked to α1(II) collagen chains [35]. As seen in Fig. 3; panel A, the antibody reacted strongly with the α1(II) chain in extracts at days 14 & 28 of culture indicating that pepsin-generated C-terminal stubs of the α1(IX) collagen chain were cross-linked to α1(II) and hence heteropolymeric type IX–type II cross-links had formed as early as 14 days, subsequent to the type XI–type II cross-links (see antibody 5890 below) and persisted as the construct developed. The reactive band migrating slightly slower in the control cartilage and in day 28 digests is an incomplete cleavage product of the α1(IX) collagen chain cross-linked to an α1(II) chain and has been previously characterized [35].
Fig. 3.
Type II, type IX and type XI collagen heteropolymer formation in cartilage constructs. A type II collagen based chondrogenic phenotype was observed in the constructs. Equivalent sample loads were electrophoresed and probed in Western blots. A. The mAb 2B4 recognized the pepsin cleavage stub in the C-telopeptide domain α1(IX) chain when it is cross-linked to α1(II) chains. Reactivity was observed only after days 14 and 28 in culture. B. The mAb 10F2 reacted with cross-linked α1(II) chains in all lanes. This antibody specifically recognizes the C-telopeptide domain of type II collagen when it is cross-linked to α1(II) collagen chains as observed for fetal human type II collagen. C. pAb 5890 specifically recognizes N-telopeptide domain of α1(XI) collagen when cross-linked to chains of α1(II) chains in all lanes. This antibody also reacted with the slower moving α3(XI) chains at day 28. D. The monoclonal antibody (mAb) 1C10 confirmed the presence of type II collagen chains in cultures from all days. In all blots, signal increased in intensity with time in culture indicating a temporal increase in type II collagen deposition and heteropolymer formation. 15-week human fetal cartilage collagen was used as a control.
Stripping and re-probing the blot using the monoclonal antibody 10F2 revealed evidence of a cross-linked polymer of the type II collagen formed in the matrix by 7 days. As seen in Fig. 3; panel B, the antibody recognizes the α1(II) collagen chains in pepsin extracts of day 7, 14 and 28 day constructs indicating that the C-telopeptide α1(II) stubs were linked to α1(II) and that a cross-linked type II–type II collagen network was progressively assembled in the cultures [33]. An α1(XI) band was also faintly detected in the control fetal cartilage and on higher exposures in the day 28 extracts (data not shown) indicating that α1(II)-C-telopeptide was also linked to α1(XI) chains.
To determine if type XI–type XI collagen cross-links had also formed in the cartilage constructs, the same blot was stripped and re-probed with polyclonal antibody 5890. This antibody recognizes an epitope in the N-telopeptide of the α1(XI) chain, even when cleaved from α1(XI), but cross-linked laterally to a different chain. Fig. 3; panel C, shows that this antibody strongly recognizes the α1(II) band in extracts from days 7,14 and 28 indicating cross-linking of the N-telopeptide of α1(XI) to α1(II) chains and hence that type XI–type II cross-links had formed by day 7. A band migrating slightly slower than α1(II) was also detected at day 28, which has been previously identified as the α3(XI) chain of type XI collagen cross-linked to an N-telopeptide of the α1(XI) chain from another molecule and thus a homopolymer of type XI collagen had also formed [4,20]. Finally to confirm that collagen chains were still present on the blot following the extensive stripping method, the monoclonal antibody 1C10 specific to residues 934–945 in the α1(II) chain of type II collagen was used to probe the blot. As Fig. 3; panel D shows, the major reactive band was the α1(II) chain confirming that type II collagen was still abundantly present. The molecular interpretations of the western blots are diagrammed for clarity in Fig. 4.
Fig. 4.
Molecular interpretations of collagen heteropolymer assembly from Western blot analysis. A. Antibody 2B4 reacted with the α1(II) chains, showing that the C-telopeptide domains of α1(IX) (KGPDP epitope) had become cross-linked to type II collagen chains from a different molecule and thus a heteropolymer of type IX collagen and type II collagen had formed. B. Antibody 10F2 reacted with the α1(II) chains, showing that the C-telopeptide domains of α1(II) (EKGPDPLQ epitope) had become cross-linked to type II collagen chains from a different molecule and thus a homopolymer of type II collagen had formed. C. Antibody 5890 reacted with the α1(II) chains and the α3(XI) chains, indicating that the N-telopeptide of α1(XI) (DGSKGPTISA epitope) had become cross-linked to α3(XI) chains or α1(II) chains from a different molecule and thus a homopolymer of type XI collagen and a heteropolymer of type XI and type II had formed. These heterotypic cross-linking reactions have been demonstrated for type IIA/IIB collagen heteropolymers that are present in the matrix of developing cartilage [3,4,20,34,35].
Since post-translational modifications of α1(II) chains are turning out to be important for the stability and assembly of type II collagen molecules into fibrils [12,21,22], the 3-hydroxylation of specific proline residues in the α1(II) collagen chains following day 3 was evaluated by SDS-PAGE and mass spectrometry. Fig. 5A clearly showed a major pepsin-resistant Coomassie blue stained protein band in developing cartilages, increasing in intensity from days 7 to 28 that migrated identically to the α1(II) collagen chain from human fetal cartilage. Mass spectrometry (Fig. 5B) showed ≥91% 3-hydroxylation at Pro-986 in the α1(II) chains when compared to 100% 3-hydroxylation in the fetal control. The Pro-944 site in the α1(II) chains was highly 3-hydroxylated and showed 84% occupancy that decreased to 62% by day 28. 15-week fetal cartilage revealed only 3% 3-hydroxylation at this site. Candidate 3-hydroxyproline residues in the (GPP)4-repeat containing peptide at the C-terminus of the α1(II) chain [21] were below detection limits. It was noticeable that no significant Coomassie blue stained protein bands were seen in the range (45 kDa) expected for pepsin treated type X collagen [36,37] (data not shown), suggesting that there was little type X collagen assembled into the matrix.
Fig. 5.
A. SDS-PAGE showing collagen chains with time in culture. The most prevalent Coomassie blue stained chain was α1(II) chain. The amount of type II collagen progressively increased with time in culture. Two other faintly stained chains moving slightly slower than the α1(II) chain were identified as α2(XI) and α1(XI) chains. B. Occupancy of 3-hydroxyproline at Pro 986 and Pro 944. Degree (%) of 3-hydroxylation of the proline residues at 986 and 944 sites were determined by using tandem mass spectrometry and quantitated as in [21].
Discussion
Over many years, a focused attempt to generate cartilage in culture from chondrocytes and stem cells has led to progress in the field of cartilage tissue engineering [24,25,27,28]. However, biochemical and mechanobiological limitations have also become obvious [27,38]. One shortcoming is the low concentration of collagen in the constructs synthesized by chondrocytes or hMSCs in culture compared with native cartilage. Also, simply increasing collagen production does not necessarily produce mechanically sound neo-cartilage, as it is all the elements of extracellular processing that convert fibrillar collagen into a cross-linked heteropolymeric unit that underpins essential cartilage mechanical properties. We show here for the first time evidence that hMSCs in a suitable chondrogenic environment elaborate a collagen network with all the features of a rapidly maturing covalently cross-linked matrix.
The culture of hMSC in Transwells begins as a shallow multilayer of cells and the first week is accompanied by cell division, (8-fold increase in DNA), whereas thereafter there is little increase in cell number [31]. This first phase of chondrogenesis is also characterized by transient Notch signaling through Jagged 1 (days 1–3), which has to be switched off for chondrogenesis to proceed [39]. This clearly coincides with sparse matrix collagen production and predominantly type I collagen on days 0–3 of culture. However, by 7 days expression of the major chondrocyte matrix collagen gene COL2A1 increased more than 100-fold [31] and a coherent type II collagen matrix was established. Over the following culture to day 28 (Fig. 3A), collagen increased to 1% of wet weight and in other hMSC Transwell cultures has approached 2% [31]. This is still much below mature cartilage, but it is comparable to 4-month-old human fetal cartilage (3% collagen by wet weight). Of potentially greater importance was the further evidence of the rapid extracellular processing of the collagen to form the essential components of a highly organized cross-linked collagen network combining the 3 major cartilage collagens.
Most of the collagen present at 7 days was due to an increase in type II collagen protein (α1(II) chains) (Fig. 5A). A transition from type IIA to type IIB mRNA transcripts was observed in the first 5 days of chondrogenesis in Transwell cultures [31]. Whether this transition plays a part in chondrogenesis, or occurs during a stage when the splicing machinery for the type IIB isoform is not completely expressed is still debatable. We have recently shown no effect on chondrogenesis and skeletal development in a transgenic mouse expressing only the type IIA collagen isoform [20]. In the cartilage constructs at the protein level we could not determine whether there was a type IIA in the matrix since both forms of type II N-propeptides were likely cleaved either by the N-propeptidase ADAM-TS3 [40–42], or by the pepsin required to solubilize the collagen network for analysis. However, over the major phase of matrix production, days 5 to 28, type IIB was the mRNA transcript expressed [31].
Despite the continued increase in proteoglycan content, the rate of gain in wet mass was less from day 14 to day 28 (Table 1, [31]). During the same period collagen content also increased, principally type II and type XI (Fig. 5A) and the collagen trivalent cross-link content approached levels present in 15-week human fetal cartilage (Table 1). This is consistent with a robust collagen network having been formed by day 14 and continued to develop until day 28. This network appears to have restricted further swelling of the matrix by the increasing proteoglycan content [43,44] forming a more rigid cartilage-like tissue. As the formation of mature collagen cross-links requires extracellular processing by lysyl oxidase of specific lysine residues in collagen fibrils, the generation of pyridinoline cross-links by day 7 is entirely compatible with the reported 5-fold up-regulation of LOX gene expression in comparable hMSC Transwell cultures between days 3 and 6 and continued expression thereafter (Barter and Young, University of Newcastle, UK; personal communication). In an earlier comparison of the Transwell format with cell pellet cultures [31], the Transwell disk lost much less aggrecan to the medium over 28 days of culture, compared to a spherical pellet (24% compared to 36%), although the surface area of the disk was much larger. The proportion of collagen lost was not measured at this time, but as aggrecan is a diffusible secreted product its retention in the nascent construct is very dependent on the developing integrity of the collagen matrix. It was also noticeable that less aggrecan was lost as the culture matured [31], which correlates well with the increase of 63% in collagen content between days 14 and 28 and the maturation of cross-links revealed in our new analysis. The expression of lysyl oxidase from day 7 as noted above may be an important factor facilitating this rapid maturation. Different forms of mechanical stimulation have been shown to boost cartilage matrix formation [45] and it would therefore be important to investigate if these factors could further enhance the maturation of the collagen network in these Transwell cultures.
No significant amount of type X collagen was detected in the cultures up to day 28. Although this appears to contrast with the significant expression of COL10A1 mRNA during hMSC chondrogenesis in pellet cultures [30], it is consistent with barely detectable levels of type X collagen by immunohistochemistry in Transwell cultures [31]. Chondrogenesis to a permanent hyaline cartilage phenotype rather than a transient endochondral phenotype was thus maintained up to day 28 of culture.
The present findings show that as hMSCs undergo chondrogenesis, type II collagen molecules rapidly become inter-molecularly cross-linked to other type II molecules (Fig. 3, 10F2), to type XI (Fig. 3, 5890) and type IX collagen molecules (Figs. 3 and 2B4), as we have previously shown for epiphyseal and articular cartilage in vivo [2,4,46]. We have used this western blotting methodology to demonstrate the deposition of cross-linked, nascent type II N-procollagen fibrils in the matrix of the Swarm rat chondrosarcoma cell line, RCS-LTC [33], and characterize type IIA–XI fibril heteropolymers in transgenic mice expressing exclusively the type IIA isoform [20]. Under polarized light microscopy, the matrix of the hMSC constructs revealed a diffuse barely discernible birefringence typical of thin randomly ordered fibrils and electron microscopy confirmed a network of thin fibrils [31] very similar to that seen in native human epiphyseal cartilage [15]. The present data suggest that by day 28 (Figs. 3 and 4) the nascent collagen was assembled into a heteropolymeric fibril network typical of hyaline cartilage.
The post-translational quality of the type II collagen in the cartilage constructs was also examined. The degree of 3-hydroxylation of specific proline residues (Pro-944 and Pro-986) suspected to be involved in supramolecular assembly of collagen molecules was also informative. As in native cartilage Pro-986 was 90–97% 3-hydroxylated (Fig. 5) at all stages. However Pro-944 was highly 3-hydroxylated in the early days of culture and decreased to 62% 3-hydroxylation by day 28 in contrast to ≤3% 3-hydroxylation in fetal cartilage in vivo. The (GPP)4-repeat at the C-terminus of α1(II) although a candidate for proline 3-hydroxylation [21] showed negligible hydroxylation. It is notable that type II collagen fibrils in the nucleus pulposus of the intervertebral disk, vitreous of the eye and Swarm rat chondrosarcoma (RCS-LTC cell line) matrix are especially thin, ranging in diameter from 6 nm–20 nm [33,47]). The degree of 3-hydroxylation of α1(II) Pro-944 in all these collagens is high, ranging from 40% in the nucleus pulposus, 57% in the RCS-LTC matrix [21] to 90% in the vitreous [5]. To speculate, perhaps thin fibrils of type II collagen deposited in the early stages of chondrogenesis are in part regulated by increased 3-hydroxylation of P944 due to increased prolyl 3-hydroxylase 2 (P3H2) expression [21]. From a tissue engineering perspective this could be significant as the ability to modulate the thickness of the collagen fibrils by selectively inhibiting or stimulating P3H2, is an attractive hypothesis. Further studies are required to test this concept.
Experimental procedures
Mesenchymal stem cell culture, chondrogenic differentiation and histology
Human bone marrow mesenchymal stem cells were isolated by adherence to tissue culture plastic from the mononuclear cell fraction of human bone marrow (Stem Cell Technologies, Cambridge, UK) and expanded in monolayer culture in MSC Growth Medium supplemented with FGF2 [31]. Cells (at passages 2–5) were dissociated with trypsin and transferred to serum-free chondrogenic medium containing TGFβ3, dexamethasone, ascorbic acid 2-phosphate, sodium pyruvate, proline and ITS +1 as previously described [32]. Aliquots of cells (5 × 105) were centrifuged in 6.5 mm diameter Corning Trans-well™ filter units. Cultures were maintained at 37 °C, 5% CO2 for 3, 7, 14, and 28 days for cartilage formation. Growth media were replenished on a two-day interval. Cartilage constructs were frozen for future biochemical analysis.
For histology cultures were fixed in 4% paraformaldehyde and processed into paraffin wax. Sections were cut at 5 μm and distribution of polyanions revealed by staining with Safranin O/Fast Green.
Total collagen and collagen cross-link analysis
Samples were thawed, drained of excess liquid and weighed. They were hydrolyzed in 6 M HCl at 110 °C for 24 h. An aliquot of the hydrolysate was colorimetrically assayed for hydroxyproline as a measure of total collagen content [48]. Collagen content was expressed as percentage of collagen by mass of wet tissue. An acid hydrolysate of human fetal cartilage (15 weeks in utero) was used as a comparative control. Quantitation of pyridinoline cross-linking residues was carried out by reverse-phase HPLC [49,50]. Pyridinoline content was expressed as moles per mole of collagen.
Analysis of post-translational modifications of type II collagen chains
Following SDS-PAGE and Coomassie blue staining of collagen extracted from hMSC-derived engineered cartilage, in-gel trypsin digests of collagen chains were analyzed on an LCQ Deca XP ion-trap mass spectrometer (MS) with an in-line a C8 column (Thermo Finnigan). Tryptic peptides were identified by calculating the theoretical MS ion profile and matching these to the experimentally determined mass spectrometric peptides [21]. The percentage 3-hydroxylation at a particular proline site was determined from the relative abundance of the 3Hyp-containing ion as a fraction of the sum of both 3Hyp- and Pro-containing tryptic peptide ions [21].
Collagen heteropolymer analysis
The collagen network laid down by differentiating hMSCs was depolymerized and extracted using 500 μg/mL pepsin in 3% acetic acid. A pepsin extract of type II collagen from fetal cartilage (15 weeks in utero) was used as a standard. Collagen chains and chain fragments were resolved by SDS-PAGE [51] and transferred to PVDF membrane for western blot analysis using a series of primary antibodies by probing, stripping and reprobing between each blot. [34]. The monoclonal antibody (mAb) 10F2 recognizes a cleavage site (neoepitope) generated on cross-linked forms of the type II collagen C-telopeptide [20,33]. The monoclonal antibody 2B4 recognizes an amino acid sequence in the α1(IX) chain that in normal cartilage tissue is cross-linked to the α1(II) chain [35]. Extracts were additionally probed with the polyclonal antibody 5890, that recognizes a sequence in the N-terminal region of the α1(XI) chain that is cross-linked to other chains of type XI collagen and type II collagen [20]. The blots were subsequently probed with mAb 1C10, which recognizes an epitope in the helical region of type II collagen chains [40]. A chemiluminescence reporting system was used to detect protein bands. The blots were stripped in 62.5 mM Tris–HCl buffer, pH 6.8 containing 2% SDS, and 100 mM 2-mercaptoethanol for 40 min at 50 C.
Acknowledgments
This work was supported in whole or in part by NIH grants AR057025 (R.J.F.) and AR037318 (D.R.E.) and a UK Research Council grant to the UK Centre for Tissue Engineering with support from The Wellcome Trust to Wellcome Trust Centre for Cell-Matrix Research, University of Manchester. The authors would like to thank Sara E. Funk, Geoffrey R. Traeger and Mary Ann Weis for their technical expertise. R.J.F. would like to gratefully acknowledge Dr. Audrey McAlinden for encouragement and helpful discussions.
Footnotes
Author contributions
ADM: Conception of project, experimental design, laboratory effort, manuscript writing.
THE, DRE: Conception of project, manuscript writing.
RJF: Conception of project, experimental design, laboratory effort, interpretation of results, figure design, manuscript writing.
References
- 1.Eyre DR, Wu JJ, Fernandes RJ, Pietka TA, Weis MA. Recent developments in cartilage research: matrix biology of the collagen II/IX/XI heterofibril network. Biochem Soc Trans. 2002;30:893–899. doi: 10.1042/bst0300893. [DOI] [PubMed] [Google Scholar]
- 2.Wu JJ, Eyre DR. Identification of hydroxypyridinium cross-linking sites in type II collagen of bovine articular cartilage. Biochemistry. 1984;23:1850–1857. doi: 10.1021/bi00303a041. [DOI] [PubMed] [Google Scholar]
- 3.Wu JJ, Eyre DR. Cartilage type IX collagen is cross-linked by hydroxypyridinium residues. Biochem Biophys Res Commun. 1984;123:1033–1039. doi: 10.1016/s0006-291x(84)80237-5. [DOI] [PubMed] [Google Scholar]
- 4.Wu JJ, Eyre DR. Structural analysis of cross-linking domains in cartilage type XI collagen. Insights on polymeric assembly. J Biol Chem. 1995;270:18865–18870. doi: 10.1074/jbc.270.32.18865. [DOI] [PubMed] [Google Scholar]
- 5.Weis MA, Hudson DM, Kim L, Scott M, Wu JJ, Eyre DR. Location of 3-hydroxyproline residues in collagen types I, II, III, and V/XI implies a role in fibril supramolecular assembly. J Biol Chem. 2010;285:2580–2590. doi: 10.1074/jbc.M109.068726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Morello R, Bertin TK, Chen Y, Hicks J, Tonachini L, Monticone M, Castagnola P, Rauch F, Glorieux FH, Vranka J, Bachinger HP, Pace JM, Schwarze U, Byers PH, Weis M, Fernandes RJ, Eyre DR, Yao Z, Boyce BF, Lee B. CRTAP is required for prolyl 3-hydroxylation and mutations cause recessive osteogenesis imperfecta. Cell. 2006;127:291–304. doi: 10.1016/j.cell.2006.08.039. [DOI] [PubMed] [Google Scholar]
- 7.Baldridge D, Schwarze U, Morello R, Lennington J, Bertin TK, Pace JM, Pepin MG, Weis M, Eyre DR, Walsh J, Lambert D, Green A, Robinson H, Michelson M, Houge G, Lindman C, Martin J, Ward J, Lemyre E, Mitchell JJ, Krakow D, Rimoin DL, Cohn DH, Byers PH, Lee B. CRTAP and LEPRE1 mutations in recessive osteogenesis imperfecta. Hum Mutat. 2008;29:1435–1442. doi: 10.1002/humu.20799. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Cabral WA, Barnes AM, Adeyemo A, Cushing K, Chitayat D, Porter FD, Panny SR, Gulamali-Majid F, Tishkoff SA, Rebbeck TR, Gueye SM, Bailey-Wilson JE, Brody LC, Rotimi CN, Marini JC. A founder mutation in LEPRE1 carried by 1.5% of West Africans and 0.4% of African Americans causes lethal recessive osteogenesis imperfecta. Genet Med. 2012;14:543–551. doi: 10.1038/gim.2011.44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Marini JC, Cabral WA, Barnes AM. Null mutations in LEPRE1 and CRTAP cause severe recessive osteogenesis imperfecta. Cell Tissue Res. 2010;339:59–70. doi: 10.1007/s00441-009-0872-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pyott SM, Schwarze U, Christiansen HE, Pepin MG, Leistritz DF, Dineen R, Harris C, Burton BK, Angle B, Kim K, Sussman MD, Weis M, Eyre DR, Russell DW, McCarthy KJ, Steiner RD, Byers PH. Mutations in PPIB (cyclophilin B) delay type I procollagen chain association and result in perinatal lethal to moderate osteogenesis imperfecta phenotypes. Hum Mol Genet. 2011;20:1595–1609. doi: 10.1093/hmg/ddr037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Takagi M, Ishii T, Barnes AM, Weis M, Amano N, Tanaka M, Fukuzawa R, Nishimura G, Eyre DR, Marini JC, Hasegawa T. A novel mutation in LEPRE1 that eliminates only the KDEL ER-retrieval sequence causes non-lethal osteogenesis imperfecta. PLoS One. 2012;7:e36809. doi: 10.1371/journal.pone.0036809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hudson DM, Joeng KS, Werther R, Rajagopal A, Weis M, Lee BH, Eyre DR. Post-translationally abnormal collagens of prolyl 3-hydroxylase-2 null mice offer a pathobiological mechanism for the high myopia linked to human LEPREL1 mutations. J Biol Chem. 2015;290:8613–8622. doi: 10.1074/jbc.M114.634915. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mordechai S, Gradstein L, Pasanen A, Ofir R, El Amour K, Levy J, Belfair N, Lifshitz T, Joshua S, Narkis G, Elbedour K, Myllyharju J, Birk OS. High myopia caused by a mutation in LEPREL1, encoding prolyl 3-hydroxylase 2. Am J Hum Genet. 2011;89:438–445. doi: 10.1016/j.ajhg.2011.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Guo H, Tong P, Peng Y, Wang T, Liu Y, Chen J, Li Y, Tian Q, Hu Y, Zheng Y, Xiao L, Xiong W, Pan Q, Hu Z, Xia K. Homozygous loss-of-function mutation of the LEPREL1 gene causes severe non-syndromic high myopia with early-onset cataract. Clin Genet. 2014;86:575–579. doi: 10.1111/cge.12309. [DOI] [PubMed] [Google Scholar]
- 15.Fernandes RJ, Wilkin DJ, Weis MA, Wilcox WR, Cohn DH, Rimoin DL, Eyre DR. Incorporation of structurally defective type II collagen into cartilage matrix in kniest chondrodysplasia. Arch Biochem Biophys. 1998;355:282–290. doi: 10.1006/abbi.1998.0745. [DOI] [PubMed] [Google Scholar]
- 16.Prockop DJ, Kivirikko KI, Tuderman L, Guzman NA. The biosynthesis of collagen and its disorders (second of two parts) N Engl J Med. 1979;301:77–85. doi: 10.1056/NEJM197907123010204. [DOI] [PubMed] [Google Scholar]
- 17.Prockop DJ, Kivirikko KI, Tuderman L, Guzman NA. The biosynthesis of collagen and its disorders (first of two parts) N Engl J Med. 1979;301:13–23. doi: 10.1056/NEJM197907053010104. [DOI] [PubMed] [Google Scholar]
- 18.Fernandes RJ, Seegmiller RE, Nelson WR, Eyre DR. Protein consequences of the Col2a1 C-propeptide mutation in the chondrodysplastic Dmm mouse. Matrix Biol. 2003;22:449–453. doi: 10.1016/s0945-053x(03)00077-5. [DOI] [PubMed] [Google Scholar]
- 19.Bogaert R, Wilkin D, Wilcox WR, Lachman R, Rimoin D, Cohn DH, Eyre DR. Expression, in cartilage, of a 7-aminoacid deletion in type II collagen from two unrelated individuals with Kniest dysplasia. Am J Hum Genet. 1994;55:1128–1136. [PMC free article] [PubMed] [Google Scholar]
- 20.McAlinden A, Traeger G, Hansen U, Weis MA, Ravindran S, Wirthlin L, Eyre DR, Fernandes RJ. Molecular properties and fibril ultrastructure of types II and XI collagens in cartilage of mice expressing exclusively the alpha1(IIA) collagen isoform. Matrix Biol. 2014;34:105–113. doi: 10.1016/j.matbio.2013.09.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Fernandes RJ, Farnand AW, Traeger GR, Weis MA, Eyre DR. A role for prolyl 3-hydroxylase 2 in post-translational modification of fibril-forming collagens. J Biol Chem. 2011;286:30662–30669. doi: 10.1074/jbc.M111.267906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Hudson DM, Kim LS, Weis M, Cohn DH, Eyre DR. Peptidyl 3-hydroxyproline binding properties of type I collagen suggest a function in fibril supramolecular assembly. Biochemistry. 2012;51:2417–2424. doi: 10.1021/bi2019139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Prockop DJ. Repair of tissues by adult stem/progenitor cells (MSCs): controversies, myths, and changing paradigms. Mol Ther. 2009;17:939–946. doi: 10.1038/mt.2009.62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Cheng A, Hardingham TE, Kimber SJ. Generating cartilage repair from pluripotent stem cells. Tissue Eng Part B Rev. 2014;20:257–266. doi: 10.1089/ten.teb.2012.0757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Khan WS, Hardingham TE. Cartilage tissue engineering approaches applicable in orthopaedic surgery: the past, the present, and the future. J Stem Cells. 2012;7:97–104. [PubMed] [Google Scholar]
- 26.Khan WS, Johnson DS, Hardingham TE. The potential of stem cells in the treatment of knee cartilage defects. Knee. 2010;17:369–374. doi: 10.1016/j.knee.2009.12.003. [DOI] [PubMed] [Google Scholar]
- 27.Huang AH, Farrell MJ, Mauck RL. Mechanics and mechanobiology of mesenchymal stem cell-based engineered cartilage. J Biomech. 2010;43:128–136. doi: 10.1016/j.jbiomech.2009.09.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hardingham T. Cell- and tissue-based approaches for cartilage repair. Altern Lab Anim. 2010;38(Suppl 1):35–39. doi: 10.1177/026119291003801S10. [DOI] [PubMed] [Google Scholar]
- 29.Decker RS, Koyama E, Pacifici M. Genesis and morphogenesis of limb synovial joints and articular cartilage. Matrix Biol. 2014;39:5–10. doi: 10.1016/j.matbio.2014.08.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Johnstone B, Hering TM, Caplan AI, Goldberg VM, Yoo JU. In vitro chondrogenesis of bone marrow-derived mesenchymal progenitor cells. Exp Cell Res. 1998;238:265–272. doi: 10.1006/excr.1997.3858. [DOI] [PubMed] [Google Scholar]
- 31.Murdoch AD, Grady LM, Ablett MP, Katopodi T, Meadows RS, Hardingham TE. Chondrogenic differentiation of human bone marrow stem cells in transwell cultures: generation of scaffold-free cartilage. Stem Cells. 2007;25:2786–2796. doi: 10.1634/stemcells.2007-0374. [DOI] [PubMed] [Google Scholar]
- 32.Tew SR, Murdoch AD, Rauchenberg RP, Hardingham TE. Cellular methods in cartilage research: primary human chondrocytes in culture and chondrogenesis in human bone marrow stem cells. Methods. 2008;45:2–9. doi: 10.1016/j.ymeth.2008.01.006. [DOI] [PubMed] [Google Scholar]
- 33.Fernandes RJ, Schmid TM, Eyre DR. Assembly of collagen types II, IX and XI into nascent hetero-fibrils by a rat chondrocyte cell line. Eur J Biochem. 2003;270:3243–3250. doi: 10.1046/j.1432-1033.2003.03711.x. [DOI] [PubMed] [Google Scholar]
- 34.Fernandes RJ, Weis M, Scott MA, Seegmiller RE, Eyre DR. Collagen XI chain misassembly in cartilage of the chondrodysplasia (cho) mouse. Matrix Biol. 2007;26:597–603. doi: 10.1016/j.matbio.2007.06.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Eyre DR, Pietka T, Weis MA, Wu JJ. Covalent cross-linking of the NC1 domain of collagen type IX to collagen type II in cartilage. J Biol Chem. 2004;279:2568–2574. doi: 10.1074/jbc.M311653200. [DOI] [PubMed] [Google Scholar]
- 36.Gibson GJ, Francki KT, Hopwood JJ, Foster BK. Human and sheep growth-plate cartilage type X collagen synthesis and the influence of tissue storage. Biochem J. 1991;277(Pt 2):513–520. doi: 10.1042/bj2770513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Niyibizi C, Sagarrigo Visconti C, Gibson G, Kavalkovich K. Identification and immunolocalization of type X collagen at the ligament-bone interface. Biochem Biophys Res Commun. 1996;222:584–589. doi: 10.1006/bbrc.1996.0787. [DOI] [PubMed] [Google Scholar]
- 38.Huang AH, Yeger-McKeever M, Stein A, Mauck RL. Tensile properties of engineered cartilage formed from chondrocyte- and MSC-laden hydrogels. Osteoarthr Cartil. 2008;16:1074–1082. doi: 10.1016/j.joca.2008.02.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Oldershaw RA, Tew SR, Russell AM, Meade K, Hawkins R, McKay TR, Brennan KR, Hardingham TE. Notch signaling through Jagged-1 is necessary to initiate chondrogenesis in human bone marrow stromal cells but must be switched off to complete chondrogenesis. Stem Cells. 2008;26:666–674. doi: 10.1634/stemcells.2007-0806. [DOI] [PubMed] [Google Scholar]
- 40.Fernandes RJ, Hirohata S, Engle JM, Colige A, Cohn DH, Eyre DR, Apte SS. Procollagen II amino propeptide processing by ADAMTS-3. Insights on dermatosparaxis. J Biol Chem. 2001;276:31502–31509. doi: 10.1074/jbc.M103466200. [DOI] [PubMed] [Google Scholar]
- 41.Bekhouche M, Colige A. The procollagen N-proteinases ADAMTS2, 3 and 14 in pathophysiology. Matrix Biol. 2015;44–46:46–53. doi: 10.1016/j.matbio.2015.04.001. [DOI] [PubMed] [Google Scholar]
- 42.Dubail J, Apte SS. Insights on ADAMTS proteases and ADAMTS-like proteins from mammalian genetics. Matrix Biol. 2015;44–46:24–37. doi: 10.1016/j.matbio.2015.03.001. [DOI] [PubMed] [Google Scholar]
- 43.Maroudas AI. Balance between swelling pressure and collagen tension in normal and degenerate cartilage. Nature. 1976;260:808–809. doi: 10.1038/260808a0. [DOI] [PubMed] [Google Scholar]
- 44.Katz EP, Wachtel EJ, Maroudas A. Extrafibrillar proteoglycans osmotically regulate the molecular packing of collagen in cartilage. Biochim Biophys Acta. 1986;882:136–139. doi: 10.1016/0304-4165(86)90065-6. [DOI] [PubMed] [Google Scholar]
- 45.Bian L, Fong JV, Lima EG, Stoker AM, Ateshian GA, Cook JL, Hung CT. Dynamic mechanical loading enhances functional properties of tissue-engineered cartilage using mature canine chondrocytes. Tissue Eng A. 2010;16:1781–1790. doi: 10.1089/ten.tea.2009.0482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Wu JJ, Woods PE, Eyre DR. Identification of cross-linking sites in bovine cartilage type IX collagen reveals an antiparallel type II–type IX molecular relationship and type IX to type IX bonding. J Biol Chem. 1992;267:23007–23014. [PubMed] [Google Scholar]
- 47.Le Goff MM, Bishop PN. Adult vitreous structure and postnatal changes. Eye (Lond) 2008;22:1214–1222. doi: 10.1038/eye.2008.21. [DOI] [PubMed] [Google Scholar]
- 48.Fernandes RJ, Harkey MA, Weis M, Askew JW, Eyre DR. The post-translational phenotype of collagen synthesized by SAOS-2 osteosarcoma cells. Bone. 2007;40:1343–1351. doi: 10.1016/j.bone.2007.01.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Eyre DR, Koob TJ, Van Ness KP. Quantitation of hydroxypyridinium crosslinks in collagen by high-performance liquid chromatography. Anal Biochem. 1984;137:380–388. doi: 10.1016/0003-2697(84)90101-5. [DOI] [PubMed] [Google Scholar]
- 50.Fernandes RJ, Eyre DR. The elastin-like protein matrix of lamprey branchial cartilage is cross-linked by lysyl pyridinoline. Biochem Biophys Res Commun. 1999;261:635–640. doi: 10.1006/bbrc.1999.1092. [DOI] [PubMed] [Google Scholar]
- 51.Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–685. doi: 10.1038/227680a0. [DOI] [PubMed] [Google Scholar]