Human collagen X contains a highly conserved 161-amino acid C-terminal non-triple helical domain that is homologous to the C-terminal domain of collagen VIII and to the C1q module of the human C1 enzyme. We have expressed this domain (residues 545–680) inEscherichia coli as a glutathione S-transferase fusion protein. The purified fusion protein trimerizes spontaneouslyin vitro, and after thrombin cleavage, the purified C-terminal domain trimer (46.2 kDa) is extremely stable and trypsin-resistant. Mutations within the C-terminal domain have been observed in patients with Schmid's metaphyseal chondrodysplasia (SMCD). Some of these mutations (Y598D, G618V, W651X, or H669X; X is the stop codon) were constructed by site-directed mutagenesis. Each mutation had identical consequences regarding the fusion protein: 1) absence of trimeric formation, 2) copurification of the ∼60-kDa GroEL chaperone protein, and 3) sensitivity of the monomeric fusion protein to trypsin digestion. These results show that the C-terminal domain of collagen X is sufficient to produce a very stable and compact trimer in the absence of collagen Gly-X-Y repeats. Moreover, mutations causing SMCD interfere in this system with the correct folding of the C-terminal domain. The existence of a similar mechanism in chondrocytes might explain the relative homogeneity of phenotypes in SMCD despite the diversity of mutations. Human collagen X contains a highly conserved 161-amino acid C-terminal non-triple helical domain that is homologous to the C-terminal domain of collagen VIII and to the C1q module of the human C1 enzyme. We have expressed this domain (residues 545–680) inEscherichia coli as a glutathione S-transferase fusion protein. The purified fusion protein trimerizes spontaneouslyin vitro, and after thrombin cleavage, the purified C-terminal domain trimer (46.2 kDa) is extremely stable and trypsin-resistant. Mutations within the C-terminal domain have been observed in patients with Schmid's metaphyseal chondrodysplasia (SMCD). Some of these mutations (Y598D, G618V, W651X, or H669X; X is the stop codon) were constructed by site-directed mutagenesis. Each mutation had identical consequences regarding the fusion protein: 1) absence of trimeric formation, 2) copurification of the ∼60-kDa GroEL chaperone protein, and 3) sensitivity of the monomeric fusion protein to trypsin digestion. These results show that the C-terminal domain of collagen X is sufficient to produce a very stable and compact trimer in the absence of collagen Gly-X-Y repeats. Moreover, mutations causing SMCD interfere in this system with the correct folding of the C-terminal domain. The existence of a similar mechanism in chondrocytes might explain the relative homogeneity of phenotypes in SMCD despite the diversity of mutations. polyacrylamide gel electrophoresis Schmid's metaphyseal chondrodysplasia glutathione S-transferase matrix-assisted laser desorption ionization-time of flight 3-(cyclohexylamino)-1-propanesulfonic acid Collagen X is a non-fibril-forming collagen with a triple helical domain half the length of fibril-forming collagens. It is primarily expressed in hypertrophic cartilage of the epiphyseal growth plate of long bones, ribs, and vertebrae (1Schmid T.M. Linsenmayer T.F. Dev. Biol. 1985; 107: 373-381Crossref PubMed Scopus (183) Google Scholar, 2Gibson G.J. Flint M.H. J. Cell Biol. 1985; 101: 277-284Crossref PubMed Scopus (150) Google Scholar, 3Grant W.T. Sussman M.D. Balian G. J. Biol. Chem. 1985; 260: 3798-3803Abstract Full Text PDF PubMed Google Scholar); in bone fracture callus (4Grant W.T. Wang G.J. Balian G. J. Biol. Chem. 1987; 262: 9844-9849Abstract Full Text PDF PubMed Google Scholar); and in osteoarthritic cartilage (5Hoyland J.A. Thomas J.T. Donn R. Marriott A. Ayad S. Boot Handford R.P. Grant M.E. Freemont A.J. Bone Miner. 1991; 15: 151-163Abstract Full Text PDF PubMed Scopus (110) Google Scholar, 6von der Mark K. Kirsch T. Nerlich A. Kuss A. Weseloh G. Gluckert K. Stoss H. Arthritis Rheum. 1992; 35: 806-811Crossref PubMed Scopus (407) Google Scholar). Human collagen X is a homotrimer composed of a 38-amino acid non-triple helical domain (NC2) at the N terminus, a 463-amino acid triple helical domain, and a 161-amino acid non-triple helical domain (NC1) at the C terminus (see Fig. 1). Unlike the C-terminal propeptide of fibrillar collagens, the C-terminal NC1 domain is not cleaved following secretion into the extracellular space. Recent cloning data have shown that there is a family of molecules that contain C-terminal domains homologous to the NC1 domain of collagen X. These include collagen VIII (7Yamaguchi N. Mayne R. Ninomiya Y. J. Biol. Chem. 1991; 266: 4508-4513Abstract Full Text PDF PubMed Google Scholar); the a, b, and c chains of subunit C1q of the human C1 enzyme complex (8Sellar G.C. Blake D.J. Reid K.B. Biochem. J. 1991; 274: 481-490Crossref PubMed Scopus (183) Google Scholar); the adipocyte complement-related protein of 30 kDa (ACRP30) (9Hu E. Liang P. Spiegelman B.M. J. Biol. Chem. 1996; 271: 10697-10703Abstract Full Text Full Text PDF PubMed Scopus (1916) Google Scholar); an inner ear-specific collagen molecule (10Davis J.G. Oberholtzer J.C. Burns F.R. Greene M.I. Science. 1995; 267: 1031-1034Crossref PubMed Scopus (63) Google Scholar); two chains of precerebellin (11Kavety B. Jenkins N.A. Fletcher C.F. Copeland N.G. Morgan J.I. Brain. Res. Mol. Brain Res. 1994; 27: 152-156Crossref PubMed Scopus (25) Google Scholar); three plasma proteins associated with mammalian hibernation (12Takamatsu N. Ohba K. Kondo J. Kondo N. Shiba T. Mol. Cell. Biol. 1993; 13: 1516-1521Crossref PubMed Scopus (88) Google Scholar); and multimerin (13Hayward C.P. Hassell J.A. Denomme G.A. Rachubinski R.A. Brown C. Kelton J.G. J. Biol. Chem. 1995; 270: 18246-18251Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar). This domain has been called the C1q module (14Bork P. Koonin E.V. Curr. Opin. Struct. Biol. 1996; 6: 366-376Crossref PubMed Scopus (124) Google Scholar, 15Hegyi H. Bork P. J. Protein Chem. 1997; 16: 545-551Crossref PubMed Scopus (29) Google Scholar). A sequence comparison between the C termini of collagens X and VIII, subunit C1q of the human C1 enzyme complex, and the fibrillar collagens has revealed a conserved cluster of aromatic residues (16Brass A. Kadler K.E. Thomas J.T. Grant M.E. Boot Handford R.P. FEBS Lett. 1992; 303: 126-128Crossref PubMed Scopus (69) Google Scholar). This cluster is found in a region of ∼130 amino acids that exhibits marked similarities in hydrophilicity profiles between the different collagens, despite a low level of sequence similarity. Fourier transform infrared spectroscopy and structure predictions suggested that the trimeric C1q module presents a pattern of 10 β-strands interspersed by β-turns and/or loops (17Smith K.F. Haris P.I. Chapman D. Reid K.B. Perkins S.J. Biochem. J. 1994; 301: 249-256Crossref PubMed Scopus (30) Google Scholar). This was confirmed by the determination of the three-dimensional structure of the homotrimeric ACRP30 C-terminal domain, which is clearly homologous to its C1q counterpart. This structure revealed an asymmetrical trimer of β-sandwich protomers. Each of the protomers has a 10-strand jelly-roll folding topology similar to the tumor necrosis factor family (18Shapiro L. Scherer P.E. Curr. Biol. 1998; 8: 335-338Abstract Full Text Full Text PDF PubMed Google Scholar). It was inferred from sequence homologies that this folding is common to all C1q module family proteins. The trimer is bell-shaped, with a wide base. Trimer contacts form primarily through a cluster of hydrophobic interactions near the base. The trimer interface near the apex is largely hydrophilic and contains many ordered water molecules. The importance of the interactions of collagen X NC1 domains in the assembly of the molecule was suggested by the observation that chick collagen X containing the triple helical and NC1 domains renatures much more rapidly than collagen X containing only the triple helical domain (19Schmid T.M. Linsenmayer T.F. Biochemistry. 1984; 23: 553-558Crossref PubMed Scopus (40) Google Scholar). The isolated NC1 domain of collagen X purified after bacterial collagenase digestion was studied (19Schmid T.M. Linsenmayer T.F. Biochemistry. 1984; 23: 553-558Crossref PubMed Scopus (40) Google Scholar, 20Barber R.E. Kwan A.P. Biochem. J. 1996; 320: 479-485Crossref PubMed Scopus (14) Google Scholar). The existence of very stable multimeric forms, even in the absence of covalent binding between the subunits, was established. The NC1 domain from human procollagen X expressed as recombinant protein in a human kidney epithelial cell line (HEK293) migrates as multimers with an apparent molecular mass of 40 kDa on SDS-PAGE,1 even after reduction and heat denaturation, and gives rise to monomers of 18–20 kDa after treatment with trichloroacetic acid (21Frischholz S. Beier F. Girkontaite I. Wagner K. Poschl E. Turnay J. Mayer U. von der Mark K. J. Biol. Chem. 1998; 273: 4547-4555Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar). Schmid's metaphyseal chondrodysplasia (SMCD; MIM 156500) is a mild autosomal disorder of the osseous skeleton associated with growth plate abnormalities, short stature, and waddling gait. All the mutations detected so far have been located in the NC1 domain of collagen X and include substitutions, deletions, and premature terminations (Refs.22Warman M.L. Abbott M. Apte S.S. Hefferon T. McIntosh I. Cohn D.H. Hecht J.T. Olsen B.R. Francomano C.A. Nat. Genet. 1993; 5: 79-82Crossref PubMed Scopus (222) Google Scholar, 23Wallis G.A. Rash B. Sweetman W.A. Thomas J.T. Super M. Evans G. Grant M.E. Boot Handford R.P. Am. J. Hum. Genet. 1994; 54: 169-178PubMed Google Scholar, 24McIntosh I. Abbott M.H. Warman M.L. Olsen B.R. Francomano C.A. Hum. Mol. Genet. 1994; 3: 303-307Crossref PubMed Scopus (52) Google Scholar; for review, see Ref. 25Chan D. Jacenko O. Matrix Biol. 1998; 17: 169-184Crossref PubMed Scopus (80) Google Scholar). Several experiments have demonstrated the role of the NC1 domain of collagen X in the trimerization process. Pre-α1(X) collagen chains have been synthesized with NC1 mutations of those found in SMCD, including G618V (26Chan D. Cole W.G. Rogers J.G. Bateman J.F. J. Biol. Chem. 1995; 270: 4558-4562Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar) and 1952delC, 1953del10, and Y598D (27Chan D. Weng Y.M. Hocking A.M. Golub S. McQuillan D.J. Bateman J.F. J. Biol. Chem. 1996; 271: 13566-13572Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). Contrary to the control construct, the mutant molecules were unable to assemble into homotrimers in vitro. Furthermore, the mutant chains did not associate with normal α1(X) chains and did interfere with the efficiency of normal chain assembly in cotranslation experiments. Expression in transiently transfected cells showed that the mutant chains are not secreted from the cells and are probably rapidly degraded intracellularly. In contrast, in-frame deletions within the triple helical domain and the NC2 domain, obtained by site-directed did not and the mutant chains were secreted as triple helical homotrimers (27Chan D. Weng Y.M. Hocking A.M. Golub S. McQuillan D.J. Bateman J.F. J. Biol. Chem. 1996; 271: 13566-13572Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). In this have expressed in the residues of the NC1 domain of collagen X. This region to a C1q module as by Bork and Koonin (14Bork P. Koonin E.V. Curr. Opin. Struct. Biol. 1996; 6: 366-376Crossref PubMed Scopus (124) Google Scholar) and has been expressed as a glutathione S-transferase fusion protein by as were to the trimerization of the NC1 domain and to the of SMCD mutations on this process. all were obtained from The bacterial were obtained from mass were obtained from A to amino acids to of human collagen X was obtained by chain from S.S. M. Olsen B.R. J. Biochem. 1992; PubMed Scopus (88) Google Scholar) of S. were within the sequence and the with in a sequence to the C-terminal region The obtained was purified and into the cloning to the of the the recombinant was with and into the with the The the C1q module domain of human collagen X with the control of The is to as the fusion protein The recombinant was by the the chain F. S. U. S. A. PubMed Scopus Google Scholar). coli cells were in containing of at with at The of was by of of the A cells were by and in containing 40 of with and of as was at a of and the cells were by on of by of 10 for a of Cell of the cell at for 30 the was on a The was at with of gel in an of the gel with 10 of was with of A containing 10 was of to the glutathione and with thrombin of in an for at was by of The was at with of gel in an to the cleaved protein. The NC1 domain was in the A protein was to a system The was with and by a in the was on in a to PubMed Scopus Google Scholar) a the as and the were for at The were with in and with A with the sequence to amino acids to of human collagen was synthesized by a an was on a and was for The purified was in and was in a of of The was to with a of the was as and the was for at in was with and for at The was by the with of of of and of the The was for at The to was and and for the of The of the to the NC1 domain of human collagen X was confirmed by the that all the mutant and fusion proteins were the mutant molecule that the sequence to the human collagen X NC1 domain SMCD mutations were by site-directed from by the from containing the of U. S. A. 1985; PubMed Scopus Google Scholar). The was into The was with the and site-directed was the from the and the and for the mutations G618V, W651X, and for premature The mutant were by the the chain F. S. U. S. A. PubMed Scopus Google Scholar) to the of the mutations and to the absence of acid were by in an protein the acid by the The was on line with an were obtained with a of flight mass with a A mass of was by laser at the All were in the with a the was to and the to of this or The in the flight was at of the of of the protein and of the were on the plate and in the to All experiments were a of acid in a of and was with mass was a triple mass with a The was at and the at The interface was at of were at a of with acid as were in the at from to including three of the protein, with a of and a of was in the were a data system the gel was in and for this a was with and in The between a of and of in was into a at in the was with with for a and with in and acid for The was with and with The of was and to as the a and three of were in and for The three of were on the plate of a and by the and the of in and were on of the at 40 the was rapidly in to the proteins and the molecular mass It was with water and with containing and low The were for with different in the for the for the and for the of S. M. der of in containing and for with for of the NC1 and GroEL proteins or with for of the protein. A of human collagen X similar to the C1q module of the human C1 complex (residues 545–680) has been into the cells expressed a protein with This apparent molecular mass is to the from the sequence of of and kDa were on a with with the and not that the to monomeric and the to a multimeric form of this protein. 3) was with and the of the was by with apparent molecular of and kDa were to the not to the The kDa) were only to the and to the protein and a The was by and the NC1 domain was by protein on a to the and purified proteins The an by the with an apparent molecular mass of kDa of the only the of kDa the was with trichloroacetic acid electrophoresis and and the of kDa and only the of kDa was not The purified NC1 domain was to The obtained sequence to the NC1 domain. The two amino acids from the protein as are after the thrombin 1). A molecular mass of was for the NC1 domain by mass This molecular mass is by with the mass of the NC1 domain This to the of the three mass detected three 3) with molecular of and comparison with the these are to to the the and the trimer of the NC1 domain of collagen X a of the of the molecular of the the and the trimer of the NC1 domain by and by with the mass by in a All the expressed mutant proteins (see Fig. 1) had apparent molecular on that were with the mutations molecular mass were observed for the with the G618V and Y598D amino acid and the premature stop at amino acids and in of the apparent molecular of and and with the mutant containing the premature stop at amino acid was not by the the was by the of revealed the of with apparent molecular from to kDa The sequence of the to the GroEL chaperone protein from E. coli The sequence of the a of amino acids at the of the protein. The chaperone protein was not found in these not We to the of the of the GroEL chaperone protein with the We that the fusion protein the glutathione on the gel by its and that the GroEL chaperone protein with the NC1 domain the results for the Y598D mutant are in are identical for the three not The mutant was on and the gel was into The were and by as in the of Fig. The was in the of as a these all the proteins to the gel 1). the of the of GroEL by the with A of the GroEL chaperone protein, by with the was A of the detected by the and not by was The was with 10 glutathione to the mutant from the gel of the mutant and of the GroEL chaperone protein was detected by and was confirmed by with the different Some protein and degraded were these The was with thrombin to the GroEL chaperone protein to thrombin molecular mass of and to the mutant NC1 domain molecular mass of kDa) were The GroEL chaperone protein and the mutant NC1 were by A with an apparent molecular mass of kDa was detected with the It to the of the mutant NC1 domain, as the electrophoresis was A very molecular mass of detected only with the was observed in and was not of the Y598D The Y598D mutant fusion protein was purified as gel to which the mutant protein was was with A. The gel was in and with of A containing A containing or A containing 40 glutathione These three were for at in an by a at The 3) was with of A containing of thrombin and for at The were through and of the were by and with the with the with the of the molecular mass proteins are to and mutant proteins were with trypsin for at and by reduction and In this a after was only as a treatment of the protein two to the NC1 trimer and to the protein. A at kDa was not to the and has not been The sequence of the NC1 domain was identical to the obtained after thrombin treatment for at or for at in the of did not in the digestion of the NC1 domain, to a which was not trypsin treatment of the all were degraded the protein and the with an of All the mutant NC1 domains are to trypsin to the protein. The NC1 domain of collagen X is for the trimerization of the It has been suggested that this for the of collagen X is in SMCD patients NC1 mutations. the molecular SMCD, have expressed in E. coli the and mutant C1q of collagen the C-terminal residues of the NC1 domain. The protein was as a fusion with and the NC1 domain was purified to homogeneity following by and NC1 in as a stable and compact trimer to the following line of the apparent molecular mass of the was for a at is with the molecular mass of a the the NC1 domain is found as a and this is probably to the of the process. of the purified NC1 domain by mass revealed the of three with molecular to those of the and trimeric of that mass produce and these results only of the of these three NC1 in have found a of for NC1 in the of by not of was to the NC1 to a level with the The for NC1 are for the for the and for the the of might have an on the these data only as an of in The between the molecular mass of the trimer by or mass and by that the NC1 trimer has a compact This is by the and of the protein. Moreover, the three-dimensional structure of ACRP30 (18Shapiro L. Scherer P.E. Curr. Biol. 1998; 8: 335-338Abstract Full Text Full Text PDF PubMed Google a of that the C1q domains form that are by an hydrophobic The of a low for the of the NC1 trimer into its monomers that interactions in the of the The that the of an NC1 trimer stable a after from that the trimerization is or that a trimer a that the The is by the that the trimer into a and/or these results that the of collagen Gly-X-Y is for the of a stable and compact C-terminal domain trimer of collagen X. All the mutations so far in SMCD are located in the NC1 domain of collagen X. of these mutations have been in and had identical consequences regarding the fusion protein. the of in the of the bacterial was at than that of Moreover, the of a trimer was even after a of of the the sensitivity of the monomeric fusion protein to trypsin digestion was with the protein. This that the NC1 not into a coli GroEL chaperone protein with the mutant NC1 the GroEL chaperone protein with the that the not The of of the and the fusion protein did not between the and was very different not It is that the NC1 from the with the GroEL chaperone protein, in the bacterial The of GroEL chaperone protein was found to to the mutant an of the of the This was confirmed by the of the complex, not of from the complex to a after by the of the GroEL chaperone protein by a of NC1 The of chaperone proteins the and folding of collagen molecules has been The which is the of the enzyme as a for and premature assembly or of fibrillar procollagen R. J.F. J. Biol. Chem. 1998; 273: Full Text Full Text PDF PubMed Scopus Google Scholar). The transiently with collagen X in a role to its in the complex Biochem. J. 1998; PubMed Scopus (46) Google Scholar). chick are or stable triple is by treatment with of with procollagen is This is to the of procollagen secretion and to the of procollagen in the A. M. K. K. J. Cell Biol. 1992; PubMed Scopus Google Scholar). It has been shown that procollagen molecules mutations in the C-terminal propeptide to the protein or J. Biol. Chem. 1993; Full Text PDF PubMed Google Scholar). observation that NC1 domains of collagen X SMCD mutations with the GroEL chaperone protein of E. coli that in mammalian the mutant NC1 domains with We have a molecular of the NC1 domain of collagen X on the structure of the ACRP30 protein (18Shapiro L. Scherer P.E. Curr. Biol. 1998; 8: 335-338Abstract Full Text Full Text PDF PubMed Google Scholar). The of the was by the of between the of C1q of ACRP30 and collagen X. The mutation Y598D is located in the containing the that was suggested to in the of trimeric results the that folding of the protein of the mutant protein the of This is by the that three which are from the aromatic the in SMCD patients with despite the diversity of the mutations. A mechanism of mutant protein by a molecular chaperone as in similar to the observed in E. explain this We M. for mass S. M. der for the of the S. Apte for the of the human collagen X J. P. for sequence J. F. for J. C. for the of the C1q module of collagen L. for the for the structure of and S. for the
No takes yet. Share an insight, caveat, or question.
Dublet et al. (1999) studied this question.
Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context: