Eur. J. Biochem. 192, 703-708 (1990) FEBS 1990
The complete primary structure of two distinct forms of human al(1X) collagen chains Yasuteru MURAGAKI, Tomoatsu KIMURA, Yoshifumi NINOMIYA and Bjorn Reino OLSEN Department of Anatomy and Cellular Biology, Harvard Medical School, Boston, USA (Received April 30, 1990) - EJB 90 0492
Type IX collagen molecules contain three genetically distinct subunits. One of the subunits, a2(IX), contains a covalently attached glycosaminoglycan side chain. A second subunit, a1(IX), has been found to be synthesized in two forms. The two forms are generated by the alternative use of two transcription start sites and splice patterns. The two forms have been found in chicken, mouse and human but cDNAs encoding both forms have only been reported for chicken. In the present report we describe the isolation of cDNA clones encoding the complete translated portion of both forms of human a1 (IX) collagen chains. Nucleotide sequence analysis has permitted the determination of the primary structure of both forms. These probes and sequences should prove useful in future studies of chondrodysplasias involving type IX collagen Type IX collagen-proteoglycan (collagen IX) is a multidomain extracellular matrix component [l, 21 which is associated with type-11-containing collagen fibrils in cartilage [3] and non-cartilage tissues such as the vitreous [4] and embryonic cornea [5, 61. The chicken molecule has been extensively studied and shown to contain three different polypeptide subunits [2]. One of the subunits, a2(IX), contains a covalently attached glycosaminoglycan side chain [7, 81. A second subunit, al(IX), has been found to be synthesized in two forms. In cartilage, most of the al(IX) chains contain a large globular amino-terminal domain, while in the embryonic chick cornea, this globular domain is replaced by a short, alternative amino acid sequence [6]. The two forms are generated by the alternative use of two transcription start sites and RNA splice patterns. Although most of the al(1X) transcripts in cartilage are generated from an upstream start site, analysis using the polymerase chain reaction (PCR) has demonstrated that a small fraction of transcripts are synthesized from the alternative, downstream start site. Conversely, in embryonic chick cornea, transcripts from both start sites are present, but the transcripts from the 3’ site predominate [6]. We have previously reported the isolation and sequencing of cDNA clones that cover the carboxyl half of human al(1X) chains [9], and we have mapped the al(1X) gene to the long arm of the human chromosome 6 [9]. We have also found that the human a1(IX) collagen gene gives rise to two transcripts with different sequences in their 5’ regions through the alternative use of two transcription start sites and RNA splice patterns [lo]. However, the presence of these alternative transcripts has been demonstrated only by PCR using primers based on limited sequences of exons within the 5‘ region of Correspondence to B. R. Olsen, Department of Anatomy and Cellular Biology, Harvard Medical School, Boston, MA 021 15, USA Abbreviation. PCR, polymerase chain reaction. Note. The novel nucleotide sequence data published here have been deposited with the EMBL sequence data bank and are available under the accession numbers X54412 and X54413. The novel amino acid sequence data have been deposited with the EMBL sequence data bank.
the human al(1X) gene, and the complete primary structures of the alternative forms of al(1X) chains are not available. Given the potential involvement of type IX collagen-proteoglycan in human chondrodystrophies and chondrodysplasias, the determination of the complete primary structures and isolation of DNA probes covering the entire al(IX) mRNA are of considerable importance. We have therefore isolated and sequenced cDNAs encoding the two mutually exclusive forms of the human al(1X) collagen chain, and we report here for the first time the complete amino acid and corresponding nucleotide sequences of the two forms. MATERIALS AND METHODS Isolation of R N A
Human chondrocytes, isolated and cultured as described [9], were generously provided by Dr Mary Goldring. RNA was isolated from the cells by extraction with guanidine isothiocyanate [I 11and poly(A)-rich RNA was separated from rRNA by oligo(dT)-cellulose chromatography [12]. Human fetal RNA, obtained from a 14-week fetus following therapeutic termination and with the approval of the local ethical committee [lo], was generously provided by Dr Adriano Henney.
Synthesis of cDNA The synthesis and cloning of the human aI(1X) collagen cDNAs KThlOl, KTh123 and KThl5O have been described elsewhere [9]. These cDNAs cover most of the carboxy-terminal half of the human al(1X) collagen chain, and they were isolated from a cDNA library made with human chondrocyte poly(A)-rich RNA as template. For preparation of the cDNA library, cDNA was synthesized according to Gubler and Hoffman [13] using oligo(dT) as primer and ligated to the arms of Lgtl 1 using synthetic EcoRI linkers [9] (Fig. 1). Poly(A)-rich RNA isolated from cultured human chondrocytes was also used as template for primer extension
704 No1
NC3
{
NC2
exons
NCl
COL~H C O L 2 H W
Gene
/ -
6
I
E
X
-
E
E
I KThlOl
1 KThl50 X E
E E
I
YMh754
I
1
.~. . .. .... .-...... ..... . . . . . . . . 143bp ...-/
The complete primary structure of two distinct forms of human al(1X) collagen chains Yasuteru MURAGAKI, Tomoatsu KIMURA, Yoshifumi NINOMIYA and Bjorn Reino OLSEN Department of Anatomy and Cellular Biology, Harvard Medical School, Boston, USA (Received April 30, 1990) - EJB 90 0492
Type IX collagen molecules contain three genetically distinct subunits. One of the subunits, a2(IX), contains a covalently attached glycosaminoglycan side chain. A second subunit, a1(IX), has been found to be synthesized in two forms. The two forms are generated by the alternative use of two transcription start sites and splice patterns. The two forms have been found in chicken, mouse and human but cDNAs encoding both forms have only been reported for chicken. In the present report we describe the isolation of cDNA clones encoding the complete translated portion of both forms of human a1 (IX) collagen chains. Nucleotide sequence analysis has permitted the determination of the primary structure of both forms. These probes and sequences should prove useful in future studies of chondrodysplasias involving type IX collagen Type IX collagen-proteoglycan (collagen IX) is a multidomain extracellular matrix component [l, 21 which is associated with type-11-containing collagen fibrils in cartilage [3] and non-cartilage tissues such as the vitreous [4] and embryonic cornea [5, 61. The chicken molecule has been extensively studied and shown to contain three different polypeptide subunits [2]. One of the subunits, a2(IX), contains a covalently attached glycosaminoglycan side chain [7, 81. A second subunit, al(IX), has been found to be synthesized in two forms. In cartilage, most of the al(IX) chains contain a large globular amino-terminal domain, while in the embryonic chick cornea, this globular domain is replaced by a short, alternative amino acid sequence [6]. The two forms are generated by the alternative use of two transcription start sites and RNA splice patterns. Although most of the al(1X) transcripts in cartilage are generated from an upstream start site, analysis using the polymerase chain reaction (PCR) has demonstrated that a small fraction of transcripts are synthesized from the alternative, downstream start site. Conversely, in embryonic chick cornea, transcripts from both start sites are present, but the transcripts from the 3’ site predominate [6]. We have previously reported the isolation and sequencing of cDNA clones that cover the carboxyl half of human al(1X) chains [9], and we have mapped the al(1X) gene to the long arm of the human chromosome 6 [9]. We have also found that the human a1(IX) collagen gene gives rise to two transcripts with different sequences in their 5’ regions through the alternative use of two transcription start sites and RNA splice patterns [lo]. However, the presence of these alternative transcripts has been demonstrated only by PCR using primers based on limited sequences of exons within the 5‘ region of Correspondence to B. R. Olsen, Department of Anatomy and Cellular Biology, Harvard Medical School, Boston, MA 021 15, USA Abbreviation. PCR, polymerase chain reaction. Note. The novel nucleotide sequence data published here have been deposited with the EMBL sequence data bank and are available under the accession numbers X54412 and X54413. The novel amino acid sequence data have been deposited with the EMBL sequence data bank.
the human al(1X) gene, and the complete primary structures of the alternative forms of al(1X) chains are not available. Given the potential involvement of type IX collagen-proteoglycan in human chondrodystrophies and chondrodysplasias, the determination of the complete primary structures and isolation of DNA probes covering the entire al(IX) mRNA are of considerable importance. We have therefore isolated and sequenced cDNAs encoding the two mutually exclusive forms of the human al(1X) collagen chain, and we report here for the first time the complete amino acid and corresponding nucleotide sequences of the two forms. MATERIALS AND METHODS Isolation of R N A
Human chondrocytes, isolated and cultured as described [9], were generously provided by Dr Mary Goldring. RNA was isolated from the cells by extraction with guanidine isothiocyanate [I 11and poly(A)-rich RNA was separated from rRNA by oligo(dT)-cellulose chromatography [12]. Human fetal RNA, obtained from a 14-week fetus following therapeutic termination and with the approval of the local ethical committee [lo], was generously provided by Dr Adriano Henney.
Synthesis of cDNA The synthesis and cloning of the human aI(1X) collagen cDNAs KThlOl, KTh123 and KThl5O have been described elsewhere [9]. These cDNAs cover most of the carboxy-terminal half of the human al(1X) collagen chain, and they were isolated from a cDNA library made with human chondrocyte poly(A)-rich RNA as template. For preparation of the cDNA library, cDNA was synthesized according to Gubler and Hoffman [13] using oligo(dT) as primer and ligated to the arms of Lgtl 1 using synthetic EcoRI linkers [9] (Fig. 1). Poly(A)-rich RNA isolated from cultured human chondrocytes was also used as template for primer extension
704 No1
NC3
{
NC2
exons
NCl
COL~H C O L 2 H W
Gene
/ -
6
I
E
X
-
E
E
I KThlOl
1 KThl50 X E
E E
I
YMh754
I
1
.~. . .. .... .-...... ..... . . . . . . . . 143bp ...-/
