US 6,265,157 B1Grant
Compositions and methods for detecting altered COL1A1 gene sequences
Issue Date:2001-07-24
•41 Claims
•307 Drawing Sheets
Abstract
Compositions and methods useful for determining whether a subject has an alteration in a gene encoding a protein chain of Type I or Type IX collagen are described. Novel intronic sequences of five human genes, COL1A1, COL1A2, COL9A1, COL9A2, and COL9A3 are described. Methods of determining the existence in a subject of a pathological condition associated with an altered gene encoding a Type I or Type IX collagen protein chain are provided, wherein such pathological conditions include diseases and disorders which are known to be associated with an altered gene encoding a Type I or Type IX collagen protein chain. Primers, probes, and methods of detecting a genetic predisposition of a subject for a pathological condition associated with an altered gene encoding a Type I or Type IX collagen protein chain are provided. Diseases and disorders for which the methods and compositions of the invention are useful for diagnostic or prognostic purposes include, but are not limited to osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease.
Metadata
Assignees
- Allegheny University of the Health Sciences
- Thomas Jefferson University
- University of Oulu
Inventors
- Darwin J. Prockop
- Loretta D. Spotila
- Constantinos D. Deltas
- Larisa Sereda
- Andrea Westerhausen Larson
- Michael Pack
- Alain Colige
- James Early
- Jarmo Körkkö
- Leena Ala-Kokko
- Susanna Annunen
- Tero Pihlajamaa
- Mirko Vuoristo
- Petteri Paassilta
Application Information
Application Number:US 08/943,731
Filing Date:1997-10-03
Priority Date:1991-12-03
Art Unit:7
Classifications
IPC:
C12Q 168C12P 1934C07H 2104
Field of Search:
435 6435 912536 231536 235536 243536 2431536 2433
Patent Drawings (307 sheets)
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] The present application is a continuation-in-part of U.S. application Ser. No. 08/212,322, filed Mar. 13, 1994 (now abandoned), which is a file wrapper continuation of U.S. application Ser. No. 07/803,628, filed Dec. 3, 1991 (now abandoned).
FIELD OF THE INVENTION
[0004] The field of the invention is detecting altered collagen gene sequences.
BACKGROUND OF THE INVENTION
[0005] The collagen genes are an important family of genes, the products of which provide the extracellular framework for virtually all multicellular organisms (Bornstein et al., 1980 Ann. Rev. Biochem. 49:957-1003). More than nineteen distinct types of collagen have been described (Ramirez et al., 1985, Ann. New York Acad. Sci. 460:117-129; Vuorio et al., 1990, Annu. Rev. Biochem. 59:837-872; Chu et al., 1993, In: Connective Tissue and Its Heritable Disorders, Royce et al., eds., Wiley-Liss, New York, pp.149-165; Prockop et al., 1995, Annu. Rev. Biochem. 64:403-434). The biosynthesis of collagen has been described (Prockop et al., 1979, N. Eng. J. Med. 301:13-23).
[0006] Large collagen structures form by nucleated growth of collagen chains into triple helical collagen subunits. Collagen fibrils form by nucleated growth of collagen subunits, a fibril comprising a quarter-staggered array of subunits (Gross et al., 1958, Annu. Rev. Cell Biol. 2:421-457; Wood et al., 1960, Biochem. J. 75:588:598; Prockop et al., 1984, N. Eng. J. Med. 311:376-386; Kadler et al., 1987, J. Biol. Chem. 262:15696-15701; Na et al., 1989, Biochem. 28: 7153-7161; Kadler et al., 1990, Biochem. J. 268:339-343; Prockop et al., 1989, Biophysics (Eng. Transl. Biofizika) 3:81-89). During nucleated growth, the collagen protein chains fold into the triple helical conformation that is a unique and characteristic feature of all collagens (Engel, 1987, Adv. Meat. Res. 4:145-158; Engel et al., 1991, Annu. Rev. Biophys. Biophys. Chem. 20:137-152; Piez, 1984, In: Extracellular Matrix Biochemistry, Piez et al., eds., Elsevier Science Pub. Co. Inc., New York, pp. 1-40).
[0007] Each of the three αchains in a collagen subunit comprises a repeating tripeptide sequence having the general amino acid sequence Gly-X-Y. The presence of glycine, the smallest amino acid, in every third position is critical, since the amino acid in this position fits into a restricted space in which the three chains come together in the center of the triple helix. The X-and Y-amino acid residues are frequently proline and 4-hydroxyproline, respectively. Because the highly flexible glycine bonds flank the relatively inflexible peptide bonds of proline and 4-hydroxyproline (Hyp), individual α chains do not independently fold into any defined three-dimensional structure. Instead, the chains fold into a defined structure only by forming hydrogen bonds and water bridges that link the Gly-X-Y sequences in one α chain to equivalent Gly-X-Y sequences in the two other α chains.
[0008] It is essential to proper collagen molecule conformation that the three a chains are in register, in the sense that the Gly-X-Y tripeptide units in one chain are hydrogen-bonded to the corresponding tripeptide units in the other two α chains. Otherwise, the chains have exposed ends or internal loops of non-triple-helical tripeptide units. Substitution of one or more amino acids of the Gly-X-Y tripeptide sequence with other amino acids, particularly substitution of Gly with an amino acid having a relatively bulky side chain, can produce a structurally abnormal but partially functional collagen subunit. A large number of mutations comprising said substitutions have been described (e.g. Kuivaniemi et al., 1991, FASEB J. 5:2052-2060). Numerous diseases and disorders are associated with mutations in one or more of the Type I or Type IX collagen genes including, but not limited to, osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease.
[0009] Type 1 Collagen
[0010] Type I collagen accounts for about 80 to 90% of the protein found in bone. It is also found in large amounts in tissues such as skin, ligaments, and tendons. In many tissues, the Type I collagen fibrils are associated with other types of collagen and with other components of the extracellular matrix.
[0011] Type I collagen is synthesized as a precursor denoted Type I procollagen, which comprises two proα1(I) chains and one proα2(I) chain. Each proα chain comprises three separate domains, namely an N-propeptide domain, a central domain, and a C-propeptide domain.
[0012] The N-propeptide domain located at the amino-terminal end of each proα chain comprises a globular subdomain, a short triple-helical subdomain, and another short subdomain that forms part of the cleavage site at which the N-propeptide is separated from the mature collagen molecule.
[0013] The central domain of each proα chain is denoted the α-chain domain, which comprises about several hundred amino acid residues and, with the exception of a short sequence at the end of the domain, every third amino acid is glycine. The α-chain largely comprises the Gly-X-Y tripeptide repeating unit.
[0014] The globular C-propeptide domain located at the carboxyl-terminal end of each proα chain is responsible for association of the proα chains during biosynthesis of collagen. Hydrophobic and electrostatic interactions among the C-propeptide domains of the three proα chains direct inclusion of two proα1(I) chains and one proα2(I) chain into the procollagen molecule. Formation of interchain disulfide bonds among the proα subunits further stabilizes the structure of the procollagen molecule, provides the correct registration of the Gly-X-Y tripeptide units of the three chains, and forms a triple helical nucleus of Gly-X-Y units of the three chains. After formation of the triple helical nucleus, triple helical association of the Gly-X-Y units of the three chains proceeds in a zipper-like fashion from the carboxyl-toward the amino-terminal portions of the three chains.
[0015] Biosynthesis of the procollagen molecule involves a large number of post-translational modifications, requiring at least eight procollagen-specific enzymes and several non-specific enzymes. Over a hundred amino acids in each a chain are modified post-translationally. After procollagen is assembled, it is secreted from cells. Extracellularly, the N-propeptide is cleaved from the procollagen molecule by one enzyme and the C-propeptide is cleaved from the procollagen molecule by a second enzyme, yielding an individual mature collagen subunit. The solubility of the collagen subunit is about two thousand times lower than the solubility of the corresponding procollagen subunit. Low collagen solubility drives spontaneous polymerization of collagen subunits into collagen fibrils. Indeed, in vitro assembly of collagen subunits formed by enzymatic cleavage of procollagen subunits has been demonstrated (Prockop et al., 1989, In: Cytoskeletal and Extracellular Proteins, Aebi et al., eds., Springer Series in Biophysics, Vol. 3, pp. 81-89; Kadler et al., 1990, Biochem J. 268:339-343).
[0016] Human proα1(I) is encoded by the COL1A1 gene, which is located on chromosome 17q21.3-q22, and human proα2(I) is encoded by the COL1A2 gene, which is located on chromosome 7q21.3-q22. Oligonucleotide primers useful for amplifying and sequencing cDNA encoding the human proα1(I) chain of Type I procollagen have been described (Labhard et al., 1990, Matrix 10: 124-130).
[0017] The complete cDNA sequence corresponding to the COL1A1 gene has been reported (Chu et al., 1984, Nature 310:337-340; Tromp et al., 1988, Biochem. J. 253:919-922; Bernard et al., 1983, Biochem. 22:5213-5223). Furthermore, the nucleotide sequence of approximately 400 base pairs of the 5′-untranslated region, introns 1-26, and twenty-six nucleotides at the 5′-end of intron 27 of COL1A2 have been reported (Chu et al., 1985, J. Biol. Chem. 260:2315-2320; D'Alessio et al., 1988, Gene 67:105-113; Barsh et al., 1985 Proc. Natl. Acad. Sci. USA 82:2870-2874).
[0018] The complete cDNA sequence corresponding to the COL1A2 gene has been reported (Bernard et al., 1983, Biochem. 22:1139-1145; de Wet et al., 1987, J. Biol. Chem. 262:16032-16036; Kuivaniemi et al., 1988, Biochem. J. 252:633-640). Furthermore, the nucleotide sequence of certain non-coding regions of the COL1A2 gene have been reported, including the following sequences:
[0019] (i) 75 nucleotides located within intron 1
[0020] (ii) 318 nucleotides at the 3′-end of intron 5
[0021] (iii) 298 nucleotides at the 5′-end of intron 6
[0022] (iv) 30 nucleotides at the 3′-end of intron 26
[0023] (v) intron 27
[0024] (vi) intron 28
[0025] (vii) 25 nucleotides at the 5′-end of intron 29
[0026] (viii) intron 33
[0027] (Myers et al., 1983, J. Biol. Chem. 258:10128-10135; Myers et al., 1984, J. Biol. Chem. 259:12941-12944; Dickson et al., 1984, Proc. Natl. Acad. Sci. USA 81:4524-4528; Tromp et al., 1988, Proc. Natl. Acad. Sci. USA 85:5254-5258; Sherwood et al., 1990, Gene 89:238-244; Vasan et al., 1991, Am. J. Hum. Genet. 48:305-317; Ganguly et al., 1991, J. Biol. Chem. 266:12035-12040).
[0028] Alterations in the coding region of either the COL1A1 gene or the COL2A1 gene and gene alterations that decrease expression of either proα1(I) or proα2(I) have been associated with osteogenesis imperfecta, a genetic disease of children which is characterized by bone brittleness. Many, but not all, children afflicted with osteogenesis imperfecta also exhibit blueness of the sclerae of the eyes, poor dentition, and thin skin. These symptoms are thought to be associated with a decrease in the amount of Type I collagen in the corresponding tissue or with formation of abnormal Type I collagen fibrils therein. Bone brittleness associated with osteogenesis imperfecta is usually apparent early in childhood because the patients develop numerous fractures resulting from relatively minor trauma. It is thought that bone brittleness associated with decreased or abnormal Type I collagen expression can be confused with the symptoms of battered child syndrome. Many patients afflicted with mild osteogenesis imperfecta become fracture-free after the growth spurt associated with puberty, but develop a marked susceptibility to bone fracture later in life.
[0029] Alterations in Type I procollagen genes have been found in patients afflicted with some forms of Ehlers-Danlos syndrome (EDS; Weil et al., 1989, EMBO J. 8:1705; Weil et al., 1988, J. Biol. Chem. 263:8561; Weil et al., 1989, J. Biol. Chem. 264:16804; Vasan et al., 1991, Am. J. Hum. Genet. 48:305; Weil et al., 1990, J Biol. Chem. 265:16007). Some patients afflicted with osteoporosis have alterations in one or more of their Type I procollagen genes (Constantinou et al., 1990, Cytogenet. Cell Genet. 51:979; Nicholls et al., 1984, J. Med. Genet. 21:257-262).
[0030] Fibroblasts obtained from a patient afflicted with osteopenia and ankylosing spondylitis synthesized Type I procollagen having decreased thermal stability, an observation which suggests that an altered procollagen protein was involved in the patient's symptoms (Constantinou et al., 1990, Cytogenet. Cell Genet. 51:979). In another case, a structural defect in the proα2(I) chain was found in a family afflicted with osteoporosis and idiopathic scoliosis (Shapiro et al., 1989, Connect Tissue Res. 21:117-123). In a third case, a single base mutation that converted the codon encoding glycine-661 of proα2(I) to a codon encoding serine was reported in a woman afflicted with postmenopausal osteoporosis (Spotila et al., 1990, Am. J. Hum. Genet. 47: A237). In yet another case, a mutation that converted the codon encoding glycine-19 of proα1(I) to a codon encoding cysteine was reported in a patient afflicted with osteoporosis and joint hypermobility (Nicholls et al., 1984, J. Med. Genet. 21:257-262). Furthermore, an eleven-base-pair deletion was detected in the gene encoding proα2(I) in another patient afflicted with osteoporosis and joint hypermobility (Nicholls et al., 1984, J. Med. Genet. 21:257-262). Functional and structural abnormalities of Type I procollagen are also known to result in a number of clinically distinct inherited disorders which affect the strength of bone, ligaments, tendons, and other connective tissues (Prockop, 1990 Arth. Rheumat. 31:1-8). The significance of mutations affecting Type I procollagen structure or function no doubt remains unrecognized in numerous diseases and disorders affecting tissues which comprise Type I collagen.
[0031] Type IX Collagen
[0032] Type IX collagen is a component of hyaline cartilage and the vitreous body of the eye. The Type IX collagen molecule is a heterotrimer comprising three distinct gene products, α1(IX), α2(IX) and α3(IX), which are encoded by the COL9A1 gene, the COL9A2 gene, and the COL9A3 gene, respectively (van der Rest et al., 1987, In Structure and Function of Collagen Types, Mayne et al., eds., Academic Press, Orlando, pp. 195-221; Shaw et al., 1991, Trends Biochem. Sci. 16:191-194). The COL9A1 gene is located on chromosome 6q12-q14, and the COL9A2 gene is located on chromosome 1p32. The chromosomal location of the COL9A3 gene is located on chromosome 20q13.3 (Brewton et al., 1995, Genomics 30:329-336).
[0033] Each α chain comprises three collagenous domains, designated COL1, COL2, and COL3, numbered in the direction from the carboxyl- to the amino-terminus of the chain. The three collagenous domains are flanked by four small non-collagenous domains, designated NC1, NC2, NC3, and NC4 (van der Rest et al., 1988, J. Biol. Chem. 263:1615-1618; Vasios et al., 1988, J. Biol. Chem. 263:2324-2329; Vaughan et al., 1988, J. Cell. Biol. 106:991-997; Ninomiya et al., 1990, In Extracellular Matrix Genes, Sandell et al., eds., Academic press, San Diego, pp. 79-114; Brewton et al., 1995, Genomics 30:329-336).
[0034] The 339-amino-acid COL2 domain and the 137-amino-acid COL3 collagenous domain are identical in length in each the three α chains. The 115-amino-acid COL1 collagenous domains of α1(IX) and α2(IX) are nearly identical in length to the COL1 domain of α3(IX), which is 112 amino acids in length. As a consequence of the similar length of the collagenous regions of the three α chains, the chains are able to fold into an triple helix like that of the (proα1(I))2proα2(I) triple helix of Type I collagen, as described herein, wherein the Gly-X-Y tripeptide sequences of the three α chains are in register.
[0035] The non-collagenous domains vary in size among the three α chains of Type IX collagen. The NC3 domain consists of twelve amino acids in the α1(IX) chain, seventeen amino acids in the α2(IX) chain, and fifteen amino acids in the α3(IX) chain. The difference in size among the non-collagenous domains are thought to impart flexibility to the Type IX collagen molecule.
[0036] Type IX collagen is attached to the surface of Type II collagen fibers by lysine-derived covalent cross-links between the COL2 domain of α3(IX) and the C-telopeptide of Type II collagen and between the N-terminal end of the COL2 domains of all three α chains and the N-telopeptide of Type II collagen (Eyre et al., 1987, FEBS Lett. 220:337-341; van der Rest et al., 1988, J. Biol. Chem. 263:1615-1618; Wu et al., 1992, J. Biol. Chem. 267:23007-23014; Diab et al., 1996, Biochem. J. 314:327-332). Type IX collagen is thus a fibril-associated collagen having interrupted triple helices, and, as such, belongs to the FACIT subgroup of collagens (Gordon et al., 1990, Curr. Op. Cell Biol. 2:833-838).
[0037] When a triple-helical domain of Type IX collagen molecule is anchored to a Type II collagen fibril, the NC3 domain functions as a hinge, allowing the COL3 and NC4 domains to project away from the surface of the fibril. Thus, the COL3 and NC4 domains of Type IX collagen are capable of mediating interactions between Type II collagen fibrils in cartilage and non-collagenous proteins (van der Rest et al., 1988, J. Biol. Chem. 263:1615-1618; Vasios et al., 1988, J. Biol. Chem. 263:2324-2329; Vaughan et al., 1988, J. Cell. Biol. 106:991-997). The NC4 domain in the α1(IX) chain is unique in the sense that it occurs in two variant forms. In cartilaginous tissue, the NC4 domain of the α1(IX) chain has a longer sequence; in ocular tissue, the NC4 domain has a shorter sequence.
[0038] Type IX collagen is a proteoglycan. The NC3 domain of the α2(IX) chain comprises an attachment site for a glycosaminoglycan side chain (Bruckner et al., 1985, Proc. Natl. Acad. Sci. USA 82:2608-2612). Results from a recent study indicate that the NC1 domain of the three α chains of Type IX collagen encode all of the information necessary for glycosaminoglycan side chain selection and assembly (Mechling et al., 1996, J. Biol. Chem. 271:13781-13785).
[0039] Complete cDNA sequences of the chicken, human, and murine COL9A1 genes have been reported (Ninomiya et al., 1984, Proc. Natl. Acad. Sci. USA 81:3014-3018; Vasios et al., 1988, J. Biol. Chem. 263:2324-2329; Ninomiya et al., 1990, In: Extracellular Matrix Genes, Sandell et al., eds., Academic press, San Diego, pp. 79-114; Muragaki et al., 1990, Eur. J. Biochem. 192:703-708; Rokos et al., 1994, Matrix Biol. 14:1-8). Portions of the genomic structure of the chicken, human, murine, and rat COL9A1 genes have been reported (Lozano et al., 1985, Proc. Natl. Acad. Sci. USA 82:4050-4054; Ninomiya et al., 1990, In: Extracellular Matrix Genes, Sandell et al., eds., Academic press, San Diego, pp. 79-114; Muragaki et al., 1990, Proc. Natl. Acad. Sci. USA 87:2400-2404; Ting et al., 1993, J. Bone Min. Res. 8:1377-1387).
[0040] Complete cDNA sequences of the chicken, human, and murine COL9A2 genes have been reported (Ninomiya et al, 1985, Biochem. 24:4223-4229; Perälä et al., 1993, FEBS Lett. 319:177-180; Perälä et al., 1994, J. Biol. Chem. 269:5064-5071). The complete genomic structure of the chicken and murine COL9A2 genes have been reported (Ninomiya et al., 1990, In Extracellular Matrix Genes, Sandell et al., eds., Academic press, San Diego , pp. 79-114; Perälä et al., 1994, J. Biol. Chem. 269:5064-5071).
[0041] Complete cDNA sequences of the chicken and human COL9A3 genes have been reported (Brewton et al., 1992, Eur. J. Biochem. 205:443-449; Har-El et al., 1992, J. Biol. Chem. 267:10070-10076; Brewton et al., 1995, Genomics 30:329-336). The genomic structure of the COL9A3 gene has not been reported in any species to date.
[0042] Transgenic mice expressing a cDNA construct comprising the coding region of the COL9A1 gene having a large in-frame deletion in the COL2-domain-encoding region thereof develop abnormalities in cartilage collagen fiber structure, and exhibit a phenotype similar to human osteoarthritis and mild chondrodysplasia (Nakata et al., 1993, Proc. Natl. Acad. Sci. USA 90:2870-2874). Degenerative joint disease was also exhibited by transgenic mice which were homozygous for an inactivated COL9A1 gene (Fassler et al., 1994, Proc. Natl. Acad. Sci. USA 91:5070-5074), by transgenic mice which overexpressed the isolated NC4 domain of the α1(IX) chain (Haimes et al., 1996, Inflam. Res. 44(Suppl.2):S127-S128), and by transgenic mice which expressed a truncated COL9A2 gene having an in-frame deletion of a region which encoded 38 amino acids in the COL2 domain of α2(IX) (Perälä et al., 1994, J. Biol. Chem. 269:5064-5071). These findings indicate that Type IX collagen is not essential for cartilage development, but it is required for maintaining the integrity of cartilage structures.
[0043] Until the present invention, it has been possible to identify a mutation associated with a pathological condition in a human COL1 or COL9 gene only if the mutation was located within the coding sequence of one of the COL1A1, COL1A2, COL9A1, COL9A2, and COL9A3 genes, within one of introns 1-26 of the COL1A1 gene, within the 26 nucleotides located at the 5′-end of intron 27 of the COL1A1 gene, or within the approximately 350 nucleotides adjacent exon 1 of the COL1A1 gene, in the 5′-untranslated region thereof. Hence, a serious unmet need exists for methods and compositions which are useful for identifying mutations which are located in non-coding regions of the genes encoding the chains of Type I and Type IX collagen and which are associated with a pathological condition.
SUMMARY OF THE INVENTION
[0044] The invention relates to a method of detecting an alteration in a collagen gene of a human subject, wherein the alteration is associated with a pathological condition in the subject, the method comprising
[0045] (i) obtaining from the subject a sample nucleic acid comprising at least a portion of the gene, wherein the portion comprises at least one intronic nucleotide, a first site, and a second site;
[0046] (ii) determining the nucleotide sequence of the portion of the gene; and
[0047] (iii) comparing the nucleotide sequence of the portion of the gene with a consensus nucleotide sequence of the gene. A difference between the nucleotide sequence and the consensus nucleotide sequence is indicative of the presence in the subject of the alteration in the gene. The portion of the gene is selected from the group consisting of the segment of the COL1A1 gene extending in the 5′- to 3′-direction from and including the 78 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of the COL1A1 gene, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 3′-end thereof through intron 4, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 2600 nucleotides at the 3′-end of intron 26 through the 340 nucleotides at the 5′-end of intron 26, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 775 nucleotides at the 3′-end of intron 29 through intron 32, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including intron 34 through the 5′-end of the COL1A2 gene, the COL9A1 gene, the COL9A2 gene, and the COL9A3 gene. The consensus nucleotide sequence of the COL1A1 gene is SEQ ID NO: 1; the consensus nucleotide sequence of the COL1A2 gene is SEQ ID NO: 2; that of the COL9A1 gene is SEQ ID NO: 3; that of the COL9A2 gene is SEQ ID NO: 4; and that of the COL9A3 gene comprises SEQ ID NO: 5 and SEQ ID NO: 640.
[0048] In one embodiment, the method of the invention further comprises the step of contacting the portion of the gene with a first intronic primer prior to determining the nucleotide sequence of the portion, the first intronic primer being either substantially complementary to or substantially homologous with the first site. Preferably, the first intronic primer has a sequence selected from the group consisting of SEQ ID NO: 273 through SEQ ID NO: 336, SEQ ID NO: 339 through SEQ ID NO: 379, SEQ ID NO: 382 through SEQ ID NO: 391, SEQ ID NO: 394 through SEQ ID NO: 477, SEQ ID NO: 480 through SEQ ID NO: 499, SEQ ID NO: 502 through SEQ ID NO: 563, and SEQ ID NO: 568 through SEQ ID NO: 639.
[0049] Another embodiment of the method of the invention further comprises the steps of contacting the portion of the gene with a first intronic primer homologous to the first site, contacting the portion of the gene with a second primer complementary to the second site, and amplifying the portion of the gene to obtain an amplified polynucleotide prior to determining the nucleotide sequence of the portion.
[0050] In another embodiment of the method of the invention, the method further comprises a CSGE step performed after the step of amplifying and prior to the step of determining the nucleotide sequence of the portion of the gene, the CSGE step comprising the steps of denaturing the amplified polynucleotide, annealing the amplified polynucleotide, and determining whether the amplified polynucleotide forms a heteroduplex, and wherein the step of determining the nucleotide sequence of the portion comprises determining the nucleotide sequence of the amplified polynucleotide.
[0051] In yet another embodiment of the method of the invention, the first intronic primer is either substantially complementary to or substantially homologous with a part of a non-coding region of the portion, wherein the non-coding region of the portion is selected from the group consisting of SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 24 through SEQ ID NO: 58, SEQ ID NO: 60 through SEQ ID NO: 62, SEQ ID NO: 65 through SEQ ID NO: 83, and SEQ ID NO: 88 through SEQ ID NO: 215.
[0052] In a preferred embodiment of the method of the invention, the first intronic primer has a sequence selected from the group consisting of SEQ ID NO: 273 through SEQ ID NO: 336, SEQ ID NO: 339 through SEQ ID NO: 379, SEQ ID NO: 382 through SEQ ID NO: 391, SEQ ID NO: 394 through SEQ ID NO: 447, SEQ ID NO: 480 through SEQ ID NO: 499, SEQ ID NO: 502 through SEQ ID NO: 563, and SEQ ID NO: 568 through SEQ ID NO: 639. Also preferably, the first primer and the second primer are selected from the pairs of primers listed in FIGS. 21 through 25.
[0053] In still another embodiment of the method of the invention, the steps of contacting the portion of the gene with a first intronic primer homologous to the first site, contacting the portion of the gene with a second primer complementary to the second site, and amplifying the portion of the gene to obtain an amplified polynucleotide comprise the steps of contacting the portion of the gene with a plurality of pairs of intronic primers and amplifying the portion of the gene to obtain a plurality of amplified polynucleotides. In a variation of this embodiment, the method further comprises a CSGE step performed after the step of amplifying the portion and prior to the step of determining the nucleotide sequence of the portion, the CSGE step comprising the steps of denaturing the amplified polynucleotides, annealing the amplified polynucleotides, and determining whether any of the amplified polynucleotides forms a heteroduplex, wherein the step of determining the nucleotide sequence of the portion comprises determining the nucleotide sequence of any of the amplified polynucleotides which forms a heteroduplex. Preferably, each of the intronic primers is either substantially complementary to or substantially homologous with a part of a non-coding region of the portion. Also preferably, the length of every of the amplified polynucleotides is between about two hundred and about five hundred nucleotides. Also preferably, the length of the polynucleotide amplified using any one pair of the plurality of pairs of primers is different from the length of the polynucleotides amplified using every other pair of the plurality of pairs of primers.
[0054] In one embodiment of the method of the invention, a plurality of pairs of primers comprising pairs of primers sufficient to amplify substantially all exons and exon flanking regions of the gene is used. In this method, it is preferred that the length of every of the amplified polynucleotides is between about two hundred and about five hundred nucleotides, and wherein the length of the polynucleotide amplified using every single pair of the plurality of pairs of primers is different from the length of the polynucleotides amplified using every other pair of the plurality of pairs of primers.
[0055] In a variation of this embodiment of the method of the invention, the method further comprises a CSGE step performed after the step of amplifying the portion and prior to the step of determining the nucleotide sequence of the portion, the CSGE step comprising the steps of denaturing the amplified polynucleotides, annealing the amplified polynucleotides, and determining whether any of the amplified polynucleotides forms a heteroduplex, wherein the step of determining the nucleotide sequence of the portion comprises determining the nucleotide sequence of any of the amplified polynucleotides which forms a heteroduplex.
[0056] In a first embodiment of the method of the invention, the portion of the gene is a segment of the COL1A1 gene not including any non-coding sequence disclosed in the scientific literature prior to the invention and each of the pairs of primers is selected from the pairs of primers listed in FIG. 21. Preferably, the pathological condition is selected from the group consisting of osteoporosis, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, and low bone density.
[0057] In a second embodiment of the method of the invention, the portion of the gene is a segment of the COL1A2 gene not including any non-coding sequence disclosed in the scientific literature prior to the invention and each of the pairs of primers is selected from the pairs of primers listed in FIG. 22. Preferably, the pathological condition is selected from the group consisting of osteoporosis, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, and low bone density.
[0058] In a third embodiment of the method of the invention, the portion of the gene is the COL9A1 gene and each of the pairs of primers is selected from the pairs of primers listed in FIG. 23. Preferably, the pathological condition is selected from the group consisting of osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, scoliosis, and degenerative joint disease.
[0059] In a fourth embodiment of the method of the invention, the portion of the gene is the COL9A2 gene and each of the pairs of primers is selected from the pairs of primers listed in FIG. 24. Preferably, the pathological condition is selected from the group consisting of osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, scoliosis, and degenerative joint disease.
[0060] In a fifth embodiment of the method of the invention, the portion of the gene is the COL9A3 gene, and each of the pairs of primers is selected from the pairs of primers listed in FIG. 25. Preferably, the pathological condition is selected from the group consisting of osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, scoliosis, and degenerative joint disease.
[0061] The invention also relates to an isolated nucleic acid comprising at least about fifteen consecutive nucleotides and having a nucleotide sequence which is either substantially complementary to or substantially homologous with a portion of a human collagen gene, wherein the portion of the gene includes at least one nucleotide located in a non-coding region.
[0062] The portion of the gene is preferably selected from the group consisting of the segment of the COL1A1 gene extending in the 5′- to 3′-direction from and including the 78 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of the COL1A1 gene, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 3′-end thereof through intron 4, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 2600 nucleotides at the 3′-end of intron 26 through the 340 nucleotides at the 5′-end of intron 26, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 775 nucleotides at the 3′-end of intron 29 through intron 32, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including intron 34 through the 5′-end of the COL1A2 gene, the COL9A1 gene, the COL9A2 gene, and the COL9A3 gene.
[0063] Preferably, the isolated nucleic acid is completely complementary to or is completely homologous with the portion of the collagen gene. Also preferably, the portion includes at least three nucleotides located in a non-coding region. Preferably, the portion of the collagen gene comprises only nucleotides located in the non-coding region, wherein the non-coding region has a nucleotide sequence selected from the group consisting of SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 24 through SEQ ID NO: 58, SEQ ID NO: 60 through SEQ ID NO: 62, SEQ ID NO: 65 through SEQ ID NO: 83, and SEQ ID NO: 88 through SEQ ID NO: 215.
[0064] In another embodiment, the isolated nucleic acid has a nucleotide sequence selected from the group consisting of SEQ ID NO: 273 through SEQ ID NO: 336, SEQ ID NO: 339 through SEQ ID NO: 379, SEQ ID NO: 382 through SEQ ID NO: 391, SEQ ID NO: 394 through SEQ ID NO: 447, SEQ ID NO: 480 through SEQ ID NO: 499, SEQ ID NO: 502 through SEQ ID NO: 563, and SEQ ID NO: 568 through SEQ ID NO: 639.
[0065] The invention also relates to a kit for detecting an alteration in a gene segment of a human collagen gene, the kit comprising a consensus sequence for the gene and a pair of intronic primers, each having a length of at least about fifteen consecutive nucleotides, wherein the gene segment comprises at least one nucleotide located in a non-coding region of the gene.
[0066] In the kit, the gene is selected from the group consisting of the segment of the COL1A1 gene extending in the 5′- to 3′-direction from and including the 78 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of the COL1A1 gene, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 3′-end thereof through intron 4, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 2600 nucleotides at the 3′-end of intron 26 through the 340 nucleotides at the 5′-end of intron 26, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including the 775 nucleotides at the 3′-end of intron 29 through intron 32, the segment of the COL1A2 gene extending in the 5′- to the 3′-direction from and including intron 34 through the 5′-end of the COL1A2 gene, the COL9A1 gene, the COL9A2 gene, and the COL9A3 gene.
[0067] The consensus nucleotide sequence of the COL1A1 gene is SEQ ID NO: 1; the consensus nucleotide sequence of the COL1A2 gene is SEQ ID NO: 2; that of the COL9A1 gene is SEQ ID NO: 3; that of the COL9A2 gene is SEQ ID NO: 4; and that of the COL9A3 gene is listed in SEQ ID NO: 5 and SEQ ID NO: 640.
[0068] In one embodiment of the kit, the length of the gene segment is between about two hundred and about five hundred nucleotides. In another embodiment of the kit of the invention, the kit comprises at least one sequencing primer for determining the nucleotide sequence of at least a portion of the gene segment. In yet another embodiment of the kit of the invention, the pair of primers is selected from the group consisting of the pairs of primers listed in FIGS. 21 through 25. In still another embodiment, the kit comprises a plurality of the pairs of primers. In a first embodiment of the kit of the invention, the gene is a segment of the COL1A1 gene not including any non-coding sequence disclosed in the scientific literature prior to the invention and the first primer and the second primer are selected from the pairs of primers listed in FIG. 21. In a second embodiment of the kit of the invention, the gene is a segment of the COL1A2 gene not including any non-coding sequence disclosed in the scientific literature prior to the invention and the first primer and the second primer are selected from the pairs of primers listed in FIG. 22. In a third embodiment of the kit of the invention, the gene is the COL9A1 gene and the first primer and the second primer are selected from the pairs of primers listed in FIG. 23. In a fourth embodiment of the kit of the invention, the gene is the COL9A2 gene and the first primer and the second primer are selected from the pairs of primers listed in FIG. 24. In a fifth embodiment of the kit of the invention, the gene is the COL9A3 gene and the first primer and the second primer are selected from the pairs of primers listed in FIG. 25.
BRIEF DESCRIPTION OF THE DRAWINGS
[0069] FIG. 1 is a schematic depiction of the 3′-portion of the human COL1A1 gene, comprising the portion extending from a portion of intron 25 through the 3′-end of the gene.
[0070] FIG. 2 is a schematic depiction of the human COL1A2 gene, extending from the 5′-untranslated region to the 3′-untranslated region thereof. Numbers in the larger font refer to intron numbers; numbers in the smaller font refer to the nucleotide length of each of the introns. The scale is indicated in the lower right, the bar representing a polynucleotide having a length of one thousand residues.
[0071] FIG. 3, comprising FIGS. 3A-3G, depicts the nucleotide sequence of the introns of the human COL1A1 gene extending from a portion of intron 25 to the three nucleotides at the 5′-end of exon 52. Capital letters indicate terminal codons of exon sequences or five ambiguous bases in introns. The alternative bases in some cloned PCR products were G/T at position +185 of intron 46, A/C at position +291 of intron 46, T/C at position +112 of intron 49, A/G at position +126 of intron 49, and C/T at position +179 of intron 50, wherein numbering is in the 3′-direction from the 5′-end of the intron on the coding strand of the gene.
[0072] FIG. 4, comprising FIGS. 4A-1 through 4B-54, depicts the human COL1A1 gene. FIGS. 4A-1 through 4A-15 list a consensus nucleotide sequence of the human COL1A1 gene (SEQ ID NO: 1). FIGS. 4B-1 through 4B-54 list the consensus nucleotide sequences of the 5′-untranslated region (FIG. 4B-1; SEQ ID NO: 6), introns 1-50 (FIGS. 4B-2 through 4B-52; SEQ ID NOs: 7-56, respectively), and the 3′-untranslated region (FIGS. 4B-53 and 4B-54; SEQ ID NO: 57) of the COL1A1 gene.
[0073] FIG. 5, comprising FIGS. 5A-1 through 5B-58, depicts the human COL1A2 gene. FIGS. 5A-1 through 5A-31 list a consensus nucleotide sequence of the human COL1A2 gene (SEQ ID NO: 2). FIGS. 5B-1 through 5B-58 list the consensus nucleotide sequences of the 5′-untranslated region (FIG. 5B-1 and; SEQ ID NO: 58), introns 1-51 (FIGS. 5B-3 through 5B-57; SEQ ID NOs: 59-109, respectively), and the 3′-untranslated region (FIGS. 5B-58; SEQ ID NO: 110) of the COL1A2 gene.
[0074] FIG. 6 is a graph depicting a comparison of the lengths of the introns of the human and chicken COL1A2 genes. Intron sizes in the human gene are indicated by open circles, and intron sizes in the chicken gene are indicated by filled triangles.
[0075] FIG. 7 is an image depicting the results of CSGE analysis of amplification products derived from human subjects having single-base alterations in one allele of the individual's COL1A1 gene. Samples were obtained from individuals described in Table 2.
[0076] FIG. 8, comprising FIGS. 8A and 8B, depicts schematic representations of the human COL9A1 and COL9A2 genes and the gene products thereof. In FIG. 8A, a diagram of the genomic organization of the COL9A1 gene is depicted, wherein exon numbers are indicated above the diagram, and relative positions of the sequence on P1 clones P1-A and P1-B are indicated below the diagram. The position of alternate exon 1*, which is expressed in ocular tissue, is indicated. Also in FIG. 8A, a diagram of the genomic organization of the COL9A2 gene is depicted, wherein exon numbers are indicated above the diagram, and relative position of the sequence on a PAC clone, PAC-1, is depicted below the diagram. The scale used in FIG. 8A, in kilobases, is indicated along the bottom of the Figure. In FIG. 8B, the protein chains encoded by the COL9A1 and COL9A2 genes are represented. The amino acid length of each region of the proteins is indicated. The COL1, COL2, and COL3 regions are shaded. The exons which encode each of the regions is indicated, wherein E16, for example, refers to a portion of the protein encoded by exon 16. The α1(IX) chain depicted in FIG. 8B is designated “9α1” and represents the form of the protein expressed in cartilaginous tissue.
[0077] FIG. 9 is a schematic representation of the human COL9A3 gene. The scale used, in kilobases, is indicated along the central portion of the Figure. A restriction map of the COL9A3 gene region is depicted on the top portion of the Figure, and restriction sites for restriction endonucleases are indicated. Below the scale, the genomic organization of the gene is represented, wherein exons are numbered. Below this representation, the protein chain encoded by the gene is depicted. The lines connecting the representation of the gene with the depiction of the protein indicate the exons which encode the various regions of the protein. The numerals in parentheses next to the region names indicate the amino acid length of the region.
[0078] FIG. 10, comprising FIGS. 10A-1 through 10B-41, depicts the human COL9A1 gene. FIGS. 10A-1 through 10A-23 list a consensus nucleotide sequence of the human COL9A1 gene (SEQ ID NO: 3). FIGS. 10B-1 through 10B-41 list the consensus nucleotide sequences of the 5′-untranslated region (FIGS. 10B-1 and 10B-2; SEQ ID NO: 111), introns 1-37 (FIGS. 10B-3 through 10B-40; SEQ ID NOs: 112-148, respectively), and the 3′-untranslated region (FIG. 10B-41; SEQ ID NO: 149) of the COL9A1 gene.
[0079] FIG. 11, comprising FIGS. 11A-1 through 11B-38, depicts the human COL9A2 gene. FIGS. 11A-1 through 11A-19 list a consensus nucleotide sequence of the human COL9A2 gene (SEQ ID NO: 4). FIGS. 11B-1 through 11B-38 list the consensus nucleotide sequences of the 5′-untranslated region (FIG. 11B-1; SEQ ID NO: 150), introns 1-31 (FIGS. 11B-2 through 11B-37; SEQ ID NOs: 151-181, respectively), and the 3′-untranslated region (FIG. 10B-38; SEQ ID NO: 182) of the COL9A2 gene.
[0080] FIG. 12, comprising FIGS. 12A, 12B, and 12C, depicts the human COL9A3 gene. FIGS. 12A-1 through 12A-16 list a portion of the gene, comprising about three kilobases of the promoter through a 5′-portion of intron 26 thereof. FIGS. 12B-1 through 12B-5 list another portion of the gene, comprising a 3′-portion of intron 26 through the 3′-end of exon 32. The sequence depicted in FIG. 12A is SEQ ID NO: 5; the sequence in FIG. 12B is SEQ ID NO: 640. FIGS. 12C-1 through 12C-37 list the consensus nucleotide sequences of the 5′-untranslated region (FIGS. 12C-1, 12C-2, and 12C-3; SEQ ID NO: 183), introns 1-25 and the 5′-portion of intron 26 (FIGS. 12C-4 through 12C-34; SEQ ID NOs: 184-208 and 209, respectively), the 3′-portion of intron 26, and introns 27-31 (FIGS. 12C-35, 12C-36, and 12C-37; SEQ ID NOs: 210 and 211-215) of the COL9A3 gene.
[0081] FIG. 13 depicts the 3′-untranslated region of the COL9A1 gene, as determined by 3′-RACE analysis, as described herein.
[0082] FIG. 14 depicts alternate splicing patterns for exons 1 through 8 of the COL9A1 gene. The gene is depicted as a series of boxes and lines across the center of the Figure. The lines and legends above the gene indicate how the exons of the gene are spliced in, for example, cartilaginous tissue, wherein exons 1 through 8 are spliced to exon 6, which is spliced to exon 7, which is spliced to exon 8, which is spliced to the higher-numbered exons of the gene, which are not shown in the Figure. The lines and legends below the gene indicate how the exons of the gene are spliced in, for example, corneal tissue, wherein alternate exon 1* is spliced to exon 8, which is spliced to the higher-numbered exons of the gene.
[0083] FIG. 15 lists the nucleotide sequences of a region of intron 6 of each of the human (“H”), murine (“M”), and chicken (“C”) COL9A1 genes, as described herein.
[0084] FIG. 16 depicts a comparison of the sizes of the introns in the human COL9A1 gene, depicted by filled circles connected by solid lines, and the sizes of the introns in the murine COL9A1 gene, depicted by open circles connected by broken lines. Intron number is indicated on the horizontal axis, and the size, in nucleotides, of each intron is indicated on the vertical axis. Introns in which Alu sequences have been detected are indicated by “+Alu”.
[0085] FIG. 17 depicts the number of Gly-X-Y amino acid triplet sequences encoded by exons 2, 3, and 4 of the human COL9A3 and COL9A2 genes. Exons are shown to scale. The junction of the NC4 and COL3 domains is indicated.
[0086] FIG. 18, comprising FIGS. 18A and 18B, depicts the results of CSGE analysis of exon 30 of the COL9A3 gene obtained from seven subjects from an MED-affected family. Affected family members are indicated by shaded symbols. Round symbols indicate female subjects and square symbols indicate male subjects. The five subjects shown below the other two subjects are the offspring of the two subjects. The pair of male subjects connected by a ‘V’-shaped line are twins. Corresponding symbols and CSGE results are numbered identically in FIGS. 18A and 18B.
[0087] FIG. 19 depicts the nucleotide sequence of a region of the COL9A3 gene, including a repetitive GC-rich region in which a deletion occurred in individuals afflicted with MED. The normal sequence of the gene is shown on the top line. Two alternate sites where the deletion described herein may have occurred are indicated on the lower two lines.
[0088] FIG. 20, comprising FIGS. 20A and 20B, depicts alternate models whereby non-triple helical domains such as the NC2 region of Type IX collagen may have a role in aligning Gly-X-Y sequences so that they can assume a triple helical conformation. In Panel A, the conformation of a portion of a normal Type IX collagen fibril is depicted. In Panel B, the conformation of a portion of an altered Type IX collagen fibril is depicted, wherein a Gly-X-Y triplet has been altered or deleted in one chain of the fibril.
[0089] FIG. 21, comprising FIGS. 21A-21L, depicts pairs of oligonucleotide primers used for PCR amplification of the promoter region, exons and flanking sequences, and the polyadenylation signals of the COL1A1 gene. Primer sequences are listed with the 5′-end of the sequence at the left. The position (“pos”) of each 5′-primer is indicated as the relative position of the 5′-most nucleotide of the gene sequence to which the 5′-primer is homologous. The position of the 5′-primer used to amplify the promoter is indicated relative to the transcription start site; the position of each 5′-primer used to amplify an exon and flanking regions is indicated relative to the 5′-end of the exon; the position of each 5′-primer used to amplify a polyadenylation signal is indicated relative to the 5′-end of the signal. The position of each 3′-primer is indicated as the relative position of the 3′-most nucleotide of the gene sequence to which the 3′-primer is complementary. The position of the 3′-primer used to amplify the promoter is indicated relative to the transcription start site; the position of each 3′-primer used to amplify an exon and flanking regions is indicated relative to the 3′-end of the exon; the position of each 3′-primer used to amplify a polyadenylation signal is indicated relative to the 3′-end of the signal. All relative positions are indicated such that a negative number represents a position in the direction of the 5′-end of the gene and a positive number represents a position in the direction of the 3′-end of the gene, relative to the corresponding reference position. All references to the gene in this figure legend refer to the coding strand thereof The length of each PCR amplification product made using a primer pair is indicated (“transcript length”). Two pairs of primers, each pair comprising the same 3′-primer, are shown for Exon 2. A SEQ ID NO is listed adjacent to each primer.
[0090] FIG. 22, comprising FIGS. 22A-22K, depicts pairs of oligonucleotide primers used for PCR amplification of the promoter region, exons and flanking sequences, and the polyadenylation signals of the COL1A2 gene. Primer sequences are listed with the 5′-end of the sequence at the left. Positions are indicated as in FIG. 21. The length of each PCR amplification product made using a primer pair is indicated (“transcript length”). The position indicated for the 3′-primer used to amplify the promoter is at nucleotide 28 of exon 1. A SEQ ID NO is listed adjacent to each primer.
[0091] FIG. 23, comprising FIGS. 23A-23H, depicts pairs of oligonucleotide primers used for PCR amplification of the promoter region and the exons and flanking sequences of the COL9A1 gene. Primer sequences are listed with the 5′-end of the sequence at the left. Positions are indicated as in FIG. 21. The length of each PCR amplification product made using a primer pair is indicated (“transcript length”). Primer positions located within an exon are indicated by “(Exon #)”, wherein “#” is the exon number. A SEQ ID NO is listed adjacent to each primer.
[0092] FIG. 24, comprising FIGS. 24A-24F, depicts pairs of oligonucleotide primers used for PCR amplification of the exons and flanking sequences of the COL9A2 gene. Primer sequences are listed with the 5′-end of the sequence at the left. Positions are indicated as in FIG. 21. The length of each PCR amplification product made using a primer pair is indicated (“transcript length”). Primer positions located within an exon are indicated by “(Exon #)”, wherein “#” is the exon number. The position of the 5′-primer used to amplify exon 24 is located in intron 22. A SEQ ID NO is listed adjacent to each primer.
[0093] FIG. 25, comprising FIGS. 25A-25F, depicts pairs of oligonucleotide primers used for PCR amplification of the exons and flanking sequences of the COL9A3 gene. Primer sequences are listed with the 5′-end of the sequence at the left. Positions are indicated as in FIG. 21. The length of each PCR amplification product made using a primer pair is indicated (“transcript length”). A SEQ ID NO is listed adjacent to each primer.
DETAILED DESCRIPTION OF THE INVENTION
[0094] The present invention provides compositions and methods for detecting altered COL1 and COL9 gene sequences in a human subject. Prior art compositions and methods are useful only for detecting mutations in the coding sequences of COL1 or COL9 genes or in certain non-coding sequences of the COL1A1 gene. The compositions and methods of the present invention are superior to prior art compositions and methods in that they are useful for detecting mutations in both the coding sequences and the non-coding sequences of any of the COL1 or COL9 genes. Thus, the compositions and methods can be used to detect collagen gene alterations which affect either the primary sequence of a collagen protein chain, splicing of the mRNA encoding such chains, or regulation of expression of the genes encoding such chains.
[0095] The ability to detect abnormalities in a COL1 or COL9 gene permits one skilled in the art to determine whether a subject is predisposed to certain pathological conditions associated with an altered COL1 or COL9 gene sequence. Similarly, the ability to detect abnormalities in a COL1 or COL9 gene permits one skilled in the art to determine whether a subject presenting symptoms potentially attributable to a pathological condition associated with an altered COL1 or COL9 gene sequence is afflicted with that pathological condition. Furthermore, identification of an abnormality in a COL1 or COL9 gene of a subject is useful for designing a therapeutic nucleotide or gene therapy agent which can be administered to the subject to correct or alleviate the abnormality.
[0096] By way of example, one skilled in the art would, upon reading the present disclosure, appreciate that the compositions and methods described herein are useful for identifying human subjects who are afflicted with, or who are predisposed to be afflicted with, any of a number of pathological conditions associated with an altered COL1 or COL9 gene sequence including, but not limited to, osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease. One skilled in the art would further appreciate that the compositions and methods described herein are useful for identifying an altered COL1 or COL9 gene sequence associated with a pathological condition, even if the association between the altered sequence and the pathological condition is not presently recognized.
[0097] The present invention also provides conformation-sensitive gel electrophoresis (CSGE) compositions and methods whereby alterations in a COL1 or COL9 gene may be easily identified.
[0098] The compositions of the present invention may be conveniently packaged in the form of a kit comprising numerous intronic primers, whereby the entire sequence of one or more COL1 or COL9 genes may be examined for the presence of genetic alterations. The kit of the invention provides the practitioner with the numerous reagents which are useful to practice the methods of the invention including, but not limited to, intronic primers, one or more consensus collagen gene sequences, oligonucleotides having sequences derived from portions of a consensus gene sequence, CSGE compositions, instructions, and the like.
[0099] It has now been determined that mutations in Type I procollagen genes increase the likelihood of developing osteoporosis. Predisposition for developing osteoporosis is therefore heritable. Although mutations in collagen genes have been demonstrated in diseases such as osteogenesis imperfecta, Ehlers-Danlos syndrome, and related disorders, it was heretofore unknown that mutations in Type I procollagen genes are linked to osteoporosis in the absence of any other evidence of a connective tissue disease or syndrome. Mutations in type I procollagen have been discovered to cause osteoporosis in a subset of patients afflicted with the disease.
[0100] When the compositions and methods of the invention are used to identify a mutation in a COL1 or COL9 gene of a first family member by scanning the entire gene of the first family member, it is possible to determine whether a second family member also has the mutation in his or her same COL1 or COL9 gene by using only a single primer pair which amplifies the mutated region of the affected gene. The invention conveniently provides kits comprising compositions which can be used to identify alterations anywhere within a COL1 or COL9 gene of a first family member and compositions which can be used to analyze a particular portion of an affected COL1 or COL9 gene in other family members to determine whether the other family members have the mutated gene.
[0101] If the existence in a subject of a pathological condition associated with an altered COL1 or COL9 gene sequence or a predisposition to develop such a pathological condition is detected early, effective treatment is more likely to be available prior to the onset of severe, less treatable symptoms. Also, members of families identified as being predisposed to such pathological conditions may choose to seek appropriate genetic counseling prior to reproduction.
[0102] PCR/CSGE Methods for Detecting an Altered Type I or Type IX Collagen Gene
[0103] The methods of the invention comprise detecting an alteration in a gene encoding a chain of Type I or Type IX collagen, relative to the consensus nucleotide sequence of the gene. The consensus nucleotide sequence of the human COL1A1 gene is listed in FIGS. 4A-1 through 4A-15 (SEQ ID NO: 1). The consensus nucleotide sequence of the human COL1A2 gene listed in FIGS. 5A-1 through 5A-31 (SEQ ID NO: 2). The consensus nucleotide sequence of the human COL9A1 gene is listed in FIGS. 10A-1 through 10A-23 (SEQ ID NO: 3). The consensus nucleotide sequence of the human COL9A2 gene is listed in FIGS. 11A-1 through 11A-19 (SEQ ID NO: 4). The consensus nucleotide sequence of the human COL9A3 gene is listed in FIGS. 12A-1 through 12B-5 (SEQ ID NOs: 5 and 640, respectively).
[0104] An alteration in a gene encoding a chain of Type I or Type IX collagen, relative to the consensus nucleotide sequence of the gene, may be detected by obtaining from a subject a nucleic acid comprising at least a portion of one allele of the gene, wherein the portion comprises at least one intronic nucleotide, determining the nucleotide sequence of the nucleic acid, and comparing the nucleotide sequence of the nucleic acid with the consensus nucleotide sequence of the gene, whereby a difference between the nucleotide sequence of the nucleic acid and the consensus nucleotide sequence indicates the presence of an alteration of the gene in the allele of the subject. Preferably, the collagen gene is selected from the group consisting of the segment of the COL1A1 gene extending in the 5′- to 3′-direction from and including the 68 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of the COL1A1 gene, the COL1A2 gene, the COL9A1 gene, the COL9A2 gene, and the COL9A3 gene, and excluding published non-coding sequences of the COL1A2 gene. Published non-coding sequences of the COL1A2 gene include the following sequences:
[0105] (i) 75 nucleotides located within intron 1
[0106] (ii) 318 nucleotides at the 3′-end of intron 5
[0107] (iii) 298 nucleotides at the 5′-end of intron 6
[0108] (iv) 30 nucleotides at the 3′-end of intron 26
[0109] (v) intron 27
[0110] (vi) intron 28
[0111] (vii) 25 nucleotides at the 5′-end of intron 29
[0112] (viii) intron 33
[0113] (Myers et al., 1983, J. Biol. Chem. 258:10128-10135; Myers et al., 1984, J. Biol. Chem. 259:12941-12944; Dickson et al., 1984, Proc. Natl. Acad. Sci. USA 81:4524-4528; Tromp et al., 1988, Proc. Natl. Acad. Sci. USA 85:5254-5258; Sherwood et al., 1990, Gene 89:238-244; Vasan et al., 1991, Am. J. Hum. Genet. 48:305-317; Ganguly et al., 1991, J. Biol. Chem. 266:12035-12040). Because the nucleotide sequences of the non-coding regions in these collagen genes are described herein for the first time, it is now possible to construct primers which are useful for replicating the sequence of an exon and intronic sequences flanking one or both ends of the exon. Thus, one skilled in the art, armed with the present disclosure, can detect both alterations affecting the primary sequence of the collagen protein chains encoded by these genes and alterations affecting mRNA splicing, transcriptional regulation, or translational regulation in these genes.
[0114] Using a pair of such primers and well known PCR reagents, a portion of a genomic DNA molecule or a non-spliced mRNA molecule comprising a Type I or Type IX collagen gene may be amplified to provide an amplified DNA sequence. The amplified DNA sequence comprises an exon of the gene, a portion of a first non-coding region of the gene which is located 3′ relative to the exon, and a portion of a second non-coding region of the gene which is located 5′ relative to the exon. Thus, the amplified DNA sequence may comprise both the exon and non-coding sequences flanking the exon. Each non-coding sequence may comprise an intron, the 5′-untranslated region, or the 3′-untranslated region of the gene.
[0115] A pair of such primers comprises a first primer which is complementary to a sequence comprising a portion of the first non-coding region on one strand of the genomic DNA molecule and a second primer which is complementary to a sequence comprising a portion of the second non-coding region on the opposite strand of the genomic DNA molecule. Thus, “one strand” and “the opposite strand” of the genomic DNA molecule refer to the sense strand and the antisense strand thereof, or vice versa. Addition of PCR components, such as Thermus aquaticusDNA polymerase and deoxynucleotide triphosphates (dNTPs), to the suspension of primers and genomic DNA and thermal cycling of the suspension, as is well known in the art of PCR, results in amplification of the exon and non-coding sequences flanking the exon.
[0116] Primers which are useful for amplifying an exon and non-coding sequences flanking the exon of a Type I or Type IX collagen gene may be constructed using the corresponding consensus COL1A1, COL1A2, COL9A1, COL9A2, or COL9A3 gene sequence as described herein. Construction of primers useful for PCR amplification of a region of a gene is well within the level of ordinary skill in the art of PCR, when the worker is provided with one of the Type I or Type IX collagen gene consensus sequences described herein. Such construction is described, for example, in Innis et al.(ed., 1990, In: PCR Protocols, Academic Press, Inc., San Diego). Preferably, at least one primer of a pair of primers used to amplify an exon of a Type I or Type IX collagen gene is complementary to at least one nucleotide of a non-coding region of the gene, whereby at least one nucleotide of a non-coding region of the gene is amplified along with a nucleic acid comprising at least a portion of the exon or a longer portion of the non-coding region or both. More preferably, at least three nucleotides of a non-coding region of the gene are amplified along with the nucleic acid. Preferably, at least fifteen nucleotides of a non-coding region of the gene are amplified along with the nucleic acid. It is particularly preferred that the each primer of the pair of primers is complementary to a sequence which is located entirely within a non-coding region of the gene.
[0117] Preferably, the primers of the invention have a relationship with a non-coding sequence of one of the Type I or Type IX collagen genes described herein, wherein the relationship is selected from the group consisting of a homologous relationship and a complementary relationship. Restated another way, it is preferred that the primers of the invention are complementary to either a portion of a non-coding region on the coding strand of a nucleic acid comprising a portion of one of the Type I or Type IX collagen genes described herein or a portion of a non-coding region on the non-coding strand of a nucleic acid comprising a portion of one of these genes. Preferably, the primer of the invention is either homologous or complementary to a portion of a Type I or Type IX collagen gene described herein, wherein the portion is located entirely within a non-coding region of the gene. Preferably, the primer of the invention is either homologous or complementary to a portion of one of the non-coding regions of the COL1A1 gene listed in FIGS. 4B-1 through 4B-52 (SEQ ID NOs: 6-57), to a portion of one of the non-coding regions of the COL1A2 gene listed in FIGS. 5B-1 through 5B-57 (SEQ ID NOs: 58-110), to a portion of one of the non-coding regions of the COL9A1 gene listed in FIGS. 10B-1 through 10B-40 (SEQ ID NOs: 111-149), to a portion of one of the non-coding regions of the COL9A2 gene listed in FIGS. 11B-1 through 11B-37 (SEQ ID NOs: 150-182), or to a portion of one of the non-coding regions of the COL9A3 gene listed in FIGS. 12C-1 through 12C-34 (SEQ ID NOs: 183-215).
[0118] A primer that is homologous to a sequence may be entirely or partially homologous to the sequence, and that a primer that is complementary to a sequence may be entirely or partially complementary to the sequence. It is most preferred that each primer described herein is either entirely homologous or entirely complementary to the corresponding sequence. However, one skilled in the art would appreciate that primers which are partially homologous or partially complementary may be used in the methods and compositions described herein. Each primer may be at least 95% homologous or at least 95% complementary to the corresponding sequence. Each primer may also be at least 90% homologous or at least 90% complementary to the corresponding sequence. Furthermore, each primer may be at least 85% homologous or at least 85% complementary to the corresponding sequence. Further, each primer may be at least 80% homologous or at least 80% complementary to the corresponding sequence. It is important that each primer is sufficiently homologous or sufficiently complementary to the corresponding sequence that the primer will bind to the sequence with which binding is desired under the conditions used in the methods and compositions described herein.
[0119] In one aspect of the invention, the nucleotide sequence of an amplified nucleic acid sequence comprising an exon of a Type I or Type IX collagen gene and at least one nucleotide from a non-coding region adjacent the exon is determined, and the sequence is compared with the consensus nucleotide sequence of the gene. A difference between the sequence and the consensus sequence is an indication that the gene is an altered gene. The amplified nucleic acid sequence preferably comprises at least five nucleotides from a non-coding region of the gene corresponding to the amplified nucleic acid. More preferably, the amplified nucleic acid sequence comprises at least about twenty nucleotides from non-coding regions of the gene corresponding to the amplified nucleic acid, and more preferably the amplified nucleic acid sequence comprises at least about eighty such nucleotides.
[0120] The method of the invention may be used to determine the sequence of a portion of a coding region of a collagen gene obtained from a subject. By comparing the sequence of the coding portion of the subject's COL1 or COL9 gene with the sequence of the corresponding coding portion of the collagen gene obtained from a subject who is not afflicted with a pathological condition associated with an altered COL1 or COL9 gene sequence, the existence, genetic location, and sequence of an alteration of the COL1 or COL9 gene sequence of the subject may be determined. Consensus sequences of COL1 and COL9 genes are presented herein in FIGS. 4 and 5 and 10 through 12. Any of the primers described herein may be used to amplify the corresponding portion of the corresponding gene.
[0121] Relevant portions of the COL1 or COL9 gene sequences of subjects having a genetic relationship to an affected subject who has been determined to possess an altered COL1 or COL9 gene sequence may be screened using a simplified procedure. In this procedure, the PCR amplification and nucleotide sequencing procedures described in this Example are performed using only the primers useful for amplifying and sequencing the region(s) of the COL1 or COL9 gene which is altered in the affected subject. Using this simplified procedure, it may be conveniently determined whether a subject having a genetic relationship to an affected subject has the altered gene sequence.
[0122] One skilled in the art would appreciate that any pair of primers described herein is useful in the methods described in this Example. One skilled in the art would also appreciate that the methods described in this Example are useful for identifying an altered COL1 or COL9 gene sequence associated with any pathological condition associated with an altered COL1 or COL9 gene sequence including, but not limited to, osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease. One skilled in the art would further appreciate that the methods described in this Example are useful for identifying an altered COL1 or COL9 gene sequence associated with a pathological condition even if the association between the altered sequence and the pathological condition is not presently recognized.
[0123] The PCR/CSGE methods of the present invention are useful for detection of any mutation in a COL1A1 or COL1A2 gene of a subject which is associated with a pathological condition. The PCR/CSGE methods of the present invention are applicable to detection of mutations in any Type I or Type IX collagen gene which is associated with a pathological condition, wherein a set of intronic primers have been adapted for use in the methods and wherein the set of intronic primers is useful for amplifying substantially all exons, exon flanking regions, and 3′- and 5′-untranslated regions of the gene. Such pathological conditions include, but are not limited to, osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease. The compositions and methods described herein are useful for identifying an altered COL1A1 or COL1A2 gene sequence associated with a pathological condition, even if the association between the altered sequence and the pathological condition is not presently recognized.
[0124] Three sets of intronic primers are described herein in FIGS. 23, 24, and 25. These intronic primers have been designed for use in PCR/CSGE methods described elsewhere herein for screening the human COL9A1, COL9A2, and COL9A3 genes for alterations associated with a pathological condition. One skilled in the art would appreciate that the intronic primers and PCR/CSGE methods described herein are useful for detection of any mutation in a COL9A1, COL9A2, or COL9A3 gene of a subject which is associated with a pathological condition. Such pathological conditions include, but are not limited to, osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease. One skilled in the art would further appreciate that the compositions and methods described herein are useful for identifying an altered COL9A1, COL9A2, or COL9A3 gene sequence associated with a pathological condition, even if the association between the altered sequence and the pathological condition is not presently recognized.
[0125] Any method of determining the nucleotide sequence of the amplified DNA sequence may be used in the methods of the invention including, but not limited to the method of Sanger et al. (1977, Proc. Natl. Acad. Sci. USA 74:5463-5467) or Sambrook et al. (1989, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, New York).
[0126] In a preferred method of determining the nucleotide sequence of an amplified DNA molecule, a sequencing primer complementary to a region of one strand of the amplified DNA molecule is used as a reactant in any of numerous DNA nucleotide sequence determination assays including, but not limited to, the method of Sanger et al. (1977, Proc. Natl. Acad. Sci. USA 74:5463-5467) or Sambrook et al. (1989, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, New York). Preferably, the sequencing primer is a primer adapted for use with the exon of the amplified DNA molecule.
[0127] In a particularly preferred method of identifying an altered Type I or Type IX collagen gene, substantially all relevant segments of at least one Type I or Type IX collagen gene are amplified using the primers described herein, wherein a relevant segment comprises an exon, at least six nucleotides located sequentially adjacent the 5′-end of the exon in the corresponding non-coding region, and at least twenty nucleotides located sequentially adjacent the 3′-end of the exon in the corresponding 3′-non-coding region. Preferably, at least twenty or thirty nucleotides sequentially adjacent the 5′-end of the exon and at least twenty or thirty nucleotides sequentially adjacent the 3′-end of the exon are amplified.
[0128] By way of example, substantially all relevant segments of the COL1A1 gene may be amplified using every pair of primers depicted in FIG. 21 (SEQ ID NOs: 216-324); substantially all relevant segments of the COL1A2 gene may be amplified using every pair of primers depicted in FIG. 22 (SEQ ID NOs: 325-434); substantially all relevant segments of the COL9A1 gene may be amplified using every pair of primers depicted in FIG. 23 (SEQ ID NOs: 435-477, 480-499, and 502-513); substantially all relevant segments of the COL9A2 gene may be amplified using every pair of primers depicted in FIG. 24 (SEQ ID NOs: 514-563 and 568-575); substantially all relevant segments of the COL9A3 gene may be amplified using every pair of primers depicted in FIG. 25 (SEQ ID NOs: 576-639).
[0129] In another particularly preferred method of identifying an altered Type I or Type IX collagen gene, the primers used to amplify one or more sequences of the gene are selected such that the amplified DNA oligonucleotide is suitable for CSGE analysis.
[0130] Oligonucleotides suitable for CSGE analysis generally range in length from about fifteen base pairs to about two thousand base pairs. Preferably, such oligonucleotides range in length from about fifty base pairs to about fifteen hundred base pairs, even more preferably from about one hundred base pairs to about one thousand base pairs. Most preferably, an oligonucleotide suitable for CSGE analysis ranges in length from about two hundred base pairs to about five hundred base pairs.
[0131] According to the data presented herein, a pair of primers useful for amplification of a sequence for CSGE analysis should be selected such that the sum of the lengths of the sequence to which one primer of the pair is homologous, the sequence to which the other primer of the pair is complementary, and the sequence located between these two sequences is greater than or equal to about two hundred and less than or equal to about five hundred nucleotides. If more than one pair of primers is to be used to generate more than one amplified oligonucleotide in the same reaction mixture, it is preferred that the each amplified oligonucleotide have a length that is different than the length of all other amplified oligonucleotides produced in the mixture, so that every amplified oligonucleotide is identifiable as a discrete band following gel electrophoretic separation of the oligonucleotides in the reaction mixture.
[0132] Pairs of primers which have been designed such that the oligonucleotide amplified using any pair of the primers is suitable for CSGE analysis are depicted in FIG. 21 (SEQ ID NOs: 216-324), corresponding to primer pairs useful for amplifying COL1A1 gene sequences, in FIG. 22 (SEQ ID NOs: 325-434), corresponding to primer pairs useful for amplifying COL1A2 gene sequences, in FIG. 23 (SEQ ID NOs: 435-477, 480-499, and 502-513), corresponding to primer pairs useful for amplifying COL9A1 gene sequences, in FIG. 24 (SEQ ID NOs: 514-563 and 568-575), corresponding to primer pairs useful for amplifying COL9A2 gene sequences, and in FIG. 25 (SEQ ID NOs: 576-639), corresponding to primer pairs useful for amplifying COL9A3 gene sequences.
[0133] The primer pairs depicted in each of FIGS. 21 through 25 have been designed such that all primer pairs depicted in any one of the Figures may be used to amplify substantially all important regions of the corresponding gene, such that the amplified oligonucleotides made thereby are suitable for CSGE analysis. In addition, the design of the primers yields unique fragments in high yield when they are used to amplify gene sequences obtained from a human genome.
[0134] CSGE methods, reagents, and compositions have been described (Ganguly et al., 1993, Proc. Natl. Acad. Sci. USA 90:10325-10329; Ganguly et al., 1995, Electrophoresis 16:1830-1835; U.S. patent application Ser. No. 08/468,551). One skilled in the art would appreciate that the selection of apparatus, sources of reagents, primer synthesis methods, and manual or automated gel imaging and analysis techniques useful in the CSGE methods described herein are within the level of ordinary skill in the art, given the disclosure provided herein.
[0135] The nucleic acid which is obtained from a subject for amplification using one or more pairs of primers described herein and subsequent CSGE analysis may be any type of nucleic acid obtained from essentially any cell or tissue of the subject. By way of example, the nucleic acid may be a single-stranded RNA molecule, an mRNA molecule, an mRNA molecule obtained prior to mRNA processing, a single-stranded DNA molecule, a double-stranded DNA molecule, or a complementary DNA (cDNA) molecule derived from an mRNA molecule. Preferably, the nucleic acid is either a cDNA molecule derived from an mRNA molecule obtained from a skin fibroblast of the subject or a double-stranded DNA molecule obtained from any cell of the subject. More preferably, the nucleic acid is a double-stranded DNA molecule obtained from a blood cell of the subject.
[0136] The subject may be a member of any species for which at least one intronic nucleotide sequence of at least one gene encoding a Type I or Type IX collagen chain is known. Preferably, the subject is a mammal, more preferably, a primate. Most preferably, the subject is a human.
[0137] In one aspect of the invention, pairs of primers which are useful for amplifying substantially all exons and exon flanking regions of a Type I or Type IX collagen gene are provided together in a kit for analyzing that gene. Such a kit comprises a plurality of pairs of primers, wherein at least one pair of primers is useful for amplification of a segment of the gene, the segment comprising at least one nucleotide located in a non-coding region of the gene. By way of example, all of the primer pairs depicted in FIG. 21 may be provided together in a kit useful for CSGE analysis of substantially all exons and exon flanking regions of the COL1A1 gene. Similarly, all of the primer pairs depicted in FIG. 22 may be provided together in a kit useful for CSGE analysis of substantially all exons and exon flanking regions of the COL1A2 gene. All of the primer pairs depicted in FIG. 23 may be provided together in a kit useful for CSGE analysis of substantially all exons and exon flanking regions of the COL9A1 gene. All of the primer pairs depicted in FIG. 24 may be provided together in a kit useful for CSGE analysis of substantially all exons and exon flanking regions of the COL9A2 gene. All of the primer pairs depicted in FIG. 25 may be provided together in a kit useful for CSGE analysis of substantially all exons and exon flanking regions of the COL9A3 gene.
[0138] In another aspect of the invention, a kit such as one of those described in the preceding paragraph further comprises a plurality of oligonucleotides generated by amplification of the consensus Type I or Type IX collagen gene using the primers of the kit. Although the kits described in the previous paragraph are useful for detecting heteromorphic alleles in a subject, these kits may not be able to detect the presence in a subject of two identical alleles of a Type I or Type IX collagen gene, wherein both alleles are altered relative to the consensus gene sequence. Kits which further comprise a plurality of oligonucleotides made by amplification of the consensus Type I or Type IX collagen gene using the primers of the kit can be used to detect altered gene sequences, even in subjects having two identical altered alleles. Sequences of a plurality of oligonucleotides are not provided herein, the method of obtaining such sequences being clear to one of ordinary skill in the art. To obtain such sequences, a set of primer pairs as described herein is used to amplify the exons and exon flanking regions of a nucleic acid having a sequence comprising the consensus gene sequence.
[0139] One skilled in the art would appreciate that the primers and probes described herein may be used in any set of reaction conditions wherein hybridization is possible between the primer or probe and a nucleic acid to which it is complementary. By way of example, the primers and probes may be used in standard PCR and hybridization conditions known in the art. The primers or probes may be suppled, for example, in solution, in lyophilized form, affixed to a well of a 96-well plate, affixed to a particular region of a solid support such as a glass slide, a microchip, and the like.
[0140] Testing Subjects Genetically Related to a First Subject for an Altered Type I or Type IX Collagen Gene
[0141] Once the location of an alteration in a gene encoding a chain of Type I or Type IX collagen is identified in a first subject using the PCR/CSGE methods described herein, simpler methods can be used to detect the presence of the same alteration in subjects who are genetically related to the first subject. The PCR/CSGE methods described herein are not only useful for detecting an altered collagen gene in a first subject who presents symptoms of a disease or disorder related to abnormal Type I or Type IX collagen production, but are also useful for designing protocols to screen subjects who are genetically related to the first subject for the presence of the altered gene. Non-limiting examples of methods of designing such protocols are described in this Example.
[0142] In one aspect, the PCR/CSGE methods described herein involve using a multiplicity of primer pairs to identify an alteration in one or more genes which encode a chain of Type I or Type IX collagen.
[0143] By way of example, a kit may be used which comprises pairs of primers which are useful for amplifying substantially all exons and exon flanking sequences of each of the COL1A1, COL1A2, COL9A1, COL9A2, and COL9A3 genes. The kit may also comprise all of the reagents necessary for CSGE analysis of the amplified sequences. Therefore, using this kit, a practitioner may detect the presence of heteromorphic alleles of any of the COL1A1, COL1A2, COL9A1, COL9A2, and COL9A3 genes in a first subject by identifying the presence of heteroduplex products in the CSGE gel. Furthermore, by identifying the particular amplified sequence in which heteroduplex formation occurs, the primer pair used to amplify the particular sequence is indicated as being useful for identifying the altered particular sequence in subjects genetically related to the first subject. Once this primer pair is identified, PCR amplification of the DNA of subjects who are genetically related to the first subject may be performed using only this primer pair, instead of using each of the multiplicity of primer pairs of the kit. Thus, once this primer pair is identified, PCR/CSGE methods of the invention may be performed using only a single pair of primers to identify heteroduplexes of the particular sequence in subjects genetically related to the first subject.
[0144] If the kit described in the preceding paragraph further comprises sequencing primers useful for determining the nucleotide sequence of each amplified sequence, then the sequencing primers may be used to determine the nucleotide sequence of the particular sequence in subjects genetically related to the first subject without the need to use the PCR/CSGE methods described herein for those subjects. Subjects who are homozygous for an altered Type I or Type IX collagen gene sequence may be identified using this method because the nucleotide sequence of the altered gene, rather than heteroduplex formation is used to determine whether the subjects have an allele comprising the altered sequence.
[0145] Alternately, once an altered Type I or a Type IX collagen gene has been identified in a first subject, the nucleotide sequence of the amplified sequence comprising the alteration may be determined, and an oligonucleotide probe which binds specifically to the altered sequence of the gene, but not to the normal sequence of the gene, may be designed using methods well known in the art. A single-stranded nucleic acid obtained from a subject genetically related to the first subject may be contacted with the oligonucleotide probe under hybridizing conditions, and the degree of hybridization between the probe and the single-stranded nucleic acid may be observed. Hybridization between the probe and the single-stranded nucleic acid is an indication that the subject has the same collagen gene alteration as the first subject.
[0146] By way of example, if an alteration of the COL1A1 gene of a first subject is identified at nucleotide 100, wherein the altered gene consists of a first nucleotide at that position and the normal COL1A1 gene consists of a different nucleotide at that position, an oligonucleotide probe may be made which is complementary to the altered gene sequence from nucleotide 90 through nucleotide 108 and which hybridizes with a portion of the gene comprising the altered region, but not with the same portion of the gene obtained from a normal individual. This probe may be contacted with a single-stranded nucleic acid obtained from a subject, whereby hybridization between the probe and the single-stranded nucleic acid is an indication that the single-stranded nucleic acid comprises a sequence which is homologous to nucleotides 90 to 108 of the altered COL1A1 gene. Alternately, primers may be designed based on the sequence of the COL1A1 gene described herein which are useful for amplifying a nucleic acid sequence comprising nucleotides 90 to 108 of the COL1A1 gene. DNA obtained from the subject may be amplified using PCR methods prior to contacting the oligonucleotide probe with the amplified DNA.
[0147] Single-stranded nucleic acid may be obtained from the subject in the form of a single-stranded nucleic acid or in the form of a double-stranded nucleic acid or may be synthesized using methods well known in the art from a single- or double-stranded nucleic acid obtained from the subject (e.g. Sambrook, et al., 1989, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, New York).
[0148] Any known method of detecting the presence of a nucleotide sequence in a sample obtained from a subject may be used to detect the presence in the subject of an altered COL1 or COL9 gene sequence identified by the methods described herein. By way of example, nucleic acid hybridization assays such as dot or slot blot assays or Southern transfer of DNA fragments after separation on agarose gel are well known. Other methods apparent to one skilled in the art may be used including, but not limited to, restriction endonuclease digestion of amplified DNA followed by agarose gel electrophoresis and visualization of the DNA by ethidium bromide, if the alteration in the collagen gene has created or destroyed a restriction endonuclease recognition site.
[0149] The methods of detecting an alteration in a Type I or Type IX collagen gene are useful for diagnosing the cause of an existing disease or disorder associated with abnormal production of a Type I or Type IX collagen chain as well as for detecting a genetic predisposition of a subject to develop such a disease or disorder.
[0150] The compositions and methods described herein may be used to detect a predisposition for or to diagnose any disease or disorder associated with abnormal production of a Type I or Type IX collagen chain. Diseases and disorders associated with abnormal production of a Type I or Type IX collagen chain include both those diseases and disorders which are presently known to be associated with abnormal production of a Type I or Type IX collagen chain and those disease and disorders which are, but are not presently recognized as being, associated with abnormal production of a Type I or Type IX collagen chain. Diseases and disorders which are presently known to be associated with abnormal production of a Type I or Type IX collagen chain include, but are not limited to, osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease.
[0151] Pathological Conditions Associated with Altered COL1 and COL9 Genes
[0152] The structures of various types of collagen in heritable collagen diseases has been reviewed (Prockop et al., 1984, New Eng. J. Med. 311:376-386; Prockop, 1985, J. Clin. Invest. 75:783-787; Prockop, 1986, Hosp. Pract., Feb. 15, 1986; Prockop, 1990, J. Biol. Chem. 265:15349-5352; Kuivaniemi et al., 1991, FASEB J. 5:2052-2060; Kuivaniemi et al., 1997, Hum. Mutat. 9:300-315). Numerous pathological conditions are associated with an alteration in a COL1 or a COL9 gene, including, but not limited to, osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease.
[0153] Osteoporosis
[0154] Osteoporosis is an important cause of serious disability in humans. Osteoporosis is characterized by a marked decrease in bone mass (osteopenia) and an associated susceptibility to bone fracture resulting from relatively minor trauma. Two forms of osteoporosis have been described. Type I osteoporosis, also designated ‘post-menopausal osteoporosis,’ primarily affects women beginning ten to fifteen years after menopause. Type I osteoporosis is associated with a transient acceleration of bone loss that coincides with a decrease in estrogen level in women at menopause and is believed to be familial. Type II osteoporosis, also designated ‘age-related osteoporosis,’ occurs in patients over the age of about seventy and afflicts each gender with approximately equal frequency. Both types of osteoporosis are associated, at least in part, with insufficient accumulation of skeletal mass in young adulthood. Normally, bone mass increases until about age thirty and decreases thereafter. When a patient's bone density decreases below an approximate threshold value, the patient becomes susceptible to bone fracture caused by relatively minor trauma.
[0155] Bone loss due to osteoporosis cannot be effectively reversed by any known therapy. However, the rate of bone loss can be decreased by administration of estrogen or a related agent. It is generally not advisable to administer estrogen or a related agent in the relatively large dose necessary for efficacy against osteoporosis to all individuals. Estrogen therapy can have significant side effects, such as feminizing effects in men and increasing the incidence of cervical and breast cancer in women. However, if women predisposed to Type I osteoporosis can be identified early, it could be justifiable to accept the risk of administering estrogens or a related agent and to carefully monitor carcinogenesis so as to avoid the devastating effects of osteoporosis.
[0156] Bone is a complex structure, and its strength and resistance to fracture depends on a number of factors. A major contributor to bone strength is the network of Type I collagen fibrils that form a scaffold on which the mineral components of bone are deposited. Therefore, mutations which affect the amount or structure of Type I collagen are expected to affect the pathology of and the symptoms associated with osteoporosis.
[0157] Osteoarthritis
[0158] Osteoarthritis is a progressive disease of joints that is a cause of serious disability in humans. Osteoarthritis is defined as a progressive degeneration of the cartilaginous surfaces of joints which leads to stiffness, pain, and loss of mobility. Degeneration of the cartilaginous surfaces of joints associated with osteoarthritis can have a number of causes. For example, severe trauma to a joint or a bacterial infection in a joint can produce immediate or slowly progressive degeneration of the joint. A number of metabolic disturbances are also know to result in joint degeneration.
[0159] Cartilage and the membranes that line joints are complex structures. A major source of the strength of cartilage is the fibrils of Type II collagen. Type IX collagen, which is associated with Type II collagen, is essential for the maintenance of cartilage integrity under conditions of normal physical activity. Type II collagen fibrils are stretched into three-dimensional arcades primarily by the presence of another group of macromolecules called proteoglycans including, but not limited to, Type IX collagen. Proteoglycans are highly charged and, therefore, absorb water and salts and thereby extend the arcades of Type II collagen fibrils. As a result, a highly resilient structure is formed that can withstand the intermittent pounding and pressures that joints must undergo.
[0160] Collagens, proteoglycans, and other proteins found in the matrix of cartilage are synthesized by cells embedded within the matrix. The matrix is actively synthesized during embryonic development of certain tissues and during periods of growth. The rates of synthesis and degradation of the matrix are lower during adult life than earlier in life. However, throughout life, a continual slow synthesis and degradation of cartilage occurs, particularly in response to the pressures associated with physical activity.
[0161] The degeneration of joint cartilages that occurs in osteoarthritis is caused by a failure of the cartilage to maintain its structural integrity. In this process, the cartilage surface is eroded by physical pressures and is not adequately replaced by the new synthesis of cartilage. Instead of adequate repair of cartilage, secondary changes occur in the joint surface and in the joint. These changes include, for example, inflammatory responses characterized by invasion of white cells and macrophages, abnormal deposition of mineral in the form of calcium and phosphate within the joint space and in the cartilage itself, deposition of fibers of Type I and other collagens; that are not normally part of cartilage or the joint, abnormal growth of cartilage calls and matrix at locations adjacent to the joint surface and abnormal calcifications of the joints and associated structures. As part of the complex changes that occur, the cells of the cartilage or the invading cells from the blood stream begin to secrete degradative enzymes that further contribute to the degradation of the joint structures. As a primary component of cartilage that accounts for much of the maintenance of the structural integrity thereof, Type IX is likely to be associated with many diseases and disorders characterized by abnormal cartilage degradation, in addition to the diseases and disorders with which Type IX collagen is presently associated.
[0162] The term “osteoarthritis”, as used herein, includes, but is not limited to, ankylosing spondylitis.
[0163] Chondrodysplasia
[0164] In the more severe diseases of cartilage known as chondrodysplasias, serious defects in the formation of cartilage are apparent early in life and there is a failure of joints to develop their normal size and shape. There is also a secondary failure of bone growth seen in these diseases. Moreover, there can be a failure of normal development of many tissues such as failure to achieve closure of normal partitions between oral and nasal passages, known as cleft palate, and improper development of the vitreous gel of the eye that causes severe myopia and retinal detachment.
[0165] Research has demonstrated that some forms of osteoarthritis and related conditions are caused by mutations in the genes that code for and, therefore, determine the structure of the collagens that are the major source of the strength of cartilage. Type IX collagen, for instance, is associated with Type II collagen in hyaline cartilage and in the vitreous body of the eye. Thus, alterations in one or more of the COL9 genes can be expected to result in the production of cartilage that does not have the strength of normal collagen, resulting in joint cartilage degeneration associated with normal physical activity.
[0166] Mutations affecting cartilage strength and durability produce drastic effects during growth and development. Some individuals who inherit mutated collagen genes develop severe chondrodysplasias and die in utero or shortly after birth. Alternatively, such individuals can have serious deformities such as dwarfism which shows severe malformation of joints and may be associated with conditions of severe myopia, myopia with retinal detachment and blindness, cataracts, cleft palate, and unusual facial appearance other similar mutations in the same genes may produce much milder effects and cause progressive generalized osteoarthritis in which affected individuals are apparently normal until middle age when they develop progressive stiffness, pain and then immobility of many joints.
[0167] Mutations of the gene for Type II procollagen and collagen have been shown to cause these disorders. Research suggests that some of the conditions are caused by mutations in other genes, such as the COL9 genes.
[0168] Multiple Epiphyseal Dysplasia
[0169] MED is a relatively common disorder which is characterized by symptoms ranging from mild joint stiffness, pain in large joints, early onset osteoarthritis, and mild to moderate shortness of stature (Briggs et al., 1994, Am. J. Hum. Genet. 55:678-684).
[0170] A mutation of a 5′-donor splice site was recently found in the third intron of the human COL9A2 gene in a family afflicted with multiple epiphyseal dysplasia (MED; Muragaki et al., 1990, Proc. Natl. Acad. Sci. USA 87:2400-2404). The mutation caused an in-frame deletion of a portion of the gene encoding twelve amino acids in the COL3 domain of α2(IX). Mutations in COMP, the gene encoding cartilage oligomeric matrix protein, cause Ribbing, Fairbank, or unclassified forms of MED (Briggs et al., 1995, Nature Genet. 10:330-336; Cohn et al., 1996, Ann. N.Y. Acad. Sci. 785:188-194; Ballo et al., 1997, Am. J. Med. Genet. 68:396-400). Furthermore, a linkage to the COMP or the COL9A2 genes has been excluded at least in one family afflicted with MED (Deere et al., 1995, Am. J. Hum. Genet. 56:698-704). ecause the COL9A1, COL9A2, and COL9A3 genes encode different chains of Type IX protein, and because mutations in COL9A2 are associated with MED, mutations in the COL9A1 and COL9A3 genes are also likely to be associated with MED.
[0171] Osteogenesis Imperfecta
[0172] Osteogenesis imperfecta (OI) is a heritable disorder that causes varying degrees of bone fragility and defects in several other tissues which are rich in Type I collagen. Almost 200 mutations in COL1A1 and COL1A2 have been detected in subjects afflicted with OI (Prockop, 1990, J. Biol. Chem. 265:15349-15352; Byers, 1993, In: Connective tissue and its heritable disorders, Royce et al., eds., Wiley-Liss, New York, pp 317-351; Prockop et al., 1995, Annu. Rev. Biochem. 64:403-434; Kuivaniemi et al., 1997, Hum. Mutat. 9:300-315). The most severe variants of the disease (OI types II, III, and IV) are caused primarily by single base substitutions that convert a codon for an obligate glycine in the triple helix of the proα1(I) or proα2(I) protein to a codon for an amino acid with a bulkier side chain, which distorts the conformation of the triple helix of the Type I collagen molecule. Most mutations in the mildest (Type I) variants of OI cause decreased expression of proα1(I) chains. Alterations resulting in decreased proα1(I) expression comprise an alteration of an amino-acid-encoding codon to a premature termination codon or an alteration of an RNA splicing site of the COL1A1 gene (Willing et al., 1994, Am. J. Hum. Genet. 55:638-647; Willing et al., 1992, Am. J. Hum. Genet. 51:508-515; Willing et al., 1996, Am. J. Hum. Genet. 59:799-809; Redford-Badwal et al., 1996, J. Clin. Invest. 97:1035-1040; Körkkö et al., 1997, Hum. Mutat. 9:148-156).
[0173] DNA linkage studies indicate that over 90% of subjects afflicted with OI have an altered COL1A2 or COL1A2 gene (Sykes et al., 1990, Am. J. Hum. Genet. 46:293-307). Practitioners employing prior art protocols have not been able to identify many altered COL1 and COL9 genes, no doubt owing in part to the capacity of such protocols to identify mutations only in the coding regions of the genes and in a very limited number corresponding non-coding sequences. Hence, a critical unmet need remains for compositions and protocols which are capable of identifying mutations in any part of any of the COL1 and COL9 genes.
[0174] Shortness of Stature
[0175] Because alteration of a subject's COL1A1 and COL1A2 gene can result in abnormalities in the bulk and strength of the subject's bones, particularly during development and growth of the subject, detection of an altered COL1A1 and COL1A2 gene in a subject is an indication that the subject is genetically predisposed to shortness of stature. By way of example, repeated bone fractures caused by an alteration in the subject's COL1A1 or COL1A2 gene can impede proper stature development. If the alteration in the COL1A1 or COL1A2 gene is identified prior to the onset of clinical symptoms of shortness of stature, the subject can be monitored closely for such symptoms and therapeutic intervention may begin before symptoms develop or shortly thereafter. Detection in an alteration in the COL1A1 or COL1A2 gene of a subject who has been diagnosed with shortness of stature can reveal the physiological cause of the condition, potentially eliminating the involvement of other causes and, where the subject is experiencing bone growth and development, permitting therapeutic intervention. It is understood that bone tissue generally develops continuously over the course of a subject's lifetime, old bone tissue being replaced by new bone tissue.
[0176] Similarly, the presence in a subject of abnormal cartilage can result in improper development of the stature of the subject, particularly where the abnormality of the subject's cartilage is caused by an alteration in the COL9A1, COL9A2, or COL9A3 gene of the subject.
[0177] Scoliosis
[0178] In light of the observation that a structural defect in the proα2(I) chain was found in a family afflicted with osteoporosis and idiopathic scoliosis (Shapiro et al., 1989, Connect Tissue Res. 21:117-123), the compositions and methods of the invention can be used to detect an altered COL1 or COL9 gene, wherein the altered gene causes the subject harboring the altered gene to be afflicted with scoliosis.
[0179] Degenerative Joint Disease
[0180] Because Type IX collagen appears to be required for maintaining the integrity of cartilage structures, it is expected that detection of altered COL9 gene sequences are associated with degenerative joint diseases including. Degenerative joint diseases include arthritis, osteoarthritis, rheumatism and other diseases which are characterized by irritation, inflammation, or erosion of joint cartilage or with stiffness, pain, or deformity of joints.
[0181] Other COL1- or COL9-Associated Diseases
[0182] It is clear that the compositions and methods of the invention can be used to identify any alteration in a COL1 or COL9 gene, whether the alteration is a neutral variant of the gene, an alteration which causes a disease or disorder recognized as being associated with an altered COL1 or COL9 gene, or an alteration which causes a disease or disorder which is, but is not recognized to be, associated with an altered COL1 or COL9 gene. By way of example, it is known that Type IX collagen is a component of the vitreous of the eye. It is therefore to be expected that certain alterations of the genes encoding Type IX collagen will result in degenerative and other disorders of the eye. Furthermore, the presence of Type IX collagen in cartilaginous structures causes one skilled in the art to expect that alterations in the genes encoding Type IX collagen can result in pathological conditions involving such structures, such as cleft palate. Type I collagen is a component of tooth tissue, and both Type I and Type IX collagens are components of the sound-transmitting and -detecting structures of the mammalian ear. Thus, it is to be expected that alterations in the genes encoding Type I collagen can be associated with diseases and disorders of the teeth and ear, and that alterations in the genes encoding Type IX collagen can be associated with diseases and disorders of the ear, such as deafness. It is emphasized that the compositions and methods of the invention are useful for detecting an alteration in a COL1 or COL9 gene, regardless of whether that alteration is presently known to result in affliction of the subject harboring the altered gene with a disease or disorder.
[0183] Disease- or disorder-causing alterations in a COL1 or COL9 gene may be distinguished from neutral variants of the gene in numerous ways. By way of example, alterations in a protein-encoding region of the gene which do not cause an alteration of the amino acid sequence of the protein can be neutral variants, although it is recognized that substitution of a codon encoding a particular amino acid for a rare codon encoding the same amino acid can result in decreased expression of the gene. Where the alteration in the gene results in an alteration of the amino acid sequence of a protein encoded by the gene, the consequence of the alteration of the amino acid sequence can be predicted by reference to data in the scientific literature regarding the function of the protein. By way of example, as described herein, substitution of the Gly residue in a Gly-X-Y triplet in the α chain portion of a collagen chain can cause deformation of the three-dimensional conformation of a collagen subunit comprising the collagen chain. Further by way of example, a recombinant collagen protein comprising a chain encoded by the altered collagen gene can be expressed by cells in culture, and abnormal characteristics of the cells, or of the collagen produced by the cells, can be determined. Disease- or disorder-causing alterations in collagen genes may also be distinguished from neutral variants of the genes by observing whether symptoms experienced by a subject afflicted with the disease or disorder who harbors a gene having an alteration are similarly experienced by other subjects, such as genetic relatives of the subject, who harbor a copy of the gene having the alteration. Coinheritance of the altered collagen gene and the symptoms of the disease or disorder indicates that the disease or disorder is associated with the alteration of the collagen gene. Furthermore, the presence of an alteration in a collagen gene obtained from a subject who is not afflicted with a disease or disorder is an indication that the alteration does not cause the disease or disorder.
[0184] Definitions
[0185] As used herein, the term “COL1 gene” means any gene encoding a chain of Type I collagen including, but not limited to, the human COL1A1 and COL1A2 genes and the vertebrate and mammalian homologs thereof including, but not limited to, the chicken and murine homologs thereof.
[0186] As used herein, the term “COL9 gene” means any gene encoding a chain of Type IX collagen including, but not limited to the human COL9A1, COL9A2, and COL9A3 genes and the vertebrate and mammalian homologs thereof including, but not limited to, the chicken, murine, and rat homologs thereof.
[0187] As used herein, the terms “alteration” or “mutation” of a gene mean that the altered or mutated gene has a nucleotide sequence which differs from a consensus nucleotide sequence of the gene at one or more positions.
[0188] As used herein, a “neutral variant” of a gene means a gene which has an altered sequence which is not associated with a pathological condition.
[0189] As used herein, a “chain” of a protein means a single polyamino acid molecule in the protein. A protein “chain,” as used herein, is frequently referred to in the art as a protein subunit. The term “chain” has been used herein to distinguish a chain of a protein from a “subunit” of a collagen molecule, as used herein.
[0190] As used herein, a “subunit” of a collagen molecule means an individual collagen protein molecule which is capable of being incorporated into a collagen fibril. It is understood by one skilled in the art that a single collagen fibril may include numerous collagen subunits, and that a single collagen subunit may comprise numerous protein chains. By way of example, a Type I collagen fibril obtained from bone tissue may comprise numerous Type I collagen protein subunits, each of which subunits comprises two proα1(I) chains and a proα2(I) chain.
[0191] As used herein, a “portion” of a collagen gene means a polymer comprising at least about fifteen consecutive nucleotides having a sequence substantially complementary to or substantially homologous with the nucleotide sequence of a human collagen gene, wherein said gene is selected from the group consisting of the segment of the COL1A1 gene extending in the 5′- to 3′-direction from and including the 78 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of the COL1A1 gene, the COL1A2 gene, the COL9A1 gene, the COL9A2 gene, and the COL9A3 gene, and excluding published non-coding sequences of the COL1A2 gene. Published non-coding sequences of the COL1A2 gene include the following sequences:
[0192] (i) 75 nucleotides located within intron 1
[0193] (ii) 318 nucleotides at the 3′-end of intron 5
[0194] (iii) 298 nucleotides at the 5′-end of intron 6
[0195] (iv) 30 nucleotides at the 3′-end of intron 26
[0196] (v) intron 27
[0197] (vi) intron 28
[0198] (vii) 25 nucleotides at the 5′-end of intron 29
[0199] (viii) intron 33
[0200] (Myers et al., 1983, J. Biol. Chem. 258:10128-10135; Myers et al., 1984, J. Biol. Chem. 259:12941-12944; Dickson et al., 1984, Proc. Nati. Acad. Sci. USA 81:4524-4528; Tromp et al., 1988, Proc. Natl. Acad. Sci. USA 85:5254-5258; Sherwood et al., 1990, Gene 89:238-244; Vasan et al., 1991, Am. J. Hum. Genet. 48:305-317; Ganguly et al., 1991, J. Biol. Chem. 266:12035-12040).
[0201] As used herein, a “coding region” of a collagen gene is a portion of the nucleotide sequence of said gene which encodes at least a portion of the amino acid sequence of the protein chain encoded by the gene. The terms “coding region” and “exon” are used synonymously herein.
[0202] As used herein, an “exon flanking sequence” means an intronic nucleotide or an intronic polynucleotide which is sequentially adjacent a coding region of a gene. An exon flanking sequence preferably comprises at least about six, and preferably about twenty, nucleotides.
[0203] As used herein, two nucleotides or polynucleotides are “sequentially adjacent” if both of the nucleotides or polynucleotides are part of the same polymeric nucleic acid and if no other nucleotides are located between the two nucleotides or polynucleotides in the polymeric nucleic acid.
[0204] As used herein, “substantially all exons and exon flanking sequences” of a gene means at least most exons and exon flanking sequences of a gene. Preferably, “substantially all exons and exon flanking sequences” means at least about 75% of such sequences; more preferably, the term means at least about 90% of such sequences; even more preferably, the term means at least about 95% of such sequences.
[0205] As used herein, a “non-coding region” of a collagen gene is a portion of the nucleotide sequence of said gene which does not encode a portion of the amino acid sequence of the protein chain encoded by the gene. Non-coding regions of collagen genes include, but are not limited to, 5′-untranslated regions of such genes, introns in such genes, and 3′-untranslated regions of such genes.
[0206] As used herein, an “intronic nucleotide” is a nucleotide located in a non-coding region of a collagen gene including, but not limited to, a nucleotide located in the 5′-untranslated regions of said gene, a nucleotide located in an intron in said gene, and a nucleotide located in the 3′-untranslated region of said gene.
[0207] As used herein, a “part of a non-coding region” of a collagen gene means at least one nucleotide located in a non-coding region of said gene. Preferably, said part of a non-coding region comprises at least three such nucleotides, more preferably five, and even more preferably about fifteen such nucleotides.
[0208] As used herein, the terms “a first site” and “a second site” refer to nucleotide sequences of a portion of a collagen gene, wherein each of said first site and said second site comprise at least about fifteen consecutive nucleotides of the gene.
[0209] As used herein, an oligonucleotide “probe” means a polymer comprising nucleotide residues having a sequence whereby the probe is complementary to a nucleotide sequence of interest. An oligonucleotide probe preferably comprises at least about fifteen nucleotide residues, more preferably at least about eighteen. Probes preferably have a length between about fifteen and about thirty nucleotides. The nucleotide sequence of an oligonucleotide probe is determined by the nucleotide sequence of the nucleic acid to which binding of the probe is sought, the probe being complementary to that nucleic acid.
[0210] As used herein, an oligonucleotide “primer” means a polymer comprising nucleotide residues, having a sequence complementary to a known sequence, and having a free 3′-end suitable for extension by a nucleic acid polymerase, such as a DNA polymerase, whereby when the primer is annealed with the known sequence in the presence of the components of a nucleic acid polymerase reaction mixture, the primer is extended in the 3′-direction thereof, whereby the nucleotide sequence of the extended region of the primer has a sequence complementary to the sequence located adjacent the known sequence in the 5′-direction. An oligonucleotide primer preferably comprises at least about fifteen nucleotide residues, more preferably at least about eighteen. Primers preferably have a length between about twenty and about fifty nucleotides, more preferably between about thirty and forty nucleotides. The nucleotide sequence of a primer is determined by the nucleotide sequence of the nucleic acid to which binding of the primer is sought, the primer being complementary to that nucleic acid. As is well known in the art, the primers of a pair of primers useful for PCR methods are complementary to sequences which flank the nucleotide sequence of which amplification is desired.
[0211] As used herein, an “intronic primer” is a oligonucleotide primer comprising at least one nucleotide which is either complementary to or homologous with a nucleotide located in a non-coding region of a collagen gene including, but not limited to, a nucleotide located in the 5′-untranslated regions of said gene, a nucleotide located in an intron in said gene, and a nucleotide located in the 3′-untranslated region of said gene.
[0212] As used herein, a “sequencing primer” is an oligonucleotide primer which is complementary to at least a portion of a polynucleotide and which can be elongated by a DNA or RNA polymerizing enzyme such as DNA polymerase, whereby binding of the sequencing primer to the polynucleotide and elongation of the primer using methods well known in the art yields an oligonucleotide transcript which is complementary to at least a part of the polynucleotide.
[0213] “Homologous” as used herein, refers to the subunit sequence similarity between two polymeric molecules, e.g., between two nucleic acid molecules, e.g., two DNA molecules or two RNA molecules, or between two polypeptide molecules. When a subunit position in both of the two molecules is occupied by the same monomeric subunit, e.g., if a position in each of two DNA molecules is occupied by adenine, then they are homologous at that position. The homology between two sequences is a direct function of the number of matching or homologous positions, e.g., if half (e.g., five positions in a polymer ten subunits in length) of the positions in two compound sequences are homologous then the two sequences are 50% homologous, if 90% of the positions, e.g., 9 of 10, are matched or homologous, the two sequences share 90% homology. By way of example, the DNA sequences 3′-ATTGCC-5′ and 3′-TATGCG-5′ share 50% homology.
[0214] As used herein, a first polynucleotide is “completely homologous” to a second polynucleotide if the nucleotide in every subunit position in the first polynucleotide is identical to the nucleotide in the same subunit position in the second polynucleotide. It is understood that, in addition to the region of complete homology, the second polynucleotide may comprise other nucleotide sequences.
[0215] As used herein, a first polynucleotide is “substantially homologous” to a second polynucleotide if the nucleotides in at least most subunit positions in the first polynucleotide are identical to the nucleotides in the same subunit positions in the second polynucleotide. Preferably, “substantially homologous” means that at least about 75% of nucleotides in the first polynucleotide are identical to the nucleotides in the same subunit positions in the second polynucleotide; more preferably, the term means that at least about 90% of nucleotides in the first polynucleotide are identical to the nucleotides in the same subunit positions in the second polynucleotide; even more preferably, the term means that at least about 95% of nucleotides in the first polynucleotide are identical to the nucleotides in the same subunit positions in the second polynucleotide. It is understood that, in addition to the region of substantial complementarity, both the first and the second polynucleotides may comprise other nucleotide sequences.
[0216] “Complementary” as used herein refers to the broad concept of subunit sequence complementarity between two nucleic acids, e.g., two DNA molecules. When a nucleotide position in both of the molecules is occupied by nucleotides normally capable of base pairing with each other, then the nucleic acids are considered to be complementary to each other at this position. Thus, two nucleic acids are complementary to each other when a substantial number (at least 50%) of corresponding positions in each of the molecules are occupied by nucleotides which normally base pair with each other (e.g., A:T and G:C nucleotide pairs).
[0217] As used herein, a first polynucleotide is “completely complementary” to a second polynucleotide if the nucleotide in every subunit position in the first polynucleotide is complementary to the nucleotide in the same subunit position in the second polynucleotide. It is understood that, in addition to the region of complete complementary, the second polynucleotide may comprise other nucleotide sequences.
[0218] As used herein, a first polynucleotide is “substantially complementary” to a second polynucleotide if the nucleotides in at least most subunit positions in the first polynucleotide are complementary to the nucleotides in the same subunit positions in the second polynucleotide. Preferably, “substantially complementary” means that at least about 75% of nucleotides in the first polynucleotide are complementary to the nucleotides in the same subunit positions in the second polynucleotide; more preferably, the term means that at least about 90% of nucleotides in the first polynucleotide are complementary to the nucleotides in the same subunit positions in the second polynucleotide; even more preferably, the term means that at least about 95% of nucleotides in the first polynucleotide are complementary to the nucleotides in the same subunit positions in the second polynucleotide. It is understood that, in addition to the region of substantial complementarity, both the first and the second polynucleotides may comprise other nucleotide sequences.
[0219] The terms “complementary” and “antisense” as used herein, are not entirely synonymous. “Antisense” refers particularly to the nucleic acid sequence of the non-coding strand of a double stranded DNA molecule encoding a protein, or to a sequence which is substantially homologous to the non-coding strand.
[0220] As used herein, “identifying the nucleotide sequence” of a gene means either identifying the nucleotide sequence of the entire gene or identifying the nucleotide sequence of only part of the gene.
[0221] As used herein, a gene is “heteromorphic” if the gene exists in more than one allele in a subject, and at least one of the alleles has a nucleotide sequence which differs from the nucleotide sequence of at least one other allele at at least one nucleotide.
[0222] As used herein, the term “proband” means a subject afflicted with a pathological condition, who has been diagnosed as so afflicted independently of any diagnosis of a genetic relative.
[0223] As used herein, a first subject is “genetically related” to a second subject if the first subject and the second subject have each inherited genetic material from a common ancestor. It is understood that various degrees of genetic relationship exist between individuals, the degree of genetic relationship being greater the fewer number of generations which separate the subjects. By way of example, siblings having common parents are equally genetically related to the parents, although one sibling may be more nearly genetically related to a single grandparent than the other sibling. Continuing the example, the father and mother of the siblings are not necessarily closely genetically related to one another, and thus the siblings are more closely genetically related to each of their father and their mother than the father and mother are to one another. The degree of genetic relationship between two subjects may be determined by examining the genealogies of the two patients, by identifying common genetic markers shared by the two subjects, or by other methods known in the art.
[0224] As used herein, a “pathological condition associated with an altered collagen gene” means the existence of clinically-detectable symptoms resulting from expression of at least one of a COL1 gene or a COL9 gene having a nucleotide sequence which differs from the consensus nucleotide sequence of the gene at at least one position. The terms “pathological condition associated with an altered collagen gene,” “pathological condition associated with an altered COL1 or COL9 gene,” and “disease or disorder associated with abnormal production of a Type I or Type IX collagen chain” are generally used interchangeably herein. Pathological conditions associated with an alteration in a COL1 or a COL9 gene include, but are not limited to, osteoporosis, osteoarthritis, chondrodysplasia, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, scoliosis, low bone density, and degenerative joint disease.
[0225] As used herein, a “disease or disorder associated with abnormal production of a Type I or Type IX collagen chain” means a disease or disorder associated with production of an abnormal Type I or Type IX collagen chain or production of an abnormal amount of a normal Type I or Type IX collagen chain. Preferably, the disease or disorder associated with abnormal production of a Type I or Type IX collagen chain is one known to be associated with abnormal production of a Type I or Type IX collagen chain.
[0226] As used herein, “production of an abnormal Type I or Type IX collagen chain” means production of a Type I or Type IX collagen chain having an amino acid sequence different from the amino acid sequence encoded by the corresponding consensus gene sequence described herein.
[0227] As used herein, “production of an abnormal amount of a normal Type I or Type IX collagen chain” means production in a subject of a chain of Type I or Type IX collagen having the same amino acid sequence as that encoded by the corresponding consensus gene sequence described herein, but in an amount such that the ratio of production the chain to production of a different chain of the Type I or Type IX collagen differs from the ratio of production in individuals who do not have a pathological condition associated with an altered Type I or Type IX collagen gene. This level of collagen production is referred to herein as “normal.” The normal ratio of production of the α1(I) chain to production of the α2(I) chain is two to one. The normal ratio of production of the α1(IX) chain to production of the α2(IX) chain is one to one. The normal ratio of production of the α1(IX) chain to production of the α3(IX) chain is one to one. The normal ratio of production of the α2(IX) chain to production of the α3(IX) chain is one to one. Preferably, the difference between the normal ratio and the ratio in a subject afflicted with a pathological condition associated with an altered Type I or Type IX collagen gene is at least about 1%; more preferably the difference is at least about 10%.
[0228] As used herein, the term “procollagen” means a gene product of a collagen gene which comprises at least one amino acid sequence which is present in the chain encoded by the collagen gene and which is not present in the mature α chain encoded by the collagen gene. By way of example, proα1(I) is a procollagen encoded by the COL1A1 gene. Proα(I) comprises N-telopeptidic and C-telopeptidic amino acid sequences which are not present in the mature α(I) chain.
[0229] As used herein, the term “registration” refers to the alignment of Gly-X-Y tripeptide sequences of the collagen or procollagen chains of a collagen or procollagen subunit. Gly-X-Y tripeptide sequences are ‘in register’ when the first Gly-X-Y sequence, numbered from the carboxyl terminal end, of a first collagen or procollagen chain is able to hydrogen bond to the first Gly-X-Y sequence, numbered from the carboxyl terminal end, of each of the other two collagen or procollagen chains of the collagen or procollagen subunit.
[0230] As used herein, a “substantially pure PCR product” is an amplified nucleotide or plurality of amplified nucleotides which has been separated from the components of a PCR. Such components include a DNA polymerase, oligonucleotide primers, and the like. Methods of making substantially pure PCR products from PCR mixtures are well known in the art and include, for example, electrophoretic separation of PCR mixture components on an agarose gel following the reaction.
EXAMPLES
[0231] The invention is now described with reference to the following experimental examples. These examples are provided for the purpose of illustration only and the invention should in no way be construed as being limited to these examples but rather should be construed to encompass any and all variations which become evident as a result of the teaching provided herein.
[0232] Although the methods of the invention have been demonstrated in the first instance in humans, they are expected to be useful in other mammalian species, particularly commercially important species and laboratory animal species used in models of human disease.
Example 1
Determining the Presence of an Alteration in the Coding Region of a Collagen Gene of a Subject
[0233] The following method may be used to determine the presence, genetic location, and sequence of an alteration in the coding region of a COL1 or COL9 gene of a subject. The method involves making and sequencing cDNA from mRNA obtained from cultured fibroblasts taken from the subject, and then comparing the cDNA sequence to a reference sequence, whereby alterations in the subject's sequence may be identified.
[0234] Complementary DNA (cDNA) molecules were prepared from mRNA encoding the proα1(I) chain and from mRNA encoding the proα2(I) chain of Type I collagen. These two cDNA molecules encode the entire coding sequence of each of the two chains.
[0235] To prepare cDNA, total cellular RNA was isolated from cultured skin fibroblasts by lysing the cells with the detergent, Sarcosyl, in the presence of guanidinium isothiocyanate. Isolated RNA was pelleted through a cesium chloride solution, as described (Maniatis et al., 1982, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., p. 196). cDNA was synthesized using a commercial poly(A)+RNA kit purchased from BRL (Bethesda Research Laboratories, Bethesda, Md.) or Pharmacia (Pharmacia-LKB, Piscataway, N.J.) or using reverse transcriptase and a primer specific for the proα1(I) chain such as those provided in Table 1. Synthesis of cDNA using reverse transcriptase and primer was performed as described (Maniatis et al., 1982, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., p. 213).
[0236] Double-stranded cDNA was synthesized using the method of Gubler et al. (1983, Gene 25:263-269), as amended by the manufacturers of the cDNA preparation kits. Single-stranded cDNA was synthesized using reverse transcriptase followed by alkaline hydrolysis of the RNA, as described (Maniatis et al., 1982, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., pp. 214-216). Double- or single-stranded cDNA was amplified using a PCR kit (GeneAMp™ DNA Amplification Reagent Kit, Perkin Elmer Cetus, Norwalk, Conn.), according to the instructions provided by the manufacturer. Primers complementary to portions of cDNA of the proα1(I) gene were used in the PCR. The primers which were used are described in Table 1.
[0237] To obtain the nucleotide sequence of an amplified DNA sequence, PCR was performed using an unequal amount of two primers, so that an excess of single-stranded DNA synthesized using the prevalent primer resulted. Following PCR, single-stranded DNA (ssDNA) was sequenced using the dideoxynucleotide chain termination method using internal primers and a sequencing kit (Sequenase™; United States Biochemical Corp., Cleveland, Ohio).
[0238] In a typical experiment, cDNA was used as a PCR template with a pair of the primers described in Table 1. The PCR was permitted to proceed for thirty cycles, wherein a cycle is defined as permitting the reaction to proceed at 94° C. for ninety seconds, at 56° C. or 58° C. for sixty seconds, and at 74° C. for ninety seconds. The picomolar ratio of antisense primer to sense primer was 20:4. In a second PCR, about one percent of the first PCR product was used as template for a second PCR. The second PCR was permitted to proceed for twenty cycles, as described, in which the ratio of antisense:sense primers was 50:1. The second PCR product was purified, and the volume in which it was contained was reduced using an Ultrafree™ MC filtration unit (Millipore, Bedford, Mass.; No. UFC3TTK00) to yield a solution comprising substantially purified DNA. The sequence of the substantially purified DNA was determined using the dideoxynucleotide chain termination reaction, as described (Sanger et al., 1977, Proc. Nat. Acad. Sci. USA 74:5463-5467) using modified T7 DNA polymerase (United States Biochemical Corp., Cleveland, Ohio). Conventional and radioactive sequencing using 35S-dATP autoradiography and sequencing using fluorescently labeled primers and an ABI 370A automated sequencer (Applied Biosystems International, San Francisco, Calif.) were performed.
Example 2
Isolation and Characterization of the Nucleotide Sequence of the COL1A1 Gene, Including the Sequences of Introns 25 to 51
[0239] The nucleotide sequences of the entire coding region, the 5′-untranslated region, introns 1-26, and the twenty-six nucleotides of intron 27 located adjacent to exon 26 of COL1A1, have been reported, as described herein. This Example describes the sequencing of the entire human COL1A1 gene, including the 78 nucleotides of intron 27 which are adjacent exon 28 through and including the −3′ end of the gene.
[0240] To analyze the sequence of COL1A1 extending from intron 25 to exon 40, a genomic fragment of the COL1A1 gene was cloned from DNA obtained from cultured skin fibroblasts of a subject afflicted with osteogenesis imperfecta (Tsuneyoshi et al., 1991, J. Biol. Chem. 266:15608-15613). Genomic DNA was digested with BamHI in order to generate a COL1A1 gene fragment having a length of about five kilobases. BamHI-digested genomic DNA was separated by electrophoresis on an agarose gel, and fragments having a length between about two and about six kilobases were electroeluted therefrom. Fragments were cloned into lambda phage vectors (Lambda-2AP, Stratagene, La Jolla, Calif.) to create a phage library. Individual phage clones from the phage library were screened for the presence of the COL1A1 gene using a cDNA probe (designated Hf-404) derived from COL1A1 mRNA (Maniatis et al., 1982, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.; Bernard et al., 1983, Biochemistry 22:5213-5223). A phage clone comprising at least a portion of the COL1A1 gene was isolated. A deletion library was prepared from the nucleic acid encoded by the phage clone using exonuclease III and a commercial kit (Erase-a-base, Stratagene, La Jolla, Calif.). Double-stranded DNA obtained from the deletion library was used for dideoxynucleotide sequencing (Sanger et al., 1977, Proc. Natl. Acad. Sci. USA 74:5463-5467) using T7 polymerase (Sequenase™; U.S. Biochemical Corp., Cleveland, Ohio).
[0241] The nucleotide sequence of the 78 nucleotides of intron 27 which are adjacent exon 28 through and including the −3′ end of the human COL1A1 gene is presented in FIG. 4.
Example 3
Post-Menopausal Osteoporosis and Reduced Bone Density in Members of a Family Having an Altered COL1A2 Gene
[0242] The data presented in this Example establish an association between post-menopausal osteoporosis and an altered COL1A2 gene sequence. In addition, the data establish an association between an altered COL1A2 gene sequence and reduced bone density in subjects.
[0243] A fifty-two-year old female human subject was diagnosed with post-menopausal osteoporosis. The woman was Caucasian and was evaluated after having developed acute mid-thoracic pain. X-ray examination of the subject's spine revealed the presence of an anterior compression fracture of the ninth thoracic vertebra and generalized demineralization of the spinal column consistent with osteoporosis. Bone densitometry of the lumbar spine by dual-energy X-ray absorptiometry revealed that the subject's bone density was 0.75 g/cm2, a value which is in the second lowest percentile for women of the same age, the mean normal value being 1.13 g/cm2.
[0244] On physical examination of the subject, her sclerae were found to have a slightly bluish cast, her skin was not abnormally thin, and there was no hyperextensibility of her joints. Routine laboratory tests, including serum protein electrophoresis, were normal.
[0245] The woman had experienced a normal menopause seven years prior to examination. She had no history of any disease or use of drugs known to be associated with osteoporosis. However, she had experienced repeated fractures throughout her life, from early childhood on. Dermal fibroblasts were obtained from the woman, and Type I procollagen synthesized by those cells was examined by SDS-PAGE. Delayed migration of both the proα1(I) and proα2(I) chains derived from the secreted Type I procollagen was observed, relative to migration of chains obtained from individuals who did not exhibit symptoms of a pathological condition associated with an altered Type I or Type IX collagen gene. Differential migration was more apparent when fragments obtained by treating type I collagen obtained from the woman with a vertebrate collagenase were examined. Migration of vertebrate collagenase fragment A, comprising amino acid residues 1-775 of the woman's type I collagen, was retarded in the gel. However, migration of the B fragments, comprising amino acid residues 776-1,014 of the woman's type I collagen, was not retarded in the gel. Retarded migration of the woman's collagen fragment A was found to be caused by post-translational over-modification of the α chains. Incubation of the woman's fibroblasts in a solution comprising 0.3 mM α,α′-dipyridyl, which inhibits post-translational modification of the protein by prolyl hydroxylase and lysyl hydroxylase, abolished the difference in migration rates between collagen chains obtained from the woman and those obtained from individuals not afflicted with a pathological condition associated with an altered Type I or Type IX collagen gene.
[0246] Total RNA was extracted from the fibroblasts obtained from the woman and was used to synthesize single-stranded cDNA, as described herein. cDNA was used as a template for nine separate PCRs using primer pairs that resulted in amplification of all 3,052 base pairs of coding sequence of the triple-helical domains of each of the proα1(I) and proα2(I) chains. Heterozygous single-base mutations were detected in individual PCR products by denaturing and renaturing the PCR and treating the renatured DNA with a water-soluble carbodiimide, as described (Ganguly et al., 1990, Nucl. Acids Res. 18:3933-3939). Analysis by primer extension of each of the nine PCR products indicated that one comprised a sequence heteromorphism. The region in which the sequence difference was identified spanned nucleotides 1,951 to 2,813 of COL1A2, a portion of the coding region of the gene corresponding to amino acids 516 to 803 of the triple-helical domain of proα2(I). Further analysis indicated that a sequence variation was present in the region of the COL1A2 gene encoding amino acid residues 660 to 667 of the proα2(I) chain.
[0247] The region of the COL1A2 gene comprising the sequence variation that gave rise to the mismatch observed using the carbodiimide technique was sequenced. Nucleotide sequencing was performed using the PCR product which comprised the sequence heteromorph, as well as using three other PCR products that spanned the remainder of the coding sequences corresponding to the triple-helical domain of proα2(I). Analysis of the woman's COL1A2 gene sequence identified a single-base alteration at codon 661, which normally has the nucleotide sequence GGT and encodes glycine. One of the alleles in the woman's COL1A2 gene had the nucleotide sequence AGT at codon 661, encoding serine instead of glycine. PCR products derived from the woman's DNA contained both A and G in this codon, indicating that her DNA was heterozygous for the altered COL1A2 gene. Dideoxynucleotide sequencing of the three PCR products spanning the remainder of the coding sequences for the triple-helical of the proα2(I) chain did not reveal any nucleotide sequence difference that would alter the amino acid encoded by the corresponding codon.
[0248] To confirm the presence of the mutation in the genomic DNA of this woman, the genomic DNA was amplified using PCR using intronic primers described herein. PCR products were hybridized with an oligonucleotide having a sequence which was complementary to either the normal coding sequence of the COL1A2 gene or the normal coding sequence having a single-base alteration that converted G to A at codon 661. Both oligonucleotides hybridized to PCR products derived from the patient's genomic DNA. However, the oligonucleotide having the altered base did not hybridize to PCR products derived from fifty control samples of genomic DNA obtained from individuals who were afflicted with neither post-menopausal osteoporosis nor abnormally low bone density.
[0249] Further experiments involving hybridization of the normal primer and the primer having the altered base with PCR products derived from genetic relatives of the woman demonstrated that the mutation was not present in the genomic DNA of the patient's 89-year old mother, although she had severe thoracic kyphosis and radiographic evidence of age-related or Type II osteoporosis. The mutation was present in the genomic DNA of each of the patient's three sons, ages 24, 29, and 31. These individuals had each suffered one to four fractures following trauma as adolescents. Subsequent examination of the three sons demonstrated that none had any evidence of osteogenesis imperfecta or related genetic diseases. However, all three individuals had markedly reduced bone density.
[0250] The results presented in this example establish an association between an altered COL1A2 gene and post-menopausal osteoporosis and/or reduced bone density.
Example 4
PCR/CSGE Analysis of the COL1A1 and COL1A2 Genes of Fifteen Human Patients Afflicted with Mild (Type I) Osteogenesis Imperfecta Using Intronic Primers
[0251] Although it has been estimated that over 90% of patients afflicted with osteogenesis imperfecta (OI) have a mutation in one or both of the COL1A1 and COL1A2 genes, it has been difficult to detect such mutations in all patients afflicted with the mildest forms of the disease, designated Type I OI. In this Example, the presence or absence of COL1A1 and COL1A2 mutations was examined by analysis of protein and mRNA synthesized in fibroblasts obtained from each of ten patients afflicted with Type I OI. No evidence of a mutation at the protein level was found in two of the ten patients, and no evidence of a mutation at the cDNA level was found in five of the ten patients.
[0252] Genomic DNA was obtained from the original ten patients and from an additional five patients afflicted with Type I OI. In order to assay the genomic DNA of these fifteen patients, it was necessary to identify a consensus sequence for both of the COL1A1 and COL1A2 genes, which is indicative of a normal non-mutated gene in each case. To identify these consensus sequences, it was necessary to obtain 90% of the sequence of the 38-kb COL1A2 gene and to sequence extensive regions of the 18-kb COL1A1 gene, including some regions for which sequence data had already been reported. Previously reported COL1A1 sequences did not appear to represent a consensus gene sequence.
[0253] The consensus sequences were used to develop intronic primers useful for PCR amplification of the 103 exons of the two genes and the flanking sequences bordering each of the exons. PCR amplification products were scanned for the presence of heteroduplexes by conformation sensitive gel electrophoresis (CSGE; Ganguly et al., 1993, Proc. Natl. Acad. Sci. USA 90:10325-10329; Ganguly et al., 1995, Electrophoresis 16:1830-1835), and products comprising heteroduplexes were sequenced. This method detected disease-causing mutations in thirteen of the fifteen patients and two additional probable disease-causing mutations in the remaining two patients. Analysis of the data described herein revealed common sequences for mutations resulting in null alleles.
[0254] The PCR/CSGE technique described in this Example greatly reduced the amount of DNA sequencing necessary to identify mutations in the genes of subjects afflicted with OI or with another pathological condition which is associated with an altered COL1 or COL9 gene.
[0255] The materials and methods used in this Example are now described.
[0256] Subjects
[0257] All subjects were probands who presented the typical phenotype of OI Type I. The major clinical symptoms manifested by each of them are summarized in Table 2.
[0258] Protein Analysis
[0259] Skin biopsy samples were obtained from each of subjects one through ten, and fibroblast cultures derived from each of these samples were established under culture conditions known in the art. About 35,000 cells/cm2were seeded onto the culture medium. Labeling of fibroblasts and purification of collagen was performed as reported (Nuytinck et al., 1996, Human. Genet. 97:324-329). Briefly, cells were labeled by adding to the culture medium BME medium (Basal Medium Eagle, Life Technologies, Gaithersburg, Md.) comprising one microcurie per milliliter 14C proline, 5% (v/v) dialyzed fetal calf serum (FCS), 0.05 milligrams per milliliter β-aminoproprionitrile and 0.025 milligrams per milliliter ascorbic acid. After twenty hours, the medium was removed and supplemented with protease inhibitors such that final concentrations were 0.1 milligrams per milliliter phenylmethylsulfonyl fluoride (PMSF), 0.1 milligrams per milliliter N-methylmaleimide, and 2 millimolar ethylenediaminetetraacetic acid (EDTA). The cell layer was trypsinized, and the cells were collected by centrifugation and lysed in a solution comprising 0.5% (v/v) Triton X-100 in 0.5 M acetic acid. The medium and lysed cells were pooled and centrifuged, and the supernatant was used for collagen analysis.
[0260] Collagens were isolated from the supernatant by alcohol precipitation and were resuspended in 0.5 M acetic acid. To convert procollagen to collagen, the samples were digested with 50 micrograms per milliliter pepsin for four hours at 15° C., Digestion was halted by adding 0.5 mg/ml pepstatin. SDS-PAGE was performed as described (Laemmli et al., 1970, Nature 227:680) using a 3% (w/v) stacking gel and a 5% (w/v) separating gel. Prior to loading, samples were lyophilized, redissolved in sample buffer comprising Tris-HCl, pH 6.8, 2 M urea, and 0.04% (w/v) bromphenol blue, and denatured for twenty minutes at 55° C., Electrophoresis was performed at 8° C. overnight using an applied voltage of 3.5 V/cm2. Each gel was processed for fluorography by soaking it in a solution comprising 20% (w/v) 2,5-diphenyloxazole (PPO) in 100% (v/v) acetic acid, drying it, and exposing it to a sheet of Hyperfilm MP™ (Amersham, Arlington Heights, Ill.).
[0261] Analysis of DNA Polymorphism
[0262] Total RNA was isolated from cultured skin fibroblasts using the Trizol buffer (Life Technologies, Gaithersburg, Md.) method, according to the manufacturer's instructions. Prior to cDNA synthesis, RNA was treated with RNase-free DNase (Life Technologies, Gaithersburg, Md.) to prevent contamination by genomic DNA in the reverse transcriptase-PCR (RT-PCR) experiments. Moloney murine leukemia virus reverse transcriptase was used in combination with random hexanucleotide primers to synthesize cDNA from mRNA. MnII polymorphisms in the COL1A1 gene were detected using primers and conditions as described (Sokolov et al., 1991, Hum. Genet. 88:125-129). After enzymatic digestion of the cDNA, fragment length was evaluated either by agarose gel electrophoresis or by separation on polyacrylamide gels and an automated laser fluorescent DNA sequencer (ALF™, Pharmacia LKB Biotechnology Inc., Piscataway, N.J.).
[0263] The four-base-pair insertion polymorphism at the 3′-end of the COL1A1 gene was detected by PCR amplification, wherein the sense primer had the sequence 5′-CCT TTC TGC TCC TTT CTC CA-3′ (SEQ ID NO: 645) and the antisense primer had the sequence 5′-AGC AAC ACA GTT ACA CAA GG-3′ (SEQ ID NO: 646). About 500 nanograms of genomic DNA was amplified by subjecting it to twenty-five PCR amplification cycles as follows: 94° C. for one minute, 56° C. for one minute, and 72° C. 1 minute. Amplification products were separated on a 6% (w/v) polyacrylamide gel using an automated laser fluorescent DNA sequencer (ALF™, Pharmacia LKB Biotechnology Inc., Piscataway, N.J.). A nucleotide having a length of 430 base pairs, designated allele A1, a nucleotide having a length of 434 base pairs, designated allele A2, or both, were identified by comparison to nucleotide size markers of known length.
[0264] Defining the Consensus Sequences of the 5′-Portion of the COL1A1 Gene
[0265] Identification of the consensus sequence of the 3′-portion of the COL1A1 gene has been described elsewhere herein.
[0266] To identify a consensus sequence of the 5′-portion of the COL1A1 gene, genomic DNA obtained from eight unrelated subjects afflicted with OI was amplified by PCR using primers based on published sequences of the COL1A1 gene (Chu et al., 1985, J. Biol. Chem. 260:2315-2320; D'Alessio et al., 1988, Gene 67:105-115; Määttä et al., 1991, FEBS Lett. 279:9-13; Westerhausen et al., 1991, Matrix 11:375-379). PCR reactions were carried out using a commercial DNA polymerase (Amplitaq Gold, Perkin Elmer Cetus, Norwalk, Conn.) in a 40 microliter reaction volume. The PCR was maintained at 95° C. for ten minutes, and then was subjected to thirty-five cycles as follows: 95° C. for forty seconds, 60° C. for forty seconds, and 72° C. for fifty seconds. Amplification products had sizes ranging from about one thousand to about twenty-five hundred nucleotides.
[0267] Nucleotide sequences of amplification products were defined by automated sequencing (ABI PRISM 377™ Sequencer, Perkin Elmer Cetus, Norwalk, Conn.; ABI PRISM™ Dye Therminator Cycle Sequencing Ready Kit with AmpliTaq™ DNA polymerase, Perkin Elmer Cetus, Norwalk, Conn.). Prior to sequencing, samples were treated with exonuclease I (Amersham, Arlington Heights, Ill.) to degrade residual PCR primers and shrimp alkaline phosphatase (Amersham, Arlington Heights, Ill.) to dephosphorylate residual nucleotides, as described (Hanke et al., 1994, BioTechniques 17:858-860; Werle et al., 1994, Nucl. Acids Res. 22:4354-4355).
[0268] Sequencing of the Human COL1A2 Gene
[0269] To sequence the 5′-end of the human COL1A2 gene, a fragment having a length of about fourteen kilobases spanning intron 1 to intron 21 was obtained from an EcoRI/EcoRI genomic fragment that had been cloned into a bacteriophage vector in the course of defining a mutation that caused OI, as described (Vasan et al., 1991, Am. J. Hum. Genet. 48:305-317). The fourteen kilobase fragment was broken at random positions by sonication, and the resulting fragments were separated by gel electrophoresis. Selected fragments were subcloned into a plasmid (pUC18), and sixty clones, each comprising a fragment, were isolated from Escherichia colihost cells. Plasmid DNA was isolated from thirteen color-selected colonies. The plasmids each comprised an insert having a size between about 500 and about 4,000 base pairs. About 90% of the sequence of the fourteen kilobase COL1A2 gene fragment was determined by shotgun sequencing of the inserts from the thirteen selected clones. The remainder of the sequence was determined by sequencing the original bacteriophage clone containing the fourteen kilobase genomic fragment using manual or automated procedures described herein and well known in the art.
[0270] To identify the sequence of the portions of the COL1A2 gene not represented in the fourteen kilobase bacteriophage fragment, a genomic P1 clone comprising the complete human COL1A2 gene was obtained by PCR screening of a human P1 library (Genome Systems, Inc., St. Louis, Mo.). PCR screening was performed using primers having sequences based on the reported sequence of human COL1A2 cDNA (de Wet et al., 1987, J. Biol. Chem. 262:16032-16036; Kuivaniemi et al., 1988, Biochem. J. 252:633-640). PCR primers used were C1PF1, which had the sequence 5′-GTA CAT TTC CTA GAG AAC TTG-3′ (SEQ ID NO: 647), and C1PR1, which had the sequence 5′-CTA CTC TCA GCC CAG GAG GTC CTG-3′ (SEQ ID NO: 648). The sequence of C1PF1 corresponded to sequences in intron 19 of COL1A2, and the sequence of C1PR1 corresponded to sequences in exon 21 and intron 21. Three P1 clones were identified as comprising at least a portion of the coding sequence of COL1A2: P1 clone DMPC-HFF#1 1250-E2, GS control #7403, P1 clone DMPC-HFF#1 1365-B1, GS control #7404, and P1 clone DMPC-HFF#1 1473-F6, GS control #7405. Because P1 clone #7407 was determined to comprise the entire coding sequences of the human COL1A2 gene, this clone was selected for detailed characterization of the gene.
[0271] To increase the yield of DNA, P1 clone #7407 was transferred from E. colistrain NS3529 to E. colistrain NS3516 via transduction. P1 plasmid DNA was isolated as described by Birnboim et al. (1979, Nucl. Acids Res. 7:1513-1523). Isolated P1 plasmid DNA was dissolved in water, and was further purified by spot dialysis using water separated from the DNA by a membrane (VSWP 02500; Millipore, Bedford, Mass.).
[0272] Nucleotide sequencing was carried out by cycle sequencing of P1 plasmid DNA using the dsDNA Cycle Sequencing System obtained from Life Technologies, Inc. (Gaithersburg, Md.) or the Cycle Sequencing Kit obtained from Pharmacia LKB Biotechnology Inc. (Piscataway, N.J.). Sequences of primers used for sequencing the COL1A2 gene were based on published COL1A2 sequences (Bernard et al., 1983, Biochem. 22:1139-1145; de Wet et al., 1987, J. Biol. Chem. 262:16032-16036; Kuivaniemi et al., 1988, Biochem. J. 252:633-640). Additional primer sequences were based on the sequences determined during the work described herein. The 5′-end of each primer was labeled using T4 polynucleotide kinase (U.S. Biochemical Corp., Cleveland, Ohio) and [γ-33P]ATP (Dupont NEN, Wilmington, Del.). In cycle sequencing experiments, 0.5-2.0 micrograms of P1 plasmid DNA was used as the template, and thermal cycling was carried out using a commercial instrument (either a GeneAmp 9600™, Perkin Elmer Cetus, Norwalk, Conn., or a PTC 225 DNA Engine Tetrad™, MJ-Research, Inc., Watertown, Mass.). Some nucleotide sequences were obtained using the Terminator Cycle Sequencing Ready Reaction kit and an ABI PRISM 377™ DNA Sequencer (both obtained from Perkin Elmer Cetus, Norwalk, Conn.). Sequence data were analyzed using the Wisconsin Sequence Analysis Package, versions 8.0 and 8.1 for UNIX (Genetics Computer Group, Madison, Wis.) and the Editseq program of Lasergene software package (DNAStar, Inc., Madison, Wis.).
[0273] PCR/CSGE Mutation Analysis
[0274] Genomic DNA was extracted from blood samples or from cultured skin fibroblasts obtained from human subjects. The exons and the flanking sequences corresponding to the 51 exons of the COL1A1 gene and to the 52 exons of the COL1A2 gene were amplified using the primers depicted in FIGS. 21 and 22 [SEQ ID NOS. 216-324 and 325-434, respectively]. Genomic DNA was amplified by performing PCR in a 40 microliter reaction volume. The PCR was maintained at 95° C. for ten minutes, and then was subjected to thirty-five cycles as follows: 95° C. for forty seconds, 60° C. for forty seconds, and 72° C. for fifty seconds.
[0275] PCR amplification was followed by a heteroduplex formation step, wherein the products of the amplification were maintained at 95° C. for 5 minutes and then at 68° C. for 30 minutes. Samples containing heteroduplexes were identified by CSGE analysis, as described, except that the taurine buffer was not autoclaved (Ganguly et al., 1993, Proc. Natl. Acad. Sci. USA 90:10325-10329; Ganguly et al., 1995, Electrophoresis 16:1830-1835).
[0276] Samples containing heteroduplexes were treated with exonuclease I and shrimp alkaline phosphatase, as described (Hanke et al., 1994, BioTechniques 17:858-860; Werle et al., 1994, Nucl. Acids Res. 22:4354-4355), and were then analyzed either by PCR product sequencing using the Sequenase PCR Product Sequencing Kit (U.S. Biochemical Corp., Cleveland, Ohio) according to the manufacturer's instructions or by automated sequencing (ABI PRISM™ 377 Sequencer, and ABI PRISM™ Dye Therminator Cycle Sequencing Ready Kit with AmpliTaq™ DNA polymerase FS, both obtained from Perkin Elmer Cetus, Norwalk, Conn.). Alleles of PCR products that contained deletions in one allele were separated on and subsequently purified from agarose gels (QIAEX II™ Gel Extraction Kit, Qiagen, Chatsworth, Calif.), and about 60 nanograms of purified PCR product was cloned into plasmid pT7 (pT7 Blue T-Vector Kit, Novagen, Madison, Wis.) prior to sequencing.
[0277] The results of the experiments performed in this Example are now described.
[0278] Detection of COL1A1 and COL1A2 Mutations by Protein and mRNA Analysis
[0279] Fibroblasts obtained from ten subjects afflicted with Type I OI were examined to determine the ratio of newly synthesized chains of Type I and Type III procollagen. The fibroblasts were also examined for the presence of polymorphic COL1A1-derived mRNA, which reflects expression of two different alleles of the COL1A1 gene. As indicated in Table 2, a reduced ratio of newly synthesized Type I procollagen to newly synthesized Type III procollagen was detected in fibroblasts obtained from seven of the ten subjects tested and a slight reduction was detected in an eighth subject. This ratio was normal in fibroblasts obtained from two of the subjects.
[0280] Fibroblasts obtained from the subjects were also examined to determine the presence or absence of a polymorphic base in the mRNA which indicated expression of COL1A1 alleles A1 and A2. One allele either was not expressed or was expressed at a reduced level in fibroblasts obtained from five subjects (subjects 1-4 and 9in Table 2, including one in whom the collagen ratio was normal. The mRNA assay indicated that both COL1A1 alleles were expressed in one subject who was identified by the protein assay as expressing a decreased ratio of Type I to Type III collagen (i.e. subject 8 in Table 2). The mRNA assay results in cells obtained from four subjects were non-informative.
[0281] Examination of procollagen protein and mRNA encoding procollagen in cultured skin fibroblasts have been useful in the prior art in defining mutations that cause OI. However, as illustrated herein, these assays did not detect all COL1A1 and COL1A2 mutations, particularly in Type I OI. Furthermore, examination of protein and mRNA requires the use of skin biopsies from which fibroblasts are cultured for three weeks or more. Thus, examination of protein and mRNA is less convenient than is an assay based on genomic DNA, which DNA can easily be obtained from numerous physiological sources, including a blood sample. For these reasons, a test was developed which included PCR amplification of each of the 103 exons, the flanking sequences thereof, and the 5′- and 3′-untranslated regions of the COL1A1 and COL1A2 genes for the purpose of identifying genetic abnormalities therein.
[0282] Sequences of the COL1A1 and COL1A2 Genes
[0283] The nucleotide sequences of the COL1A1 and COL1A2 genes were determined, so that intronic PCR primers could be designed for analysis of the genes. Several groups had reported the nucleotide sequence of the coding sequence, introns 1-26, and the 5′-portion of intron 27 of the COL1A1 gene, as described herein. However, attempts to make intronic PCR primers based on published sequences for the 5′-half of the COL1A1 gene were largely unsuccessful, apparently because the published sequences did not represent the most frequent alleles of the gene. For this reason, about twelve kilobases of the sequence of the COL1A1 gene were sequenced in the present study, using samples derived from the genomic DNA of eight subjects. The consensus sequence for the complete eighteen kilobase COL1A1 gene has been submitted to GenBank (accession number AF017178), and is included herein at FIG. 4 (SEQ ID NO: 1). The genomic structure of COL1A1 is depicted herein at FIG. 1.
[0284] More than 250 differences were detected between the nucleotide sequences of the introns of the consensus COL1A1 gene and the nucleotide sequences reported for introns 1-27 thereof. It is thus apparent that the COL1A1 intron sequences reported herein and the sequences of primers derived therefrom have not been previously described.
[0285] As described herein, the complete cDNA sequence of COL1A2 and portions of the non-coding region thereof have been reported in the prior art. The complete nucleotide sequence of COL1A2 reported herein includes about 30 kb of sequence which has not previously been reported. About 11 kb of new sequences from the 5′-end of the gene were obtained using a 14 kb EcoRI genomic fragment prepared for other purposes (Vasan et al., 1991, Am. J. Hum. Genet. 48:305-317). An additional 19 kb of new sequences of the COL1A2 gene were obtained using the P1 clone #7407. The complete sequence of the COL1A2 gene has been submitted to GenBank (accession number AF004877), and is described herein in FIG. 5 (SEQ ID NO: 2). The genomic structure of COL1A2 is depicted herein in FIG. 2.
[0286] As depicted in FIG. 6, the lengths of the human COL1A2 introns were about the same as the lengths of the chicken COL1A2 introns (Boedtker et al., 1985, Ann. N.Y. Acad. Sci. 460:85-116). There were major differences in the lengths of human COL2A1 introns and chicken COL2A1 introns (Ala-Kokko et al., 1995, Biochem. J. 308:923-929).
[0287] Primers for PCR Amplification of the Exons and Flanking Regions of COL1A1 and COL1A2
[0288] The sequences of the human COL1A1 and COL1A2 genes described herein were used to design sequences for pairs of nucleotide primer useful for amplification of each of the 103 exons and at least 80 bp of both the 5′- and 3′-flanking sequences of the two genes. The non-coding regions of the COL1A1 gene are depicted herein in FIG. 4 (SEQ ID NOs: 6-57), and the non-coding regions of the COL1A2 gene are depicted herein in FIG. 5 (SEQ ID NOs: 58-110).
[0289] Conditions for PCR amplification were optimized so that each PCR amplification product could be detected as a discrete band by agarose gel electrophoresis and by polyacrylamide gel electrophoresis. In addition, the primers were designed so each of the amplification products made using the primer pairs generally had a length no greater than about 500 base pairs, so that the presence of a single-base substitution could be detected by heteroduplex analysis of an amplification product using CSGE (Ganguly et al., 1995, Electrophoresis 16:1830-1835). Thus, these primer pairs are useful for analyzing not only the coding regions of the COL1A1 and COL1A2 genes, but also for analyzing the intron-exon boundaries and the 5′- and 3′-ends of the genes for mutations associated with pathological conditions. The primers which were designed for use with the COL1A1 gene are depicted in FIG. 21 (SEQ ID NOs: 216-324), and the primers which were designed for use with the COL1A2 gene are depicted in FIG. 22 (SEQ ID NOs: 325-434). One skilled in the art would appreciate that other intronic primers can be made, based on the non-coding region nucleotide sequences described herein, which are useful for PCR/CSGE analysis of at lease a portion of the COL1A1 gene or the COL1A2 gene.
[0290] PCR/CSGE Detection of COL1A1 and COL1A2 Mutations in Human Subjects Afflicted with Type I OI
[0291] The PCR/CSGE Mutation Analysis protocol described herein was used to identify mutations in fifteen subjects, including the ten subjects from whom fibroblast samples were obtained. Twenty-five polymorphisms in the COL1A1 gene and eighteen polymorphisms in the COL1A2 gene were detected. Most of the polymorphisms were found within an intron sequence of one of the two genes, and thus could not have been detected using primers corresponding to cDNA sequences derived from the genes. In addition, the presence of unique heteroduplexes was detected in each of the fifteen subjects, as depicted in FIG. 7. Nucleotide sequence analysis of each of the PCR amplification products which comprised a unique heteroduplex identified thirteen mutations. Five of the mutations were single base deletions or insertions (Subjects 4, 6, 7, 9 and 11 in Table 2), one mutation was a two-base-pair insertion (Subject 13), and three of the mutations were single base substitutions that converted a codon encoding Arg to a premature termination codon (Subjects 5, 10 and 12). Four mutations altered consensus sites of RNA splicing in the first two or last two bases of an intron (Probands 3, 8, 14 and 15). Thus, in order to screen type I and type IX collagen genes for disease/disorder-causing mutations, it is important to examine both coding regions and potential RNA splicing sites. Consensus RNA splicing sites have been described in the scientific literature (see, e.g., Shapiro et al., 1987, Nucl. Acids Res. 15:7155-7174; Jacob et al., 1989, Nucl. Acids Res. 17:2159-2180). Two alterations identified in the two other subjects were also likely to alter RNA splicing: one was at position +3 of intron 22 (subject 1); the second was at position −12 of intron 20 (subject 2). Intronic nucleotides are numbered such that the nucleotide which is located in the intron adjacent the 5′-end of an exon on the coding strand and which is located immediately adjacent the 5′-end of the exon is numbered “−1,” the nucleotide located immediately adjacent the −1 nucleotide in the 5′-direction is numbered “−2,” and so on. Similarly, the nucleotide which is located in the intron adjacent the 3′-end of an exon on the coding strand and which is located immediately adjacent the 3′-end of the exon is numbered “+1,” the nucleotide located immediately adjacent the +1 nucleotide in the 3′-direction is numbered “+2,” and so on. Although the nucleotides at these two positions are not conserved in all introns (Nakai et al., 1994, Gene 141:171-177), these two mutations were not found in one hundred other COL1A1 alleles. Furthermore, no other nucleotide sequence alterations were identified following amplification of all other regions of the COL1A1 and COL1A2 genes of these two subjects. Protein and mRNA data indicated that each of these two subjects expressed one COL1A1 allele at an abnormally low level. Two other observations supported the conclusion that the single base change at −12 of intron 20 in subject 2 was a disease-causing mutation. The same change was found in the affected father and sister of subject 2, but not in unaffected brother of subject 2. Also, the single base change at position-12 of intron 20 produced a 3′-consensus sequence for RNA splicing, i.e. it converted the nucleotide sequence 5′-(Pyrimidine)11-CGG-3′ to the sequence 5′-(Pyrimidine)11-CAG-3′, wherein (Pyrimidine)11means a sequence comprising eleven pyrimidine residues. The results therefore establish that definitive disease-causing mutations have been identified in thirteen subjects afflicted with Type I OI and probable disease-causing mutations have been identified in the other two. As indicated in Table 2, all of these mutations were in the COL1A1 gene.
[0292] Identification of Common Sequences for Mutations Causing COL1A1 Null Alleles
[0293] Comparison of COL1A1 mutations identified herein with previously defined mutations in OI-afflicted subjects identified common sequences for null allele mutations in the COL1A1 gene (Willing et al., 1994, Am. J. Hum. Genet. 55:638-647; Willing et al., 1996, Am. J. Hum. Genet. 59:799-809; Redford-Badwal et al., 1996, J. Clin. Invest. 97:1035-1040; Körkkö et al., 1997, Hum. Mutat. 9:148-156).
[0294] Single base substitutions that converted a CGA codon encoding arginine to a premature TGA termination codon were considered first. There were nine mutations that met this criterion, as indicated in Table 3. All were found in the COL1A1 gene in the sequence context 5′-C/GCC-CGA-GG/T-3′, wherein “C/G” means that the nucleotide at that position may be C or G and “G/T” means that the nucleotide at that position may be G or T. Mutations comprising replacement of a codon encoding Arg with a termination codon were found in five of the six sequences having the indicated context in the wildtype COL1A1 gene. In contrast, no such mutations have been reported in the seven other sequence contexts in which arginine-encoding CGA codons exist in the wildtype COL1A2 gene, i.e. seven sequences in which a CGA codon is not preceded by G/CCC. These data establish that the sequence context 5′-C/GCC-CGA-GG/T-3′ represents a site in the COL1A1 gene that is highly susceptible to mutation. Accordingly, primers, including, but not limited to, primer pairs and intronic primers, which can be used to amplify regions of the COL1A1 gene which comprise this sequence context are especially useful for identifying altered COL1A1 gene sequences.
[0295] Single base deletions or insertions which created frameshifts and premature termination codons in the COL1A1 gene were considered next. A total of fourteen such mutations have been reported in the COL1A1 gene, as indicated in Table 4. Nine of the fourteen mutations in the COL1A1 gene were contained within the sequence 5′-CCC-CCT-3′. Only five of the fourteen mutations were identified in other sequences. In the COL2A1 gene, six such mutations were reported and two are located in the sequence 5′-CCC-CCT-3′ (Brown et al., 1992, Arch. Ophthalmol. 110:1589-1593; Brown et al., 1995, Hum. Molec. Genet. 4:141-142; Ahmad et al., 1995, Arch. Ophthalmol. 113:1454-1457; Ritvaniemie et al., 1993, Genomics 17:218-221). These observations are in accord with the observation of by Jego et al. (1992, Oncogene 8:209-213), that both single base insertions, single base deletions are likely to occur in duplicated sequences. In the case of collagens, the sequence 5′-CCC-CCT-3′, encoding Pro-Pro or Pro-Hyp, is a common sequence in the COL1A1 gene.
[0296] Table 4. Null Allele Mutations Produced by Single Nucleotide Insertions or Deletions. Mutations were identified herein and by Körkkö et al. (1997, Hum. Mutat. 9:148-156), Willing et al. (1996, Am. J. Hum. Genet. 59:799-809), Willing et al. (1994, Am. J. Hum. Genet. 55:638-647), and Redford-Badwal et al. (1996, J. Clin. Invest. 97:1035-1040).
[0297] No mutations in the COL1A2 gene were found by PCR/CSGE analysis of the genomic DNA obtained from the fifteen subjects afflicted with Type I OI. Therefore, mutations in the COL1A2 gene appear to be a rare cause of Type I OI, even though mutations in the gene can cause some severe variants of OI (Kuivaniemi et al., 1997, Hum. Mutat. 9:300-315). Individuals and mice which are heterozygous for a non-functional proα2(I) chain have decreased bone density (Dickson et al., 1984, Proc. Natl. Acad. Sci. USA 81:4524-4528; Chipman et al., 1993, Proc. Natl. Acad. Sci. USA 90:1701-1705; Saban et al., 1996, BioTechniques 21:190-192). Therefore, individuals with non-functional COL1A2 genes may have phenotypes that are milder than Type I OI and that overlap with osteoporosis.
Example 5
PCR/CSGE Analysis of the COL9A1 and COL9A2 Genes Using Intronic Primers
[0298] In this Example, the complete nucleic acid sequences of each of the human COL9A1 and COL9A2 genes are described. Primers useful for PCR/CSGE analysis of the genes are also described. PCR/CSGE screening methods have been described elsewhere herein. Sequences of the non-coding regions of these two genes are reported herein for the first time.
[0299] The materials and methods used in this Example are now described.
[0300] Screening of Genomic Library
[0301] PCR screening of a human genomic P1 library (Genome Systems, Inc., St. Louis, Mo.) was performed using primers based on the reported cDNA sequence of the human α1(IX) collagen chain (Muragaki et al., 1990). Initial PCR screening was performed using oligonucleotide MV-9B3, which has the nucleotide sequence 5′-CTT TCG CTA AGA GAG CCT GTG (SEQ ID NO: 649) and oligonucleotide MV-9R3, which has the nucleotide sequence 5′-GGA CTG AGC ACG CAG CTC TG (SEQ ID NO: 650). One P1 clone was identified which comprised at least a portion of the COL9A1 gene. This clone was designated P1-A and was clone DMC-HFF#1-1378-C8, GS control #5173.
[0302] A second screening involved the use of two primer pairs. The first primer pair comprised oligonucleotide PRIM-1, which has the nucleotide sequence 5′-GTT TCT GTG AGC CAG CCT CCT G (SEQ ID NO: 651), and oligonucleotide PRIM-2, which has the sequence 5′-GGC ACA GTG GCC CAC GAT AAG AC (SEQ ID NO: 652). The second primer pair comprised oligonucleotide PRIM-3, which has the nucleotide sequence 5′-TTT ATA TGA GTA TGA AGC AGG CAC (SEQ ID NO: 653), and oligonucleotide PRIM-4, which has the sequence 5′-TCT TGT TTA CCC TTG TGT ATC TAC (SEQ ID NO: 654). PCR screening of the P1 library described herein using these two primer pairs identified a P1 clone which comprised at least of portion of the COL9A1 gene. This clone was designated P1-B and was P1 clone DMPC-HFF#1-837-D10, GS control #13295.
[0303] To obtain genomic clones comprising the human COL9A2 gene sequence, PCR screening of a human genomic PAC library (Genome Systems, Inc., St. Louis, Mo.) was performed using a primer pair, the sequences of which were based on published sequences of cDNA encoding the human α2(IX) collagen chain (Perälä et al., 1994, J. Biol. Chem. 269:5064-5071). This primer pair comprised oligonucleotide TP-9, which has the nucleotide sequence 5′-GGC AGC CCA GGT ATT CGT GG (SEQ ID NO: 655) and oligonucleotide TP-10, which has the nucleotide sequence 5′-GCC CTG GTG GCC TGG ACT TC (SEQ ID NO: 656). The screening yielded one positive PAC clone, PAC-39-22G, GS control#6111, herein designated PAC-1.
[0304] Characterization of Genomic Clones
[0305] In order to increase the yield of DNA, both P1 clones were transferred from E. colistrain NS3529 to E. colistrain NS3516 via phage transduction. Isolation of P1 plasmid DNA and PAC plasmid DNA was done using the method of Birnboim et al. (1979, Nucl. Acids Res. 7:1513-1523). Isolated DNA was dissolved in water and further purified by spot dialysis using water separated from the DNA by a membrane as described elsewhere herein. The COL9A1 and COL9A2 genes were sequenced using primers based on published cDNA sequences of the genes. Additional sequencing primers were based on sequences identified in the experiments described herein.
[0306] Nucleotide sequencing was performed by cycle sequencing of P1 and PAC clones using the dsDNA Cycle Sequencing System obtained from Life Technologies, Inc (Gaithersburg, Md.) or the Cycle Sequencing Kit obtained from Pharmacia LKB Biotechnology Inc. (Piscataway, N.J.). Labeling was achieved by either labeling the 5′-end of the primer with T4 polynucleotide kinase (U.S. Biochemical Corp., Cleveland, Ohio) and [γ-33P]ATP (Dupont NEN, Wilmington, Del.) or by including [α-35S]dATP (Amersham Life Science, Arlington Heights, Ill.) in the sequencing reaction mixture. For cycle sequencing, 0.5 to 2.0 micrograms of P1 or PAC DNA was used as template and thermal cycling was performed using either a GeneAmp™ 9600 (Perkin Elmer Cetus, Norwalk, Conn.) or PTC-225 DNA Engine Tetrad™ (MJ-Research, Inc., Watertown, Mass.) instrument. Some nucleotide sequences were determined by automated sequencing of P1 or PAC DNA using the Dye Terminator Cycle Sequencing Ready Reaction kit and ABI PRISM™ 377 DNA Sequencer (both obtained from Perkin Elmer Cetus, Norwalk, Conn.).
[0307] Some intronic sequences for the COL9A1 and COL9A2 genes were PCR amplified and cloned into a pUC 18 vector (SureClone Ligation Kit™, Pharmacia LKB Biotechnology Inc., Piscataway, N.J.). Cloned plasmid DNA was isolated using the Wizard Plus Maxiprep DNA Purification System (Promega Corp., Madison, Wis.), and was sequenced using the dideoxynucleotide sequencing method (Sanger et al., 1977, Proc. Natl. Acad. Sci. USA 74:5463-5467) and T7 DNA polymerase (T7 Sequencing Kit, Pharmacia LKB Biotechnology Inc., Piscataway, N.J.).
[0308] The sizes of the longest introns of the COL9A1 gene were determined by PCR amplification of the intron (Expand Long Template PCR System; Boehringer-Mannheim, Indianapolis, Ind.) and by estimation of the size of the intron after agarose gel electrophoresis by comparison to nucleotides of known length. For each intron this was done using at least two forward and two reverse primers in all possible combinations and using the P1 DNA as template. Computerized analysis of all sequences was performed using the Wisconsin Sequence Analysis Package, versions 8.0 and 8.1 for UNIX (Genetics Computer Group, Madison, Wis.).
[0309] 3′-RACE analyses
[0310] To define the 3′-end of the cDNA encoding α1(IX), a RT-PCR was performed. About 0.4 micrograms of total RNA obtained from human fetal cartilage (Baldwin et al., 1989, Biochem. J. 262:521-528) was reverse transcribed using the GeneAmp™ RNA PCR Kit obtained from Perkin Elmer Cetus (Norwalk, Conn.) and an oligo(dT) primer linked to a random sequence, namely 5′-GAC TGA TCA GCG AAT TCT ACG TCG C(T20) (SEQ ID NO: 657). Single-stranded cDNA was amplified by two sequential PCRs using nested forward primers designed to hybridize to the 3′-end of the cDNA, namely a primer designated RACE-1, which had the sequence 5′-CAG GGC TGG CAG GAA TTC CTG (SEQ ID NO: 658) and a primer designated RACE-2, which had the sequence 5′-AAT TCC TGG AGT GCC TGG AC (SEQ ID NO: 659). The reverse primer used with each of RACE-1 and RACE-2 had a sequence identical to the random sequence at the 5′-end of the oligo(dT) primer. The first PCR was performed in a volume of 25 microliters using 10 picomoles of the reverse primer and 10 picomoles of RACE-1. The PCR was maintained at 95° C. for ninety seconds, and was then subjected to forty cycles as follows: 95° C. for thirty seconds, 62° C. for thirty seconds, and 72° C. for thirty seconds. Two microliters of the reaction mixture was used as template for the second PCR. The reaction volume for the second PCR was 50 microliters, and 20 picomoles of the reverse primer and 20 picomoles of RACE-2 were used. The second PCR was subjected to the same reaction conditions as the first PCR, except that at the annealing temperature was 60° C. instead of 62° C., and that thirty cycles were performed instead of forty.
[0311] Agarose gel electrophoretic separation of the products of the second PCR demonstrated the presence of a product having a length of 380 base pairs. Cloning and sequencing of this RT-PCR product revealed a 229-base-pair stretch of 3′-untranslated sequence, followed by a stretch of adenine residues. No polyadenylation signal was detected. Because the signal was not detected, a second RT-PCR was performed as before using a new oligonucleotide, RACE-3, which had the sequence 5′-GGA AGA CAG CAG AGT CAT CAG (SEQ ID NO: 660). The sequence of RACE-3 was based on the sequence located 20 nucleotides in the 5′-direction relative to the stretch of adenine residues.
[0312] PCR was performed in a volume of 25 microliters using 10 picomoles of the reverse primer described herein and 10 picomoles of RACE-3. The PCR was maintained at 95° C. for ninety seconds, and was then subjected to forty cycles as follows: 95° C. for thirty seconds, 60° C. for thirty seconds, and 72° C. for ninety seconds. In addition to the expected 50-base-pair-length product derived by amplification of the intragenic stretch of adenine residues, another product having a length of about 200 nucleotides was detected. This RT-PCR product was cloned and the sequence was determined.
[0313] Cloning and Sequencing of the Murine COL9A 1 Gene
[0314] Genomic DNA was isolated from murine liver, as described (Ausubel et al., 1989, In: Current Protocols in Molecular Biology, Wiley, New York, Vol. 1, pp. 2.2.1-2.2.3) and 200 nanograms of this DNA was used as a template for PCR amplification of the sequences of exon 6, the alternative promoter, and exon 7 of the murine COL9A1 gene. The PCR was performed in a reaction volume of 50 microliters using 10 picomoles of primer M6F, which had the nucleotide sequence 5′-AGT TTG AAC TCC AGT GGG TGC (SEQ ID NO: 661), and 10 picomoles of M7R, which had the nucleotide sequence 5′-ACC TCA TCA GTG GTC TGG CTG (SEQ ID NO: 662). The sequences of primers M6F and M7R were based on published sequences of exons 6 and 7, respectively, of the murine COL9A1 gene (Muragaki et al., 1990, Eur. J. Biochem. 192:703-708). The PCR was maintained at 94.5° C. for ninety seconds, and was then subjected to 33 cycles as follows: 94.5° C. for thirty seconds, 60° C. for thirty seconds, and 72° C. for seventy-five seconds. The PCR product was cloned into a pUC18 vector and sequenced. The sequence for this portion of the murine COL9A 1 gene has been described in the literature.
[0315] The results obtained in the experiments of this Example are now described.
[0316] Characterization of Genomic Clones
[0317] PCR screening of a human P1 library for clones comprising at least a portion of the human COL9A1 gene sequence yielded two positive clones, P1-A and P1-B, which are schematically depicted in FIG. 8A. Clone P1-A comprised the 5′-end of the gene and extended in the 3′-direction therefrom beyond intron 36. Clone P1-B comprised the 3′-end of the gene and extended in the 5′-direction therefrom beyond intron 18. PCR screening of a human PAC library for clones comprising at least a portion of the human COL9A2 gene sequence identified one positive clone, PAC-1, which is schematically depicted in FIG. 8A. Clone PAC-1 comprised the entire coding sequence for the gene.
[0318] Genomic Structure and Domain Organization of the Human COL9A1 and COL9A2 Genes
[0319] The exon boundaries of the COL9A1 and COL9A2 genes were defined by nucleotide sequencing using primers based on corresponding cDNA sequences. The sizes of several large introns of the COL9A1 gene were also defined by PCR amplification and subsequent agarose gel electrophoresis. It was determined that the COL9A1 gene spans about 90 kb and comprises thirty-eight exons. The complete genomic organization and sequence were defined for the COL9A2 gene. It was determined that the gene is considerably more compact than the COL9A1 gene, in that it spans only about 15 kb even though it comprises thirty-two exons. The relative sizes of the introns and exons of the COL9A1 and COL9A2 genes are diagrammatically listed in FIG. 8A.
[0320] Because COL9A1 and COL9A2 encode the α1(IX) and α2(IX) chains of Type IX collagen, it was not surprising that the exon organization of these genes was identical except for the exons coding for NC4 domain, as schematically listed in FIG. 8B. The NC4 domain comprises the cartilage-specific domain of the α1(IX) collagen chain. Exon 16 of the COL9A1 gene is 33 base pairs in length, whereas the corresponding exon 10 of the COL9A2 gene is 48 base pairs in length. This difference of 15 base pairs explains the presence of five additional amino acids in the NC3 domain of the proα1(XI) chain, relative to the NC3 domain of the proα2(IX) chain.
[0321] The intron sizes of the two genes are markedly different, as schematically listed in FIG. 8A. The longest intron in the COL9A2 gene is intron 16 and has a length of 1,690 base pairs. Also, only five introns of COL9A2 are over 1,000 base pairs in length. In contrast, there are nineteen introns that are over 1,000 base pairs in length in the COL9A1 gene. Furthermore, three of the introns of the COL9A1 gene are about 10,000 base pairs in length. The largest introns of the two genes are not located in the 5′-end of the genes, as they are in the genes encoding fibrillar collagen (Ala-Kokko et al., 1995, Biochem. J. 308:923-929).
[0322] Comparison of the Human COL9A1 Gene with the Chicken, Murine, and Rat COL9A1 Genes
[0323] The gene encoding the α1(IX) chain in chicken is about 100 kilobases in length (Ninomiya et al., 1990, In Extracellular Matrix Genes, Sandell et al., eds., Academic press, San Diego, pp. 79-114), and is similar in size to the human gene. However, the chicken gene has been only partially characterized, in that the sizes of only the first seventeen and the last two exons are known (Lozano et al., 1985, Proc. Natl. Acad. Sci. USA 82:4050-4054; Vasios et al., 1988, J. Biol. Chem. 263:2324-2329; Ninomiya et al., 1990, In Extracellular Matrix Genes, Sandell et al., eds., Academic press, San Diego, pp. 79-114). The sizes of the defined exons are the same in the chicken and human genes with two exceptions. Exon 6 is 78 base pairs in length in the chicken gene and 84 base pairs in length in the human gene (Ninomiya et al., 1990, In Extracellular Matrix Genes, Sandell et al., eds., Academic press, San Diego, pp. 79-114). The last (i.e. 3′-most) exon is about 1,000 base pairs in length in the chicken gene and about 550 base pairs in length in the human gene because of a difference in the lengths of the 3′-untranslated regions thereof.
[0324] The alternative transcript of the COL9A1 gene is initiated in intron 6 thereof, as depicted in FIG. 14. The size of intron 6 is similar in the chicken and human genes, 663 and 820 base pairs, respectively, as indicated in Table 5.
[0325] To extend the comparison of COL9A1 gene sequences, intron 6 of the mouse gene was sequenced. The size of intron 6 in the mouse gene was 799 base pairs, and was similar to the intron size in the human and chicken genes. The nucleotide sequences from the 5′-end of intron 6 to the start of translation of the alternative exon 1* were compared in the human, chicken, and mouse genes. The sizes of these sequences were 685, 549, and 612 base pairs, respectively, in the human, chicken, and murine genes, as depicted in FIG. 15 and in Table 5). In addition to conserved size, the human, chicken, and murine sequences were also conserved, in that there was 72% sequence identity between the human and murine sequences, 65% sequence identity between the human and the chicken sequences, and 65% sequence identity between the mouse and the chick sequences. Because the cornea-specific transcription start site of the chicken gene is located 13 to 20 nucleotides downstream of a TAATAA box and 41 nucleotides downstream of a CCAAT box (Nishimura et al., 1989, J. Biol. Chem. 264:20033-20041), the analogous regions were analyzed for the presence of these sequences in the human and murine genes. The TAATAA and CCAAT sequences and their locations were conserved among the chicken, human, and murine genes, as depicted in FIG. 15. Thus, a COL9A1 cornea-specific promoter was identified in the human gene. The alternative start site in exon 1* of the human gene was similar to the alternative start site in exon 1* of the chicken gene.
[0326] Short sequences separated the 3′-end of exon 1 from the 5′-end of exon 7 in the COL9A1 genes of the various species. The length of this separation was 62 base pairs in the human gene, 48 base pairs in the chicken gene, 58 base pairs in the murine gene, and 57 base pairs in the rat gene. Among these genes, the distance separating the 3′ end of exon 1* and the 5′-end of exon 7 was consistently less than 70 bp, the minimum size observed for any intron sequence (Fu et al., 1988, Mol. Cell. Biol. 8:3582-3590). The short size of this nucleotide sequence probably explains why splicing of RNA transcripts that begin in exon 1* occurs between the 3′ end of exon 1* and the 5′-end of exon 8 in the cornea.
[0327] Although the human COL9A1 and COL9A2 genes encode similar chains of a single protein, they differ in size by a factor of about six. The difference in size is accounted for by much larger introns in the 90 kilobase COL9A1 gene. The comparison of the mouse and human COL9A2 genes demonstrated that the intron sizes are conserved to a high degree in the genes of the two species. Three introns that are larger in the human sequence are probably larger because of Alu sequence insertions. However, intron 16 in the mouse gene is twice the size of the corresponding human intron, even though the human intron comprises an Alu sequence.
[0328] In addition to providing information on conservation of gene structures, the results described herein provide a basis for identifying mutations in the COL9A1 and COL9A2 genes that are associated with human disorders including, but not limited to, multiple epiphyseal dysplasia (MED), osteoarthritis, and early onset osteoarthritis. Recent results demonstrate that mutations in the COL9A2 gene are associated with MED. Also, experiments in transgenic mice demonstrate that mutations in the COL9A1 gene can produce cartilage degeneration similar to osteoarthritis.
Example 6
PCR/CSGE Analysis of the COL9A3 Gene Using Intronic Primers
[0329] In this Example, the complete nucleic acid sequence of the human COL9A3 gene is described. Primers useful for PCR/CSGE analysis of the genes are also described. PCR/CSGE screening methods have been described elsewhere herein. Sequences of the non-coding regions of COL9A3 are reported herein for the first time. Two unrelated families having different nine base pair deletions in the same region of the COL1 domain of the COL9A3 gene are also described, wherein the deletions comprise neutral variants of the COL9A3 gene.
[0330] The materials and methods used in this Example are now described.
[0331] Isolation of Phage and P1 Clones for the Human COL9A3 Gene
[0332] Sequences of the human COL9A3 gene in phages and plasmids of human genomic libraries were identified by PCR amplification using primer pairs having sequences based on published cDNA sequences encoding human α3(IX) and on murine COL9A2 gene sequences (Brewton et al., 1995, Genomics 30:329-336; Perälä et al., 1994, J. Biol. Chem. 269:5064-5071). The COL9A3 gene was amplified using a primer pair comprising an oligonucleotide designated C93-F4, which has the nucleotide sequence 5′-CAG GAA AGC CGG GGA AAC CAG (SEQ ID NO: 663), corresponding the sequence of nucleotides 200 to 220 numbered in the 3′-direction relative to the translation start site identified in the human cDNA, and a second oligonucleotide designated C93-R5, which has the nucleotide sequence 5′-GTC CAT CTC GTC CAG TCA GAC (SEQ ID NO: 664), corresponding to the sequence of nucleotides 277 to 257 numbered in the 3′-direction relative to the translation start site. Sequences at the 3′-end of the COL9A3 gene were amplified using a pair of primers comprising an oligonucleotide designated C93-F32, which has the nucleotide sequence 5′-CCT GCC AAG GAG CCG TGT TAG G (SEQ ID NO: 665), corresponding to nucleotides 1997 to 2018 numbered in the 3′-direction relative to the translation start site and a second oligonucleotide designated C93-RUTR, which has the nucleotide sequence 5′-CCT TTT GAG GTA TGC TGT CAG GC (SEQ ID NO: 666), corresponding to nucleotides 2249 to 2227 numbered in the 3′-direction relative to the translation start site. The nucleotide sequences of primers C93-F4 and C93-R5 corresponded to sequences in exons 4 and 5, respectively, of cDNA derived from the human COL9A3 gene. The nucleotide sequences of primers C93-F32 and C93-RUTR corresponded to sequences in exon 32 of the murine COL9A2 gene.
[0333] PCR amplification was performed in a 40 microliter reaction volume comprising between about 50 and 100 nanograms of genomic DNA, 0.25 mM of each of the two primers, 200 micromolar of each dNTP, 1.5 millimolar MgCl2, and 1 unit of Taq polymerase (AmpliTaq™, Perkin Elmer Cetus, Norwalk, Conn.). The PCR was subjected to thirty cycles as follows: 94° C. for one minute, 60° C. for one minute, and 72° C. for one minute. Amplification products exhibited single bands following agarose gel electrophoretic separation thereof. The amplification product made using primers C93-F4 and C93-R5 had a length of about 700 base pairs, and the amplification product made using primers C93-32F and C93-RUTR had a length of about 250 base pairs. These two amplification products were used for screening of human P1 library (Genome Systems, Inc.) to identify clones comprising at least a portion of the human COL9A3 gene. Screening of the library resulted in identification of three positive P1 clones designated P1-C93A, P1-C93B and, P1-C93C (having GS control numbers 12269, 12270, and 12271 and clone addresses DMPC-HFF#1-270-C3, DMPC-HFF#1-753-B10 and DMPC-HFF#1-1082-B5, respectively).
[0334] Characterization of P1 Clones
[0335] P1 DNA was isolated by culturing bacterial cells comprising a P1 clone overnight in 3 milliliters of Luria Broth (LB) comprising 25 μg/ml kanamycin. A 2.5 milliliter aliquot of the overnight culture was used to inoculate 75 milliliters of LB. After ninety minutes of incubation, isopropyl β-D-thiogalactopyranoside (IPTG) was added to a final concentration of 0.5 mM and the cells were incubated for an additional five hours. Cells were harvested by centrifugation of 10 milliliter aliquots of culture medium at 10,000×g for ten minutes. P1 DNA was isolated from harvested cells using a standard plasmid isolation protocol (Birnboim et al., 1979, Nucl. Acids Res. 7:1513-1523).
[0336] P1 clone DNA was sequenced by a cycle sequencing method using a Cycle Sequencing Kit supplied by Pharmacia LKB Biotechnology Inc. (Piscataway, N.J.). Sequences of the primers used for the sequencing method were based on published cDNA nucleotide sequences encoding human α3(IX) protein (Brewton et al., 1995, Genomics 30:329-336) or the nucleotide sequence of the murine COL9A2 gene (Perälä et al., 1994, J. Biol. Chem. 269:5064-5071). The introns between exons 10 and 11, between exons 12 and 13, between exons 26 and 27, and between exons 31 and 32 were amplified using an Expand Long Template PCR System kit supplied by Boehringer Mannheim (Indianapolis, Ind.). PCR products were purified using a commercial agarose gel extraction kit (QIAEX™ II Gel Extraction Kit, Qiagen, Chatsworth, Calif.), cloned into pUC18 vectors using the SureClone™ Ligation Kit supplied by Pharmacia LKB Biotechnology Inc. (Piscataway, N.J.), and sequenced using a T7 Sequencing Kit supplied by Pharmacia LKB Biotechnology Inc. (Piscataway, N.J.). Oligonucleotides produced during the sequencing reactions were separated using a 6% (w/v) polyacrylamide gel.
[0337] PCR/CSGE Mutation Analysis
[0338] PCR primers having nucleotide sequences derived from the nucleotide sequences of the introns of the human COL9A3 gene were used to amplify each exon, the 5′-untranslated region, and the 3′-untranslated region thereof. Primers were selected such that the length of each of the amplified sequences was between about 200 and about 400 base pairs. Each amplified sequence included at least 80 intron-encoded nucleotides, including a segment derived from the intron located at the 5′-end and a segment derived from the intron located at the 3′-end of the amplified exon sequence. Genomic DNA obtained from individual subjects was PCR-amplified in a 40 microliter reaction volume by subjecting the PCR to thirty cycles as follows: 94° C. for forty-five seconds, 60-62° C. for forty-five seconds, and 72° C. for one minute. Following amplification, the PCR products were maintained at 72° C. for ten minutes, denatured at 95° C. for five minutes, and reannealed at 68° C. for thirty minutes. Using this method, heteroduplexes were formed in PCR mixtures comprising genomic sequences having non-identical alleles of COL9A3.
[0339] PCR amplification products were analyzed by CSGE, as described, except that taurine buffer was not autoclaved (Ganguly et al., 1993, Proc. Natl. Acad. Sci. USA 90:10325-10329). The concentration and quality of amplification products was estimated analyzing 5 microliters of each product by gel electrophoresis using a 1.5% (w/v) agarose gel.
[0340] Each CSGE gel comprised 10% (w/v) polyacrylamide, wherein the ratio of acrylamide (Intermountain Scientific, Kaysville, Utah) to 1,4 bis(acryloyl)piperazine (BAP, Fluka, Ronkonkoma, N.Y.) was 99:1, 10% (v/v) ethylene glycol (Sigma Chemical Company, St. Louis, Mo.), 15% (v/v) formamide (Gibco, Grand Island, N.Y.), 0.1% (w/v) ammonium persulfate (U.S. Biochemical Corp., Cleveland, Ohio), 0.07% (v/v) TEMED (Sigma Chemical Company, St. Louis, Mo.), and 0.5×TTE buffer (which comprised 44 millimolar Tris, 14.5 millimolar Taurine, and 0.1 millimolar EDTA at pH 9.0). Electrophoresis was performed using a standard DNA sequencing apparatus and 0.5×TTE buffer as the electrode buffer. Prior to electrophoresis, between 3 and 15 microliters of an amplification product, comprising between about 25 and about 75 nanograms of DNA, was mixed with loading buffer to achieve final concentrations of 3% (v/v) glycerol, 0.025% (w/v) xylene cyanole FF, and 0.025% (w/v) bromphenol blue. Prior to loading the gel, electrical current was applied to the gel at a rate of 45 Watts for fifteen minutes. After loading, electrical current was applied to the gel at a rate of 45 Watts for five hours, the gel being maintained at room temperature throughout the procedure. Following electrophoresis, the gel was stained with ethidium bromide. Amplification products which were identified as containing heteroduplexes were analyzed by direct PCR sequencing using a T7 Sequenase PCR Product Sequencing Kit supplied by U.S. Biochemical Corp. (Cleveland, Ohio). PCR products were purified from agarose using a commercial kit (QIAEX™ II Gel Extraction Kit, Qiagen, Chatsworth, Calif.). PCR products were sequenced following cloning of about 60 nanograms of the purified product into a pUC18 vector, using the SureClone™ Ligation Kit supplied by Pharmacia LKB Biotechnology Inc. (Piscataway, N.J.). Several clones corresponding to each observed heteroduplex were sequenced to identify sequences for both alleles.
[0341] The results of the experiments presented in this Example are now described.
[0342] Characterization of genomic clones
[0343] Screening of the human P1 library with the two PCR primer pairs designed to amplify the 5′-end and the 3′-end of the human COL9A3 gene identified three P1 clones, each of which comprised sequences corresponding to both ends of the gene. Thus, all three clones comprised the entire coding region of COL9A3. Clone P1-C93A was selected for detailed characterization of the gene. Nucleotide sequencing of the human COL9A3 gene was performed by direct sequencing of clone P1-C93A or by sequencing of plasmids comprising subclones thereof. Over 27 kilobases of nucleotide sequence was identified, and the results indicated that the COL9A3 gene has a length of about 24 kilobases and contains thirty-two exons, as depicted in FIG. 9). The nucleotide sequence of more than 3 kilobases of the untranslated region adjacent the 5′-end of the gene was identified. Restriction endonuclease sites in this region are depicted in FIG. 9.
[0344] Exon Organization and Domain Structure
[0345] The genomic structure of the human COL9A3 gene is depicted in FIGS. 9 and 12. The exon organization of the COL9A3 gene was compared with the exon organization of the murine gene encoding the α2 chain of Type IX collagen (Perälä et al., 1994, J. Biol. Chem. 269:5064-5071). The overall exon organization of these two genes exhibited considerable similarities. There were some unexpected differences in the lengths of the exons encoding the COL3 domain of each of the proteins, even though the size of the domain is identical in these genes. The COL3 domain comprises 137 amino acids in each of the two proteins, and is thus encoded by a sequence comprising 411 nucleotides. This domain is encoded by exons 2 through 10 in both genes. Also, in both genes, exon 2 encodes the junction between the NC4 and COL3 domains of the chain.
[0346] It was discovered that exon 2 of the human COL9A3 gene encodes one fewer collagen Gly-X-Y triplet than exon 2 of the murine COL9A2 gene. Nonetheless, the COL3 domains of the two proteins remain identical, because exon 4 of the human COL9A3 gene encodes one more collagen Gly-X-Y triplet than does exon 4 of the murine COL9A2 gene, as depicted in FIG. 17. In effect, deletion of nine nucleotides encoding one Gly-X-Y triplet from exon 2 of the human COL9A3 gene is compensated for by insertion of nine nucleotides encoding an additional Gly-X-Y triplet in exon 4.
[0347] Non-Coding Regions
[0348] The lengths of the introns of the human COL9A3 gene vary from 84 nucleotides to more than 3,000 nucleotides, as depicted in FIG. 12. Six of these introns are more than 1,000 nucleotides in length. The largest intron, intron 26, has a length of about 3,000 nucleotides, which can be explained in part by the observation that the intron comprises two Alu-repeats. In contrast to the murine COL9A2 gene and each of the genes encoding fibrillar collagen proteins which have been characterized to date, the first intron of the human COL9A3 gene is relatively small, and the large introns are located mostly in the 3′-end of the gene (Ala-Kokko, 1995, Biochem. J. 308:923-929).
[0349] Mutation Screening
[0350] A subject who was afflicted with MED and who was a member of a first family was examined to determine whether the subject's COL9A3 gene comprised an MED-causing mutation. The exons and flanking sequences of the subject's COL9A3 gene were amplified by PCR, and the amplification products were analyzed for the presence of heteroduplexes using CSGE analysis, as described herein.
[0351] Several neutral polymorphisms and one potential disease-causing mutation were identified. The potential disease-causing mutation was located in exon 30. Sequencing of the amplification product corresponding to exon 30 and the flanking sequences thereof indicated the presence of a nine nucleotide deletion in the exon, which caused a deletion in a Gly-Pro-Pro triplet in the 5′-end of the COL1 domain of the α3(IX) chain in the subject's DNA.
[0352] Other members of the first family were examined for the presence of the mutation. As depicted in FIG. 18, the COL9A3 genes of three affected members of the first family exhibited the deletion. However, the mutation was not demonstrated to be the sole or sufficient cause of the MED symptoms in the subjects. The COL9A3 gene of one unaffected member of the first family exhibited the deletion; furthermore, the COL9A3 genes of identical twins that were affected did not have the deletion. Hence, the deletion in the COL9A3 gene and the MED phenotype did not exclusively co-segregate in the first family.
[0353] The nucleotide sequence of the deleted region identified in the first family, which is depicted in FIG. 19 as “Del A1,” indicated that the deletion occurred in a repetitive GC-rich region. Because of the repetitiveness of the sequences, the deletion could have occurred at either of two different sites. The nucleotide sequence of the deleted region in the second family also consisted of nine nucleotides of the COL9A3 gene which encoded a Gly-Pro-Pro triplet in the 5′-end of the COL1 domain. The deleted sequence in the second family was different from the deleted sequence in the first family, as depicted in FIG. 19 as “Del A2.”
[0354] Because the three α chains of the type (IX) collagen triple helix must be of the same length and in register, the observation that different nine nucleotide deletions encoding a Gly-X-Y triplet occurred in two unrelated MED-affected individuals suggested that the deletions were associated with the disease state of the individuals. However, examination of affected and unaffected members of the family indicated there was no co-inheritance of the deletion with the disease phenotype. Therefore, the deletion must be a neutral variant of the COL9A3 gene. In Type IX collagen protein encoded by this variant, the protein apparently retains its normal function, even though each of the two the deletions shorten the COL1 domain of the α3(IX) chain, such that it lacks a Gly-X-Y triplet found in each of the α1(IX) and α2(IX) chains. It appears sufficient for normal Type IX collagen function that the three chains retain a continuous sequence of four or five Gly-X-Hyp or Gly-Pro-Hyp at the ends of the COL1 domain (Fertala et al. 1993, Biochem. J. 289:195-199; Westerhausen et al., 1990, J. Biol. Chem. 265:13995-14000).
[0355] Fibrillar collagens have been used as prototype proteins to study chain association and triple helix formation in collagens. According to the presently-understood model, the carboxyl-terminal propeptides of the three α chains of a collagen molecule associate and form intramolecular disulfide-bonded trimers. Formation of bonded trimers ensures correct registration of the α chains. Following disulfide bond formation, helix formation progresses in the direction from the carboxyl terminus toward the amino terminus of the collagen protein or region.
[0356] It has been demonstrated that the carboxyl terminal propeptides of fibrillar collagens contain all the information necessary for correct chain selection and association (Engel et al., 1991, Annu. Rev. Biophys. Biophys. Chem. 20:137-152). It has also been demonstrated that synthetic peptides comprising the entire NC1 domain and the carboxyl-terminal end of the COL1 domain each of the three α chains of Type IX collagen contain all the information necessary for chain selection and assembly (Mechling et al., 1996, J. Biol. Chem. 271:13781-13785). The correct assembly of collagen α chains is critical for biosynthesis of functional Type IX collagen because formation of the triple helix progresses in a zipper-like fashion. In FACIT collagens, non-triple helical domains such as the NC2 region of Type IX collagen may have a role in aligning Gly-X-Y sequences so that they can assume a triple helical conformation, as depicted in FIG. 20.
[0357] Oligonucleotide probes and primers may be labeled with a detectable label, using any labeling method known in the art, preferably using a radiolabel such as 32P. By way of example, oligonucleotides may be labeled with 32P using standard methods, such as treatment of an oligonucleotide with γ-32P-ATP and T4polynucleotide kinase according to known methods (see, e.g. Sambrook, et al., 1989, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, New York). Further by way of example, non-radiolabeled probes comprising biotinylated nucleotides introduced during oligonucleotide synthesis may also be used. Detection of biotinylated nucleotides may be accomplished by streptavidin and antibody-linked enzymes which generate a color reaction using, for example, the Genius™ system (Boehringer Mannheim Biochemicals, Indianapolis, Ind.).
[0358] Polymerase chain reaction (PCR) methods, as used herein, are described, for example, in U.S. Pat. No. 4,683,195 to Mullis et al., in U.S. Pat. No. 4,683,202 to Mullis, and in Innis et al., ed., 1990, In: PCR Protocols, Academic Press, Inc., San Diego. PCR primer and DNA probe labeling methods are described, for example, in U.S. Pat. No. 4,822,731 to Watson et al.
[0359] The disclosures of each and every patent, patent application, and publication cited herein are hereby incorporated herein by reference in their entirety.
[0360] Methods which were used but not described herein are well known and within the competence of one of ordinary skill in the art of molecular biology.
[0361] While this invention has been disclosed with reference to specific embodiments, it is apparent that other embodiments and variations of this invention may be devised by others skilled in the art without departing from the true spirit and scope of the invention. The appended claims are intended to be construed to include all such embodiments and equivalent variations.
Claims
What is claimed is:
1. A method of detecting a collagen gene alteration associated with a pathological condition in a human subject, the method comprising
obtaining from the subject a sample nucleic acid comprising a portion which consists of at least fifteen consecutive nucleotides of a segment of the gene, wherein the portion comprises at least one intronic nucleotide, a first site, and a second site;
determining the nucleotide sequence of the portion; and
comparing the nucleotide sequence of the portion with the corresponding consensus nucleotide sequence of the gene,
whereby a difference between the nucleotide sequence of the portion and the corresponding consensus nucleotide sequence indicates the presence of the collagen gene alteration in the subject,
wherein the segment is the segment of the COL1A1 gene extending in the 5′- to 3′-direction from and including the 78 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of intron 51
wherein the consensus nucleotide sequence of the COL1A1 gene is SEQ ID NO: 1.
2. The method of claim 1, further comprising contacting the sample nucleic acid with a first intronic primer prior to determining the nucleotide sequence of the portion, the first intronic primer being either substantially complementary to or substantially homologous with the first site.
3. The method of claim 2, wherein the first intronic primer has a sequence selected from the group consisting of SEQ ID NOs: 273-299 and 301-319.
4. The method of claim 1, further comprising
contacting the sample nucleic acid with a first intronic primer homologous with the first site,
contacting the sample nucleic acid with a second primer complementary to the second site, and
amplifying the portion to obtain an amplified polynucleotide prior to determining the nucleotide sequence of the portion, wherein determining the nucleotide sequence of the portion comprises determining the nucleotide sequence of the amplified polynucleotide.
5. The method of claim 4, further comprising performing conformation-sensitive gel electrophoresis (CSGE) after amplifying the portion and prior to determining the nucleotide sequence of the portion, the CSGE comprising
denaturing the amplified polynucleotide,
annealing the amplified polynucleotide, and
determining whether the amplified polynucleotide forms a heteroduplex.
6. The method of claim 4, wherein the first intronic primer is either substantially complementary to or substantially homologous with a part of a non-coding region of the segment that has a sequence selected from the group consisting of SEQ ID NOs: 6, 7, 24-57.
7. The method of claim 4, wherein the first intronic primer has a sequence selected from the group consisting of SEQ ID NOs: 273-299 and 301-319.
8. The method of claim 4, wherein the first primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
9. The method of claim 4, wherein
the sample nucleic acid is contacted with a plurality of pairs of primers and a plurality of portions of the segment are amplified to obtain a plurality of amplified polynucleotides.
10. The method of claim 9, further comprising performing CSGE after amplifying the portions and prior to determining the nucleotide sequence of the portions, the CSGE comprising
denaturing the amplified polynucleotides,
annealing the amplified polynucleotides, and
determining whether any of the amplified polynucleotides forms a heteroduplex.
11. The method of claim 10, wherein each of the primers is either substantially complementary to or substantially homologous with a part of a non-coding region of the segment.
12. The method of claim 9, wherein the length of each amplified polynucleotide is between about two hundred and about five hundred nucleotides.
13. The method of claim 12, wherein the length of the polynucleotide amplified using each pair of primers is different than the length of the polynucleotides amplified using every other pair of primers.
14. The method of claim 9, wherein the plurality of pairs of primers comprises pairs of primers sufficient to amplify substantially all exons and exon flanking regions of the gene.
15. The method of claim 14, wherein the length of each amplified polynucleotide is between about two hundred and about five hundred nucleotides and is different from the length of every other amplified polynucleotide.
16. The method of claim 15, further comprising performing CSGE after amplifying the portions and prior to determining the nucleotide sequence of the portions, the CSGE comprising
denaturing the amplified polynucleotides,
annealing the amplified polynucleotides, and
determining whether any of the amplified polynucleotides forms a heteroduplex.
17. The method of claim 16, wherein the segment is the segment of the COL1A1 gene extending in the 5′- to 3′-direction from and including the 78 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of intron 51, and wherein each pair of primers comprises a first intronic primer and a second primer, the first intronic primer and second primer being selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
18. The method of claim 17, wherein the pathological condition is selected from the group consisting of osteoporosis, multiple epiphyseal dysplasia, osteogenesis imperfecta, shortness of stature, and low bone density.
19. An isolated nucleic acid comprising at least about fifteen consecutive nucleotides,
wherein the isolated nucleic acid is either completely complementary to or completely homologous with at least fifteen consecutive nucleotides of a segment of a human collagen gene, including at least one nucleotide of the segment located in a non-coding region of the gene, and
wherein the segment is the segment of the COL1A1 gene (SEQ ID NO: 1) extending in the 5′- to 3′-direction from and including the 78 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of intron 51.
20. The isolated nucleic acid of claim 19, wherein the nucleic acid is completely complementary to the segment.
21. The isolated nucleic acid of claim 19, wherein the nucleic acid is completely homologous to the segment.
22. The isolated nucleic acid of claim 19, wherein the isolated nucleic acid is complementary to or homologous with at least three nucleotides of the segment located in the non-coding region.
23. The isolated nucleic acid of claim 19, wherein the segment comprises only nucleotides located in the non-coding region and wherein the non-coding region has a nucleotide sequence selected from the group consisting of SEQ ID NOs: 6, 7, 24-57.
24. The isolated nucleic acid of claim 19, having a nucleotide sequence selected from the group consisting of SEQ ID NOs: 273-299 and 301-319.
25. A kit for detecting an alteration in a segment of a human collagen gene, the kit comprising a consensus sequence of the gene, a first intronic primer, and a second primer,
wherein each of the first intronic primer and the second primer comprises at least about fifteen consecutive nucleotides,
wherein the segment comprises at least one nucleotide located in a non-coding region of the gene, a first site, and a second site, the first intronic primer being substantially homologous with the first site and the second primer being substantially complementary to the second site,
wherein the segment is the segment of the COL1A1 gene extending in the 5′- to 3′-direction from and including the 78 nucleotides of intron 27 located adjacent exon 28 through the 3′-end of intron 51
wherein the consensus nucleotide sequence of the COL1A1 gene is SEQ ID NO: 1.
26. The kit of claim 25, wherein the length of the segment is between about two hundred and about five hundred nucleotides.
27. The kit of claim 25, further comprising at least one sequencing primer for determining the nucleotide sequence of at least a portion of the segment.
28. The kit of claim 25, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
29. The kit of claim 28, comprising a plurality of the pairs of primers.
30. The method of claim 8, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
31. The method of claim 8, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
32. The method of claim 8, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
33. The method of claim 8, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
34. The method of claim 8, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
35. The method of claim 8, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
36. The kit of claim 28, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
37. The kit of claim 28, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
38. The kit of claim 28, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
39. The kit of claim 28, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
40. The kit of claim 28, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
41. The kit of claim 28, wherein the first intronic primer and the second primer are selected such that when the first intronic primer has a sequence listed in Column 1 of the following table, the second primer has the sequence listed on the same line in Column 2 of the following table:
Patent Citations (3)
Non-Patent Literature (120)
- Chu et al., Nucleic Acids Res. 10(19), 5925-5934 (1982).Applicant
- Ahmad et al., 1995, Arch. Ophthalmol. 113:1454-1457.Applicant
- Ala-Kokko et al., 1995, Biochem. J. 308:923-929.Applicant
- Ausubel et al., 1994, In: Current Protocols in Molecular Biology, Wiley, New York, vol. 1, pp. 2.2.1-2.2.3.Applicant
- Baldwin et al., 1989, Biochem. J. 262:521-528.Applicant
- Ballo et al., 1997, Am. J. Med. Genet. 68:396-400.Applicant
- Barsh et al., 1985 Proc. Natl. Acad. Sci. USA 82:2870-2874.Applicant
- Bernard et al., 1983, Biochem. 22:1139-1145.Applicant
- Bernard et al., 1983, Biochemistry 22:5213-5223.Applicant
- Birnboim et al., 1979, Nucl. Acids Res. 7:1513-1523.Applicant
- Boedtker et al., 1985, Ann. N.Y. Acad. Sci. 460:85-116.Applicant
- Bornstein et al., 1980 Ann. Rev. Biochem. 49:957-1003.Applicant
- Brewton et al., 1995, Genomics 30:329-336.Applicant
- Brewton et al., 1992, Eur. J. Biochem. 205:443-449.Applicant
- Briggs et al., 1995, Nature Genet. 10:330-336.Applicant
- Briggs et al., 1994, Am. J. Hum. Genet. 55:678-684.Applicant
- Brown et al., 1995, Hum. Molec. Genet. 4:141-142.Applicant
- Brown et al., 1992, Arch. Ophthalmol. 110:1589-1593.Applicant
- Bruckner et al., 1985, Proc. Natl. Acad. Sci. USA 82:2608-2612.Applicant
- Byers, 1993, In: Connective tissue and its heritable disorders, Royce et al., eds., Wiley-Liss, New York, pp. 317-351.Applicant
- Chipman et al., 1993, Proc. Natl. Acad. Sci. USA 90:1701-1705.Applicant
- Chu et al., 1993, In: Connective Tissue and Its Heritable Disorders, Royce et al., eds., Wiley-Liss, New York, pp. 149-165.Applicant
- Chu et al., 1985, J. Biol. Chem. 260:2315-2320.Applicant
- Chu et al., 1984, Nature 310:337-340.Applicant
- Cohn et al., 1996, Ann. N.Y. Acad. Sci. 785:188-194.Applicant
- Constantinou et al., 1990, Cytogenet. Cell Genet. 51:979.Applicant
- D'Alessio et al., 1988, Gene 67:105-115.Applicant
- de Wet et al., 1987, J. Biol. Chem. 262:16032-16036.Applicant
- Deere et al., 1995, Am. J. Hum. Genet. 56:698-704.Applicant
- Diab et al., 1996, Biochem. J. 314:327-332.Applicant
- Dickson et al., 1984, Proc. Natl. Acad. Sci. USA 81:4524-4528.Applicant
- Engel et al., 1991, Annu. Rev. Biophys. Biophys. Chem. 20:137-152.Applicant
- Eyre et al., 1987, FEBS Lett. 220:337-341.Applicant
- Fässler et al., 1994, Proc. Natl. Acad. Sci. USA 91:5070-5074.Applicant
- Fertala et al. 1993, Biochem. J. 289:195-199.Applicant
- Fu et al., 1988, Mol. Cell. Biol. 8:3582-3590.Applicant
- Ganguly et al., 1995, Electrophoresis 16:1830-1835.Applicant
- Ganguly et al., 1993, Proc. Natl. Acad. Sci. USA 90:10325-10329.Applicant
- Ganguly et al., 1991, J. Biol. Chem. 266:12035-12040.Applicant
- Ganguly et al., 1990, Nucl. Acids Res. 18:3933-3939.Applicant
- Gordon et al., 1990, Curr. Op. Cell Biol. 2:833-838.Applicant
- Gubler et al. 1983, Gene 25:263-269.Applicant
- Haimes et al., 1996, Inflam. Res. 44(Suppl.2):S127-S128.Applicant
- Hanke et al., 1994, BioTechniques 17:858-860.Applicant
- Har-El et al., 1992, J. Biol. Chem. 267:10070-10076.Applicant
- Innis et al., ed., 1990, In: PCR Protocols, Academic Press, Inc., San Diego (too voluminous to submit).Applicant
- Jacob et al., 1989, Nucl. Acids Res. 17:2159-2180.Applicant
- Jego et al. 1992, Oncogene 8:209-213.Applicant
- Kadler et al., 1987, J. Biol. Chem. 262:15696-15701.Applicant
- Kadler et al., 1990, Biochem. J. 268:339-343.Applicant
- Körkkö et al., 1997, Hum. Mutat. 9:148-156.Applicant
- Kuvianiemi et al., 1997, Hum. Mutat. 9:148-156.Applicant
- Kuivaniemi et al., 1991, FASEB J. 5:2052-2060.Applicant
- Kuivaniemi et al., 1988 Biochem, J. 252:633-640.Applicant
- Labhard et al., 1990, Matrix 10:124-130.Applicant
- Laemmli et al., 1970, Nature 227:680.Applicant
- Lozano et al., 1985, Proc. Natl. Acad. Sci. USA 82:4050-4054.Applicant
- Määttä et al., 1991, FEBS Lett. 279:9-13.Applicant
- Mechling et al., 1996, J. Biol. Chem. 271:13781-13785.Applicant
- Muragaki et al., 1990, Proc. Natl. Acad. Sci. USA 87:2400-2404.Applicant
- Muragaki et al., 1990, Eur. J. Biochem. 192:703-708.Applicant
- Myers et al., 1984, J. Biol. Chem. 259:12941-12944.Applicant
- Na et al., 1989, Biochem. 28: 7153-7161.Applicant
- Nakai et al., 1994, Gene 141:171-177.Applicant
- Nakata et al., 1993, Proc. Natl. Acad. Sci. USA 90:2870-2874.Applicant
- Nicholls et al., 1984, J. Med. Genet. 21:257-262.Applicant
- Ninomiya et al., 1990, In: Extracellular Matrix Genes, Sandell et al., eds., Academic press, San Diego , pp. 79-114.Applicant
- Ninomiya et al, 1985, Biochem. 24:4223-4229.Applicant
- Ninomiya et al., 1984, Proc. Natl. Acad. Sci. USA 81:3014-3018.Applicant
- Nishimura et al., 1989, J. Biol. chem. 264:20033-20041.Applicant
- Nuytincket al., 1996, Human. Genet. 97:324-329.Applicant
- Perälä et al., 1994, J. Biol. Chem. 269:5064-5071.Applicant
- Perälä et al., 1993, FEBS Lett. 319:177-180.Applicant
- Piez, 1984, In: Extracellular Matrix Biochemistry, Piez et al., eds., Elsevier Science pp. Pub. Co. Inc., New York, pp. 1-39.Applicant
- Pihlajaniemi et al., 1984, J. Biol. Chem. 259:12941-12944.Applicant
- Prockop et al., 1995, Annu. Rev. Biochem. 64:403-434.Applicant
- Prockop, 1990 Arth. Rheumat. 31:1-8.Applicant
- Prockop, 1990, J. Biol. Chem. 265:15349-15352.Applicant
- Prockop et al., 1989, In: Cytoskeletal and Extracellular Proteins, Aebi et al., eds. Springer Series in Biophysics, vol. 3, pp. 81-89.Applicant
- Prockop, 1986, Hosp. Pract., Feb. 15, 1986.Applicant
- Prockop et al., 1989, Biophysics (Eng. Transl. Biofizika) 3:81-89.Applicant
- Prockop, 1985, J. Clin. Invest. 75:783-787.Applicant
- Prockop et al., 1984, N. Eng. J. Med. 311:376-386.Applicant
- Prockop et al., 1979, N. Eng. J. Med. 301:13-23.Applicant
- Ramirez et al., 1985, Ann. New York Acad. Sci. 460:117-129.Applicant
- Redford-Badwal et al., 1996, J. Clin. Invest. 97:1035-1040.Applicant
- Ritvaniemie et al., 1993, Genomics 17:218-221.Applicant
- Rokos et al., 1994, Matrix Biol. 14:1-8.Applicant
- Saban et al., 1996, BioTechniques 21:190-192.Applicant
- Sambrook, et al., 1989, In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, New York (too voluminous to submit).Applicant
- Sanger et al., 1977, Proc. Natl. Acad. Sci.USA 74:5463-5467.Applicant
- Shapiro et al., 1989, Connect Tissue Res. 21:117-123.Applicant
- Shapiro et al., 1987, Nucl. Acids Res. 15:7155-7174.Applicant
- Shaw et al., 1991, Trends Biochem. Sci. 16:191-194.Applicant
- Sherwood et al., 1990, Gene 89:239-244.Applicant
- Sokolov et al., 1991, Hum. Genet 88: 125-129.Applicant
- Spotila et al., 1990, Am. J. Hum. Genet. 47:A237.Applicant
- Sykes et al., 1990, Am. J. Hum. Genet. 46:293-307.Applicant
- Ting et al., 1993, J. Bone Min. Res. 8:1377-1387.Applicant
- Tromp et al., 1988, Proc. Natl. Acad. Sci. USA 85:5254-5258.Applicant
- Tromp et al., 1988, Biochem. J. 253:919-929.Applicant
- Tsuneyoshi et al., 1991 Biol. Chem. 266: 15608-15613.Applicant
- van der Rest et al., 1988, J. Biol. Chem. 263:1615-1618.Applicant
- van der Rest et al., 1987, In Structure and Function of Collagen Types, Mayne et al., eds., Academic Press, Orlando, pp. 195-221.Applicant
- Vasan et al., 1991, Am. J. Hum. Genet. 48:305-31.Applicant
- Vasios et al., 1988, J. Biol. Chem. 263:2324-2329.Applicant
- Vaughan et al., 1988, J. Cell. Biol. 106:991-997.Applicant
- Vuorio et al., 1990, Annu. Rev. Biochem. 59:837-872.Applicant
- Weil et al., 1990, J Biol. Chem. 265:16007.Applicant
- Weil et al., 1989, J. Biol. Chem. 264:16804.Applicant
- Weil et al., 1989, EMBO J. 8:1705.Applicant
- Weil et al., 1988, J. Biol. Chem. 263:8561.Applicant
- Werle et al., 1994, Nucl. Acids Res. 22:4354-4355.Applicant
- Westerhausen et al., 1991, Matrix 11:375-379.Applicant
- Westerhausen et al., 1990, J. Biol. Chem. 265:13995-14000.Applicant
- Willing et al. 1996, Am. J. Hum. Genet. 59:799-809.Applicant
- Willing et al. 1994, Am. J. Hum. Genet. 55:638-647.Applicant
- Willing et al., 1992, Am. J. Hum. Genet. 51:508-515.Applicant
- Wood et al., 1960, Biochem. J. 75:588:598.Applicant
- Wu et al., 1992, J. Biol. Chem. 267: 23007-23014.Applicant