GENOMICS

8, 562-567

(1990)

SHORT COMMUNICATION Complete Amino Acid Sequence of Human Cartilage Link Protein (CRXI) Deduced from cDNA Clones and Chromosomal Assignment of the Gene’ SHERRI L. OSBORNE-LAWRENCE,*‘t ANDREA K. SINCLAIR,*,t ROBERT C. Hlcw,*,t STEPHEN W. LAcEy,t ROGER L. EDDY, JR.,* MARY G. BYERS,* THOMAS B. SHOWS,* AND ALLAN D. DUBY*,tr2 *Harold C. Simmons Arthritis Research Center, tDepartment of internal Medicine, University of Texas Southwestern Medical Center, Dallas, Texas 75235; and *Department of Human Genetics, Roswell Park Memorial Institute, Buffalo, New York 14263 Received

February

20, 1990;

Cartilage link protein binds to hyaluronic acid and the keratan sulfate/chondroitin sulfate proteoglycan of cartilage and stabilizes proteoglycan/hyaluronate aggregation (reviewed in Hascall, 1988). Link proteins have also been reported in aorta (Garde11 et al., 1980; Vijayagopal et al., 1985; Barday et al., 1985), eye 1 This work was supported by USPHS Grant GM20454 (T.B.S.) and Kll-1860 (S.W.L.). ’ Pew Scholar in the Biomedical Sciences and to whom correspondence should be addressed at Department of Internal Medicine, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75235-6664.

Copyright All rights

562

$3.00

0 1990 by Academic Press, of reproduction in any form

Inc. reserved.

June 7, 1990

(Poole et al., 1982), and synovium (Fife et al., 1985), largely on the basis of immunological criteria, although the structural relationship among these species and their functions have not been fully established. cDNAs have been reported for chicken (De&k et al., 1986), rat (Doege et al., 1986; Rhodes et al., 1988), and porcine (Dudhia and Hardingham, 1989) cartilage link protein. Additionally, a partial protein sequence has been reported for bovine nasal cartilage link protein (P&in et al., 1980) and a complete protein sequence has been reported for rat chondrosarcoma link protein (Neame et al., 1986). Depending on the species studied and the amount of carbohydrate content, link protein has a molecular weight of 42,000-49,000 and has three domains that have been labeled A, B, and Bl. All three domains share homologous domains in the amino-terminal region of the major proteoglycan of cartilage (Doege et al., 1987) and versican, a large fibroblast proteoglycan (Zimmerman and Ruoslahti, 1989). Versican contains sequences virtually identical to partial peptide sequences from a glial hyaluronate-binding protein (Perides et al., 1989). The A domain, which in its carboxy1 terminus shares homology to the corresponding region of variable domains of members of the immunoglobulin supergene family (Bonnet et al., 1986), appears to be involved in proteoglycan/link protein interaction (P&in et al., 1987). The B and Bl domains, which are homologous to each other, bind hyaluronic acid (P&in et al., 1987; Goetinck et al., 1987). Depending on the method of analysis, two or three forms of cartilage link protein have been found in human articular cartilage (Pal et al., 1978; Roughley et al., 1982; Mort et al., 1983). To begin to understand

Little is known about the primary amino acid structure of human-cartilage link protein (CXI’U). We screened a human genomic library with a cDNA encoding the 3’ untranslated region and the adjoining Bl domain of chicken link protein. One clone was isolated and characterized. A 3.6-kb EcoRI-KpnI fragment from this genomic clone that contains the human Bl exon was used to map the gene to chromosome 5q13 3 q14.1. The same fragment was used to screen a cDNA library prepared from mRNA of Caco-2, a human colon tumor cell line. Two overlapping clones were isolated and shown to encode all of CRTLl. The deduced amino acid sequence is 354 residues long. The amino acid sequence shows a striking degree of identity to the porcine (96%), rat (96%), and chicken (85%) link protein sequences. Furthermore, there is greater than 86% homology between the 3’ untranslated region of the genes encoding human and porcine link proteins. These results indicate that there has been strong evolutionary pressure against changes in the coding and 3’ untranslated regions of the gene encoding cartilage link protein. o 1990 Academic PWS, IIIC.

o&x3-7543/90

revised

SHORT

COMMUNICATION

the structural reasons for this heterogeneity in size, we have isolated and analyzed cDNAs encoding the human gene for cartilage link protein. To obtain a DNA probe specific for the human gene encoding cartilage link protein for subsequent Northern blot analysis, we screened a human genomic library (Maniatis et al., 1978) with pLPH1, a partial cDNA containing the Bl exon of the chicken link protein and adjoining 3’ untranslated (UT) region. One clone, HLP(G)-1, that hybridized strongly to the probe was isolated. A 4.4kb EcoRI-EcoRI fragment that hybridized to the chicken probe was subcloned into pBR322.A 3.5-kb KpnI-EcoRI fragment, HBl, that hybridized to pLPH1 was purified from the subclone and used as a DNA probe for Northern blot analysis. As discussed above, cartilage link protein has been identified in several tissues other than cartilage. Different proteoglycans capable of aggregating with hyaluronate are synthesized by a large number of cell types, including aortic smooth muscle cells, colon cells, glial cells, skin fibroblasts, lung tissue, embryonic bone, tendon, and sclera (reviewed in Kimura et al., 1987). Because of the above, we hypothesized that “cartilage” link protein is expressed by several different tissues where it binds hyaluronic acid toproteoglycans different from the major proteoglycan of cartilage. Therefore, we investigated the expression of the gene encoding human cartilage link protein in several human tissues and lines. A strong signal to HBl was detected by Northern blot analysis of total RNA from Caco-2, a human colon tumor line (Fogh et al., 1977) (data not shown). We suspect that link protein is binding to the hyaluronic acid and aggregating proteoglycans that are found in both normal colonic epithelium and colonic carcinomas (Iozzo and Wight, 1982). We hybridized HBl to a Xgtll cDNA library prepared from Caco-2 and isolated two independent, overlapping clones, HLP(C)-1 and HLP(C)-2. Together these two cDNAs encode 73 bp of 5’ untranslated region, 1062 bp of coding region, and 580 bp of 3’ untranslated region. (This sequence has been submitted to Genbank.) The nucleotide sequence of the cDNAs encoding the Bl exon and the adjoining 3’UT was identical in all 874 positions to the corresponding region of the genomic clone, HLP(G)-1. While this maiiuscript was in preparation, Dudhia and Hardingham (1989) published the sequence of a partial 500-bp cDNA encoding human cartilage link protein. The deduced amino acid sequence has a presumed 15-residue leader sequence, based on the reported rat link protein sequence (Neame et al., 1986) (Fig. 1). The deduced sequence of the mature polypeptide is 339 residues long and has a calculated M, of 38,485. There are two potential N-linked glycosylation sites which could account for some or all of the heterogeneity in size, discussed above.

563

In Fig. 1, we have compared the leader, amino terdomains of the huminus, and “A”, “B”, and “Bl” man sequence to their counterparts in the rat, porcine, and chicken sequences. In the mature 339~residue polypeptide, the human and porcine sequences are 96% identical (327/339), human and rat sequences are 96% identical (326/339), and human and chicken sequences are 85% (288/339) identical. The majority of the changes are in the amino terminus and “A” domains. The potential N-linked glycosylation site in the A domain is conserved among the four species, while the rat lacks the potential N-linked glycosylation site present in the amino terminus domain of the other three species. Caterson et al. (1985) produced monoclonal antibody (mAB) 8A4 against rat link protein and demonstrated its binding to the 20-amino-acid regions that are situated between the two centrally located cysteine residues in the B and Bl domains. The corresponding regions in the human sequence, 206-225 and 305-324, are identical to the rat sequence. It is obvious from sequence analysis why mAB 8A4 binds well to human link protein. As shown in Fig. 1, sequence 206-225 is identical in the porcine and chick sequence, while one of the arginines (R) present in sequence 305-324 is replaced by a lysine (K) in the chicken sequence. Goetinck et al. (1987) showed that the epitopes for mAB 8A4 could be further narrowed to a lo-peptide stretch indicated in Fig. 1. Apparently, the substitution of a lysine for an arginine at position 323 of the chicken sequence does not prevent binding of 8A4. In the same study, the authors demonstrated that these two regions were involved in binding to hyaluronic acid. Two further regions were also shown to be involved in binding hyaluronic acid: 241-250 and 338-347 in the human sequence. As can be seen in Fig. 1, sequence 338-347 is identical in all four species. Glutamic acid-arginine (ER) is conservatively replaced by aspartic acid-lysine (DK) in peptide 241250 in the rat, porcine, and human proteins. Using the Gap program of the University of Wisconsin Genetics Computer Group (Devereux et cd., 1984), we compared the 3’UT region of the human link protein gene to the 3’UT regions of the chicken (Desk et al., 1986), rat (Rhodes et aZ., 1988), and porcine (Dudhia and Hardingham, 1989) link protein genes that have been sequenced (Fig. 2). Ignoring gaps that have been inserted to optimize the alignment, there is 65% homology between the 5’ end of the 3’UT regions of the human and rat sequences (only 87 bp of the rat 3’UT region have been sequenced), 76% homology between the 4’ 476 bp of the human 3’UT sequence compared to the corresponding region of the chicken sequence, and 87% homology between the 5’ 261 bp of the human sequence compared to the corresponding

564

SHORT

I.

LEADER

MAN PIG RAT CHICK

III.

MKSLLLLVLiSICWA V R F VR T F V

COMMUNICATION

II. 15

AMINO TERMINUS

MAN PIG RAT CHICK

DHLSbNYTLDHDRAiHIQ NS-PO vv EPHP _sS E E I

33

A DOMAIN

MAN PIG RAT CHICK

AENGPHiLVEAEQAKViSHRGGNVTLPCKFYRDPT-AFGSGIHKIRI~WTKLTSDYL~EVDVFVSMG~HKKTYGGYD~RVFL R F T R R : R V I Q = HEH STA T V A NRS

MAN PIG RAT CHICK

KGGSD~DASLVITDL;LEDYGRYKC~VIEGLEDDT;VVALDLQ N A A RES Elj ; NIM A

IV.

114 H K

157 E E NE

B DOMAIN 1 mAB8A4

MAN PIG RAT CHICK

GV~FPYFPRLGR~NLNFHEAQ~~CLDDDAVIASFDQLYDAWR~GLDWCNAGW~SDGSVQYPI~KPREPCGGQ~TVPGVRNYG~WDKDKSRYD~FCFTSNFN

258

R S

SI

E

S

K

Bl DOMAIN

V.

**********

1

t mAB8A4 MAN PIG RAT CHICK

ER

**********

{

GRFYYLiHPTKLTYDE~VQACLNDGA~IAKVGQIF~WKILGYDRC~AGWLADGSV~YPISRPRRR~SPTEAAVRF~GFPDKKHKL~GVYCFRAYN' L RN L K L K N

354

FIG. 1. Comparison of the amino acid sequence of the human link protein to those of porcine, rat, and chicken link protein. The human sequence has been divided into the following domains: I, Leader; II, amino terminus; III, “A” domain; IV, “B” domain; and V, “Bl” domain. The human amino acid sequence is shown to the right of the sequence; every tenth residue is dotted. The porcine, rat, and chicken sequences are displayed below the human sequence. Only those residues that differ from the human sequence are displayed. A gap, indicated by a dash (--), has been inserted in the human, porcine, and rat sequences between positions 68 and 69 to permit optimal alignment with the chicken sequence. Sights of potential N-glycosylation are underlined. The two sites of binding of monoclonal antibody 8A4 to chicken link protein demonstrated by Goetinck et al. (14) are shown (positions 216-225 and 315-324). These authors demonstrated binding of these same two sites to hyaluronic acid; hyaluronic acid was shown to bind to two further sites that are starred (*) (positions 241-250 and 338-347).

252 bp of the porcine 3’UT region that have been sequenced. The sequence homology in the 3’UT region among the four species is intriguing. High degrees of sequence homology have been demonstrated in the 3’UT region between mRNAs for homologous (isotypic) proteins (e.g., actins) in different organisms but not between mRNAs coding for very similar isoforms differing in their function or tissue specificity (e.g., actins) (reviewed in Yaffe et al., 1985). Further-

more, it has been demonstrated that for the @-actin genes that the region of high sequence homology ends 12 nucleotides after the polyadenylation signal; i.e., it is confined to the sequence that is preserved in the mature RNA. In many cases the sequence conservation in the 3’UT region is higher than that in the introns and in the silent sites in the coding region of genes. The functions of the 3’UT region are beginning to be elucidated. The pentamer ATTTA is present sin-

MAN PIG CHICK

ATGTGCCCTTAGAGC-GCATCAGTTTTAAAGTCATTAAGAAACT GC C -A GG AA AT A C T T

MAN PIG CHICK

GTW\T-AACCCTTTTTTACTTACTGTAAAGAGTCATTTTCAT~---AGATCAATTCATTGATTTGT---------------TTTTTGTAAAGCT ___ c _____-__________ G A __ -A C ACC A A C A TG T GTTTCTGTTTAAATA

MAN PIG CHICK

ATCATTCAATATATATTATATTAATATAAATTTAAGGG~GCTCTATGTAAGGAGACTTAGAGCCA~CTGTTT~GCTGTATCATCCCAACA -_ -_ - ATG CA G AC C A ----A C G TA A G T G TC ACG

-c __________

A

__ T G

G

At

GA

FIG. 2. Comparison of the 3’UT region of the human link protein gene compared to the 3’UT region of porcine and rat link protein genes. The human sequence (top rows) at the 5’ end of the 3’UT region is compared to the comparable porcine and rat sequences. The alignment starts with the first nucleotide after the stop codon. Gaps, indicated by dashes (--), have been included to maximize the alignment. Only 252 bp of the porcine 3’UT have been sequenced. As shown in the figure, it is homologous to the 5’ 265 bp of the human sequence and to the 4’ 263 bp of the chicken sequence.

SHORT

t

565

COMMUNICATION

-MH

-9.2Kb

-5.2Kb I234 FIG. 3. Southern blot of somatic cell hybrids. A representative blot is shown. DNA (10 rg) prepared from the somatic cell hybrids, XER-‘7 (lane +) and XTR-3BSAgB (lane -), murine cells (RAG, lane M), and human cells (GM00131, lane II) was digested with Hi&II, size-fractionated on an 0.8% gel, transferred to nitrocellulose, and hybridized to the HBl probe. The HBl probe hybridized to a 5.2-kb murine restriction fragment, whereas it hybridized to a 9.2-kb human fragment. Somatic hybrid cells, XER, were scored as positive and somatic hybrid cells, XTR-BBSAgB, were scored as negative. The variation in intensity of human bands reflects the fact that the somatic cell hybrids used for the preparation of DNA are normally not diploid for all chromosomes and are a mixed population; i.e., not all cells retain the human chromosomes of interest. Lane 1, hybrid positive for chromosome 5; lane 2, hybrid negative for chromosome 5: lane 3, mouse DNA, lane 4, human DNA.

gly or in multiple reiterations in a wide variety of oncogene and cytokine 3’UT. Removal of this region confers significantly greater stability to messages produced from transfected constructs, whereas the addition of a short DNA segment coding for this motif destabilizes previously stable messages (reviewed in Malter, 1989). The 3’UT region of the transferrin receptor contains sequences for two distinct functions; one is related to an RNA instability determinant that gives this mRNA a short half-life, and the other is the site of binding of an iron-responsive protein that can regulate the use of the instability element. Of the 160 nucleotides that encode these elements, only 6 differ between human and chickens (reviewed in Klausner and Harford, 1989). It remains to be seen whether the 3’UT region is involved in regulation of the translation and stability of link protein mRNA. Southern blot analysis of somatic cell hybrids and in situ chromosomal hybridization were used to map the human gene for cartilage matrix protein. The results of Southern blot analysis of genomic DNA from 31 mouse X human hybrids (Fig. 3) probed with HBl are shown in Table 1. The discordancy scores localize the human cartilage link protein gene to chromosome 5. One can also see from Fig. 3 that on HindIII-digested human DNA there is evidence of only one gene hybridizing to this probe. Similarly, only one band is detected on EcoRI and BumHI digestion, under simi-

lar stringent conditions of hybridization and washing (data not shown). To determine the subchromosomal localization of the human gene encoding cartilage link protein, we hybridized HBl to normal metaphase chromosomes. Of 118 metaphase cells examined from this hybridization, 18.8% (52/277) of the total grains were associated with chromosome 5, corroborating the results of the Southern blot analysis of the somatic cell hybrids. Furthermore, 9.4% of the total grains (26/277) and 50% (26/52) of the grains on chromosome 5 were located at 5q13 + q14.1 (Fig. 4). We therefore conclude that the human gene for cartilage link protein is localized at position 5q13 + q14.1. TABLE Concordancy

Analysis

1

of Somatic

Cell Hybrid

Panel’

Hybridization of probe/ presence of chromosome Concordant Chromosome 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 X

+/+ 7 11 10 11 15 9 10 12 5 13 7 13 9 13 11 6 12 11 10 11 12 8 10

Discordant -/14 10 9 9 16 13 9 10 13 7 10 11 9 8 9 14 4 10 13 8 3 9 7

+/--

7 4 4 4 0 6 4 3 8 2 6 2 6 2 4 9 2 4 5 4 3 7 3

-/+

% Discordancy 1 5 6

7 0 3 6 6 3 8 6 5 7 8 7 2 12 6 3 8 13 5 6

28 30 34 35 0 29 34 29 38 33 41 23 42 32 35 35 47 32 26 39 52 41 35

’ This table is compiled from 31 hybrids involving 14 unrelated human cell lines and 4 mouse cell lines (35,37). The hybrids were characterized by karyotypic analysis, by mapped enzyme markers, and partly by mapped DNA probes (34-36). The DNA probe, HBl, was hybridized to Southern blots containing ZfindIII-digested DNA from the human-mouse hybrids. The scoring for HBl was determined by the presence (+) or absence (-) of human bands in the hybrids on the blots. Concordant hybrids have either retained or lost the human bands together with a specific human chromosome. Discordant hybrids have either retained the human bands but not a specific chromosome or the reverse. Percentage discordancy indicates the degree of discordant segregation for a marker and a chromosome. A 0% discordancy is the basis for chromosome assignment. HBl mapped to human chromosome 5.

566

SHORT

COMMUNICATION mic library. We thank Drs. R. J. Baer, B. Beutler, P. E. Lipsky, R. J. MacDonald, and D. W. Russell for useful discussion.

REFERENCES 1.

BARDAY, H. B., MURRAY, E., AND BURDEN, T. S. (1985). The isolation of putative proteoglycan link protein from bovine aorta. Biochem. Znt. 10: 829-837.

2.

BONNET, F., P~RIN, J., LORENZO, F., JOLLBS, J., AND JOLL&, P. (1986). An unexpected homology between link proteins of the proteoglycan complex and immunoglobulin-like proteins. Biochim. Biophys. Acta 873: 152-155. CARLOCK, L. R., SKARECKY, D., DANA, S. L., AND WASMUTH, J. J. (1985). Deletion mapping of human chromosome 5 using chromosome-specific DNA probes. Amer. J. Hum. Genet. 37: 839-852.

3.

4.

5. 0

32 33 34 35

.a

ti-

5 FIG. 4. Distribution of labeled sites on chromosome 5. The figure summarizes the analysis of 115 normal human metaphase cells from phytohemagglutinin-stimulated peripheral blood lymphocytes that were hybridized to the HBl probe. Each dot represents one labeled site observed in the corresponding band. Fifty percent (26/52) of the labeled sites on chromosome 5 were located at q13 -+ q14.1; this cluster represented 9.4% (26/277) of all labeled sites.

Only four genes have been mapped to the subchromosomal region of 5q13 + ql4.1:hexosaminidase B (p polypeptide) to5q13 (GilbertetaL, 1975); argininosuccinate synthetase pseudogene 10 to 5q13 + q23 (Su et al., 1984; Carlock et al., 1985); ribosomal protein S20A (gene or pseudogene) to 5q13 + q23 (Nakamichi et al., 1986); and 3-hydroxy-3-methylglutaryl-coenzyme A reductase to 5q13.3 + q14 (Humphries et al., 1985; Lindgren et al., 1985; Mohandas et al., 1986). These genes bear no obvious relationship to the gene encoding human cartilage link protein. No disorders that could be related to abnormalities of the human cartilage link protein gene have been mapped to the long arm of chromosome 5. It remains to be determined whether allelic differences in the coding regions or regulatory regions of the gene encoding cartilage link protein predispose individuals to various human disorders.

6.

7.

8.

9.

10.

11.

12.

DOEGE, K., SASAKI, M., HOFUGAN, E., HASSELL, J. R., AND YAMADA, Y. (1987). Complete primary structure of the rat cartilage proteoglycan core protein deduced from cDNA clones. J. Biol. Chem. 262: 17757-17767. DIJDHIA, J., AND HARDINGHAM, T. E. (1989). Isolation and sequence of cDNA clones for pig and human cartilage link protein. J. Mol. Biol. 206: 749-753. FIFE, R. S., CATERSON, B., AND MYERS, S. L. (1985). Identification of link proteins in canine synovial cell cultures and canine articular cartilage. J. Cell Biol. 100: 1050-1055. FOGH, J., FOGH, J. M., AND Ornnzo, T. (1977). One hundred and twenty-seven human tumor cell lines producing tumors in nude mice. J. Natl. Cancer Inst. 59: 221-226. GARDELL, S., BAKER, J., CATERSON, B., HEINEG~RD, D., AND ROD$N, L. (1980). Link protein and a hyaluronic acid-binding region as components of aorta proteoglycan. B&hem. Biophys. Res. Commun. 95: 1823-1831. GILBERT, F., KUCHEXUA~ATI, R., CREAGAN, R. P., MURNANJX, M. J., DARLINGTON, G. J., AND RUDDLE, F. H. (1975). TaySach’s and Sandhoff s diseases: The assignment of genes for hexoseaminidase A and B to individual human chromosomes. Proc. Natl. Acad. Sci. USA 72: 263-267.

14.

GOETINCK, P. F., STIRPE, N. S., TSONIS, P. A., AND CARLONE, D. (1987). The tandemly repeated sequences of cartilage link protein contain the sites for interaction with hyaluronic acid. J. Cell Biol. 105: 2403-2408. HASCALL, V. C. (1988). Proteoglycans: The chondroitin sulfate/keratan sulfate proteoglycan of cartilage. ZSZ AtZus Sci. Biochem. 1: 189-198.

15.

16. Dr. P. F. Goetinck for his generous gift of the chicken probe and T. Maniatis for his gift of the human geno-

DEVEREUX, J., HAJZBERLI, P., AND SMITHIES, 0. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12: 387-395. DOEGE, K., HASSELL, J. R., CATERSON, B., AND YAMADA, Y. (1986). Link protein cDNA sequence reveals a tandemly repeated protein structure. Proc. Natl. Acad. Sci. USA 83: 3761-3765.

13.

ACKNOWLEDGMENTS We thank link protein

CATERSON, B., BAKER, J. R., CHFUSTNER, J. E., LEE, Y., AND LENTZ, M. (1985). Monoclonal antibodies as probes for determining the microheterogeneity of the link proteins of cartilage proteoglycan. J. Biol. Chem. 260: 1134811356. DEAK, F., KISS, I., SPARKS, K. J., ARGRAVES, W. S., HAMPIKIAN, G., AND GOETINCK, P. F. (1986). Complete amino acid sequence of chicken cartilage link protein deduced from cDNA clones. Proc. Natl. Acad. Sci. USA 83: 3766-3770.

H~MPHRIES, S. E., TATA, F., HENRY, HOLM, M., JUNIEN, C., AND WILLIAMSON, tion, characterization, and chromosomal

I., BARICHARD, F., R. (1985). The isolaassignment of the

SHORT gene for human 3-hydroxy-3-methylglutaryl ductase (HMG-CoA reduetase). Hum.

Genet.

567

COMMUNICATION

coenzyme A re71: 254-258.

30.

P~RIN, J.-P., BONNET, F., THURIEAU, C., AND JOLJ&, P. (1987). Link protein interactions with hyaluronate and proteoglycans: Characterization of two distinct domains in bovine cartilage link proteins. J. Biol. Chem. 262: 13269-13272.

17.

Iozzo, R. V., AND WIGHT, terization of proteoglycans colon carcinoma. J. Biol.

T. N. (1982). Isolation and characsynthesized by human colon and Chem. 257: 11135-11144.

31.

18.

KIMURA, J. H., SHINOMURA, T., AND THONAR, E. J.-M. A. (1987). Biosynthesis of cartilage proteoglycan and link protein. In “Methods in Enzymology” (L. W. Cunningham, Ed.), Vol. 144, pp. 372-393, Academic Press, San Diego.

POOLE, A. R., Prnoux, I., REINER, A., CGSTER, L., AND HASSEL, J. R. (1982). Mammalian eyes and associated tissues contain molecules that are immunologically related to cartilage proteoglycan and link protein. J. Cell Bc’ol. 93: 910-920.

32.

19.

KLAUSNER, R. D., AND HARFORD, J. B. (1989). els for post-transcriptional gene regulation. 870-872.

RHODES, C., DOEGE, K., SASAKI, M., AND YAMADA, Y. (1988). Alternative splicing generates two different mRNA species for rat link protein. J. Biol. Chem. 263: 6063-6067.

33.

20.

LINDGREN, V., LUSKEY, K. L., RUSSEL, D. W., AND FRANCKE, U. (1985). Human gene for HMG CoA reductase maps to 5q133-q14 (HMG8). Proc. Nutl. Acd Sci. USA 82: 85678571. MALTER, J. S. (1989). Identification of an AUUUA-specific messenger RNA binding protein. Science 246: 664-666. MANLATIS, T., HARIXSON, R. C., LACY, E., LAUER, J., O’CONNELL, C., QUON, D., SIM, G. K., AND EFSTRATIDIS, A. (1978). The isolation of structural genes from libraries of eucaryotic DNA. Cell 15: 687-701.

ROUGHLEY, P. J., POOLE, A. R., AND MORT, J. S. (1982). The heterogeneity of link proteins isolated from human articular cartilage proteoglycan aggregates. J. Biol. Chem. 257: 1190811914.

34.

SHOWS, T. B., BROWN, J. A., HALEY, L. L., EDDY, R. L., COOPER, E. S., AND GOOGIN, A. P. ment of the @-glucuronidase structural gene to region of chromosome 7 in man. Cytogenet. 99-104.

35.

SHOWS, T. B., SAKAGUCHI, A. Y., AND NAYLCIR, S. L. (1982). Mapping the human genome, cloned genes, DNA polymorphisms, and inherited diseases. In “Advances in Human Genetics” (H. Harris and K. Hirschhorn, Eds.), Vol. 12, pp. 341452, Plenum Press, New York.

36.

SHOWS, T. B. (1983). Human genome organization of enzyme loci and metabolic diseases. In “Isozymes: Current Topics in Biological and Medical Research” (M. Rattazzi, M. C., Scandalios, J. G., and Whitt, G. S., Eds.) Vol. 10, pp. 323-339, Liss, New York.

37.

SHOWS, T., EDDY, R., HALEY, L., BYERS, M., HENRY, M., FuJITA, T., MATSUI, H., AND TANUGUCHI, T. (1984). Interleukin 2 (IL2) is assigned to human chromosome 4. Somatic Cell Mol. Genet. 10: 315-318.

38.

Su, T. S., NUSSBAUM, R. L., AIRHART, MOHANDAS, T., O’BRIEN, W. E., AND Human chromosomal assignments for thetase pseudogenes: Cloned DNAs netic analysis. Amer. J. Hum. Genet.

39.

VLJAYAGOPAL, P., RADHAKRISHNAMURTHY, B., SRINIVASAN, S. R., AND BERENSON, G. S. (1985). Isolation andcharacterization of a link protein from bovine aorta proteoglycan aggregate. Biochim. Biophys. Acta 839: 110-118.

40.

YAFFE, D., NUDEL, U., MAYER, Y., AND NEUMAN, S. (1985). Highly conserved sequences in the 3’ untranslated region of mRNAs coding for homologous proteins in distantly related species. Nucleic Acids Res. 10: 3723-3737.

41.

ZIMMERMAN, D. R., AM) RUOSLAHTI, E. (1989). Multiple domains of the large fibroblast proteoglycan, versican. EMBO J. 8: 2975-2981.

21. 22.

23.

24.

25.

26.

27.

28.

29.

Cis-trans Science

mod246:

MOHANDAS, T., HEINZMANN, C., SPARKES, R. S., WASMUTH, J., EDWARDS, P., AND LUSIAS, A. J. (1986). Assignment of human 3-hydroxy-3-methylyglutaryl coenzyme A reductase gene to q13-q23 region of chromosome 5. Somatic Cell Mol. Genet. 12: 89-94. MORT, J. S., POOLE, A. R., AND ROUGH-Y, P. J. (1983). Agerelated changes in the structure of proteoglycan link proteins present in normal human articular cartilage. Z&&cm. J. 214: 269-272. NAKAMICHI, N. N., KAO, F. T., WASMUTH, J., AND ROUFA, D. J. (1986). Ribosomal protein gene sequences map to human chromosomes 5, 8, and 17. Somatic Cell Mol. Genet. 12: 222-236. NEAME, P. J., CHRISTNER, J. E., AND BAKER, J. R. (1986). The primary structure of link protein from rat chondrosarcoma proteoglycan aggregate. J. Biol. Chem. 261: 3519-3535. PAL, S., STRIDER, W., MARGOLIS, R., GALLO, G., LEE-HUANG, S., AND ROSENBERG, L. (1978). Isolation and characterization of proteoglycans from human chondrosarcomas. J. Biol. Chem. 253: 1279-1289. PERIDES, G., LANE, W. S., ANDREWS, D., DAHL, D., AND BIGNAMI, A. (1989). Isolation and partial characterization of a glial hyaluronate-binding protein. J. Bid. Chem. 264: 59815987. P$RIN, J.-P., BONNET, F., PIWN, V., JOL&S, J., AND JOLL~S, P. (1980). Structural data concerning the link proteins from bovine nasal cartilage proteoglycan complex. FEBS I&t. 119: 333-335.

BYERS, M. G., (1978). Assignthe pter + q22 Cell Genet. 21:

S., LEDBE~R, D. H., BEAUDET, A. L. (1984). 14 arginosuccinate synas reagents for cytoge36: 954-964.