Key points are not available for this paper at this time.
We describe a vertebrate hyaluronan and proteoglycan binding link protein gene family (HAPLN), consisting of four members including cartilage link protein. The encoded proteins share 45–52% overall amino acid identity. In contrast to the average sequence identity between family members, the sequence conservation between vertebrate species was very high. Human and mouse link proteins share 81–96% amino acid sequence identity. Two of the four link protein genes (HAPLN2 and HAPLN4) were restricted in expression to the brain/central nervous system, while one of the four genes (HAPLN3) was widely expressed. Genomic structures revealed that all four HAPLN genes were similar in exon-intron organization and were also similar in genomic organization to the 5' exons for the CSPG core protein genes. Strikingly, all four HAPLN genes were located immediately adjacent to the four CSPG core protein genes creating four pairs of CSPG-HAPLN genes within the mammalian genome. Furthermore, the two brain-specific HAPLN genes (HAPLN2 and HAPLN4) were physically linked to the brain-specific CSPG genes encoding brevican and neurocan, respectively. The tight physical association of the HAPLN and CSPG genes supports a hypothesis that the first HAPLN gene arose as a partial gene duplication event from an ancestral CSPG gene. There is some degree of coordinated expression of each gene pair. Collectively, the four HAPLN genes are expressed by most tissue types, reflecting the fundamental importance of the hyaluronan-dependent extracellular matrix to tissue architecture and function in vertebrate species. Comparison of the genomic structures for the HAPLN, CSPG genes and other members of the link module superfamily provide strong support for a common evolutionary origin from an ancestral gene containing one link module encoding exon. We describe a vertebrate hyaluronan and proteoglycan binding link protein gene family (HAPLN), consisting of four members including cartilage link protein. The encoded proteins share 45–52% overall amino acid identity. In contrast to the average sequence identity between family members, the sequence conservation between vertebrate species was very high. Human and mouse link proteins share 81–96% amino acid sequence identity. Two of the four link protein genes (HAPLN2 and HAPLN4) were restricted in expression to the brain/central nervous system, while one of the four genes (HAPLN3) was widely expressed. Genomic structures revealed that all four HAPLN genes were similar in exon-intron organization and were also similar in genomic organization to the 5' exons for the CSPG core protein genes. Strikingly, all four HAPLN genes were located immediately adjacent to the four CSPG core protein genes creating four pairs of CSPG-HAPLN genes within the mammalian genome. Furthermore, the two brain-specific HAPLN genes (HAPLN2 and HAPLN4) were physically linked to the brain-specific CSPG genes encoding brevican and neurocan, respectively. The tight physical association of the HAPLN and CSPG genes supports a hypothesis that the first HAPLN gene arose as a partial gene duplication event from an ancestral CSPG gene. There is some degree of coordinated expression of each gene pair. Collectively, the four HAPLN genes are expressed by most tissue types, reflecting the fundamental importance of the hyaluronan-dependent extracellular matrix to tissue architecture and function in vertebrate species. Comparison of the genomic structures for the HAPLN, CSPG genes and other members of the link module superfamily provide strong support for a common evolutionary origin from an ancestral gene containing one link module encoding exon. Hyaluronan (HA) 1The abbreviations used are: HA, hyaluronan; ECM, extracellular matrix; PCM, pericellular matrix; CSPG, chondroitin sulfate proteoglycan; CNS, central nervous system; HAPLN, hyaluronan and proteoglycan link; LTP, long term potentiation; EST, expressed sequence tag; PTR, proteoglycan tandem repeat; RACE, rapid amplification of cDNA ends; CRTL1, cartilage link protein 1; MTE, multiple tissue expression. functions at multiple levels within the vertebrate extracellular matrix (ECM) and pericellular matrix (PCM) (1Fraser J.R. Laurent T.C. Laurent U.B. J. Intern. Med. 1997; 242: 27-33Crossref PubMed Scopus (1484) Google Scholar). HA may be found in variable amounts in many connective tissues, where it is usually bound by large aggregating chondroitin sulfate proteoglycans (CSPGs) (1Fraser J.R. Laurent T.C. Laurent U.B. J. Intern. Med. 1997; 242: 27-33Crossref PubMed Scopus (1484) Google Scholar), such as versican and aggrecan, or within the ECMs of the brain and central nervous system (CNS), where it is most often bound by the smaller aggregating CSPGs, brevican or neurocan. Engineered and spontaneous loss-of-function mutations in the versican (2Mjaatvedt C.H. Yamamura H. Capehart A.A. Turner D. Markwald R.R. Dev. Biol. 1998; 202: 56-66Crossref PubMed Scopus (262) Google Scholar) and aggrecan genes (3Watanabe H. Kimata K. Line S. Strong D. Gao L.Y. Kozak C.A. Yamada Y. Nat. Genet. 1994; 7: 154-157Crossref PubMed Scopus (210) Google Scholar), result in mid-gestational and early postnatal lethalities, respectively. This illustrates the fundamental importance of these two CSPGs and the HA-CSPG aggregates to normal development and tissue structure and function. In contrast, mice deficient in neurocan develop in an apparently normal fashion (4Zhou X.H. Brakebusch C. Matthies H. Oohashi T. Hirsch E. Moser M. Krug M. Seidenbecher C.I. Boeckers T.M. Rauch U. Buettner R. Gundelfinger E.D. Fassler R. Mol. Cell. Biol. 2001; 21: 5970-5978Crossref PubMed Scopus (145) Google Scholar), although they do display disturbances in long term potentiation (LTP), which suggests a role for neurocan in the normal electrophysiological function of the brain/CNS. The cartilage proteoglycan aggregate, consisting of HA and aggrecan is dependent upon cartilage link protein for its assembly and stability, with cartilage link protein and aggrecan binding along the HA chain in a 1:1 stoichiometry (5Faltz L.L. Caputo C.B. Kimura J.H. Schrode J. Hascall V.C. J. Biol. Chem. 1979; Google Scholar), i.e. one link protein monomer for each aggrecan monomer. In vitro reassociation studies and in vivo loss-of-function studies have demonstrated the critical role of cartilage link protein in the structure and function of the cartilage proteoglycan aggregate (6Morgelin M. Heinegard D. Engel J. Paulsson M. Biophys. Chem. 1994; 50: 113-128Crossref PubMed Scopus (77) Google Scholar, 7Watanabe H. Yamada Y. Nat. Genet. 1999; 21: 225-229Crossref PubMed Scopus (159) Google Scholar). Mice deficient in cartilage link protein had a phenotype that was essentially a phenocopy of the spontaneous mouse mutant, cmd, which results from a 7-bp deletion within the aggrecan gene (7Watanabe H. Yamada Y. Nat. Genet. 1999; 21: 225-229Crossref PubMed Scopus (159) Google Scholar, 3Watanabe H. Kimata K. Line S. Strong D. Gao L.Y. Kozak C.A. Yamada Y. Nat. Genet. 1994; 7: 154-157Crossref PubMed Scopus (210) Google Scholar). The name cartilage link protein may be somewhat of a misnomer as this protein has been reported to be expressed or present in many other locations within both the developing mammalian and vertebrate embryo and the adult (8Binette F. Cravens J. Kahoussi B. Haudenschild D.R. Goetinck P.F. J. Biol. Chem. 1994; 269: 19116-19122Abstract Full Text PDF PubMed Google Scholar, 9Kobayashi H. Sun G.W. Hirashima Y. Terao T. Endocrinology. 1999; 140: 3835-3842Crossref PubMed Google Scholar). Despite this, the phenotype associated with loss-of-function of cartilage link protein was restricted to the skeleton (7Watanabe H. Yamada Y. Nat. Genet. 1999; 21: 225-229Crossref PubMed Scopus (159) Google Scholar). This might suggest that, while cartilage link protein is expressed elsewhere, it is essential only to the stability and function of the cartilage proteoglycan aggregate and is not necessary to stabilize the HA-CSPG aggregates of other tissues, such as the brain and CNS. The recently described brain link protein, encoded for by the brain link protein 1 (BRAL1) gene (10Hirakawa S. Oohashi T. Su W.D. Yoshioka H. Murakami T. Arata J. Ninomiya Y. Biochem. Biophys. Res. Commun. 2000; 276: 982-989Crossref PubMed Scopus (64) Google Scholar), is clearly related to but distinct from the cartilage link protein in both amino acid sequence, predicted structure, and gene organization. This suggests that a second link protein may play a role in organizing and stabilizing the HA-CSPG aggregates of the brain and CNS. HA-CSPG aggregates do not act in isolation within the ECM and PCM. These aggregates can interact with numerous other ECM and cell surface components through binding interactions of the CSPG core protein, the chondroitin sulfate and/or the HA chain. Binding partners that have been identified to date, include HA receptors such as CD44 (11Lesley J. Hyman R. Front. Biosci. 1998; 3: D616-D630Crossref PubMed Scopus (143) Google Scholar), EGF receptors (12Zhang Y. Cao L. Yang B.L. Yang B.B. J. Biol. Chem. 1998; 273: 21342-21351Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar), sulfated glycolipids (13Miura R. Aspberg A. Ethell K. E. Y. J. Biol. Chem. 1999; Full Text Full Text PDF PubMed Scopus Google Scholar), A. R. S. M. Heinegard D. M. E. Y. U. S. A. 1997; PubMed Scopus Google Scholar, U. K. X.H. Cell. Mol. 2001; PubMed Scopus Google Scholar), M. T. R. Heinegard D. Aspberg A. J. Biol. Chem. 2001; 276: Full Text Full Text PDF PubMed Scopus Google Scholar), and cell U. K. X.H. Cell. Mol. 2001; PubMed Scopus Google Scholar, D.R. L. M. J. Biol. 1994; PubMed Scopus Google Scholar). the HA-CSPG aggregate, although present at levels in many tissues, may provide an within the ECM and which an matrix can be the components that HA-CSPG HA, a large aggregating CSPG protein, and a link protein, be to be critical to the organization and function of the ECM and of many by a of the HA-CSPG aggregate may act to or such as in to the of many to genes encoding identified multiple expressed sequence encoding predicted link of these partial that they four genes and are predicted to link to cartilage link protein. We the first of the vertebrate link protein gene family and suggest a evolutionary for the link module and the and the cartilage link protein amino acid sequence as and the were that were were to amino acid The from were used to the in an to the cDNA at the and cDNA were used to the sequence of in an to the for each in to of the gene Genomic structures for all members of the link module superfamily were also from the of and the T. D. E. Y. L. T. J. T. R. E. J. M. L. A. H. C. E. R. M. S. A. E. S. J. A. J. E. A. M. Res. PubMed Scopus Google Scholar). acid sequence were the with to and were used to and exon-intron from genomic were used as a to for each of the link protein from both and The of were by were to the and to the and with cDNA from and brain by the of the were and of were were also to the predicted and to the of for each link protein gene. to the of the and mouse genes were to the of these this an cDNA from brain for were with as described B. Biochem. PubMed Scopus Google Scholar), and to the at in with were for at in to at with two were identified and from the The were by of The predicted cDNA for was that in this were for for each HAPLN multiple were or were both and to aggrecan, and neurocan, and were from or the as or were in and the identity of each were from each and each cDNA was as described and to or at in with by the were and at as described had been to they were by for in by to and with the The was to in Human was used as a the and The of were for at to were for between Two were for each of a to the of multiple cDNA encoding predicted related to cartilage link protein. were identified in mouse and In and and the two recently described and A. K. S. C. J. S. S. Biochem. J. PubMed Scopus Google Scholar), were identified in to numerous encoding most related to cartilage link protein These be four to and link In only were and were HAPLN and proteoglycan link where was of the HAPLN has been reported as (10Hirakawa S. Oohashi T. Su W.D. Yoshioka H. Murakami T. Arata J. Ninomiya Y. Biochem. Biophys. Res. Commun. 2000; 276: 982-989Crossref PubMed Scopus (64) Google Scholar). In a partial sequence for has also been reported as E. J. 1999; Full Text Full Text PDF PubMed Scopus Google Scholar). In the four of link protein sequence were which and a four members were also identified in the and partial four genes were identified in the and and This suggest that the HAPLN gene family is present in all at as as the We first from a of the also mouse with the predicted cDNA sequence multiple that to the of The sequence of each was and a cDNA was in to cDNA for of the genomic the of only four link proteins in and The four link proteins share between and amino acid with overall of The between and mouse HAPLN proteins are: and This suggest that may be between the four link in for with one or CSPG and with four link proteins are in the with an sequence by an and two link or proteoglycan tandem The and the share the overall sequence identity. link protein to aggrecan the D. Hascall V.C. J. Biol. Chem. Full Text PDF PubMed Google Scholar). levels of sequence identity between the suggest that may be of of the HAPLN proteins with the CSPGs are the as are the of that have been or predicted to be for HA binding in other link module superfamily members J. Biol. Chem. 2001; 276: Full Text Full Text PDF PubMed Scopus Google Scholar). from the other members of the it a acid within the first and has an which is in the amino and acid amino are are upon the amino acid sequence, all four HAPLN proteins are predicted and (BRAL1) are to to HA T. S. Y. Rauch U. D.R. Su W.D. A. Murakami T. Ninomiya Y. Mol. Cell. PubMed Scopus Google Scholar). The of the is not at this one for while and do not from to for the four HAPLN expression was by with a of from this many in ECM, it was in a to the of expression of each results that is the most widely expressed of all the HAPLN with a of found in most tissues, including the The levels of expression were in and was expressed in a restricted expressed by the and In contrast, and were essentially restricted to the brain and CNS. of was in of was also only in are by which that all of the and were from not the HAPLN gene family one that is widely expressed two members that are essentially (HAPLN2 and HAPLN4) and a that is expressed in a restricted of adult Collectively, HAPLN genes are expressed by most Genomic used the first of the sequence Human 2001; PubMed Scopus Google Scholar, M. C.A. J.R. X.H. L. M. J. C. S. J. M. C. M. R. A. D. M. L. A. S. S. S. C. K. K. J. E. K. R. M. R. K. K. C. F. R.R. Y. J. Y. F. B. D. S. B. Sun J. A. J. M. R. C. C. A. J. M. H. L. F. S. S. D. S. C. A. T. F. H. A. D. H. M. K. D. A. A. L. S. L. R. S. K. L. S. A. B. J. C. C. S. D. J. T. C. J. F. L. S. A. F. D. S. T. M. L. B. K. C. E. H. M. R. D. B. R. C. M. E. Strong R. E. R. S. C. J. S. M. S. E. K. J. K. R. B. A. H. B. T. A. K. A. S. S. R. R. B. S. D. A. J. L. M. J. M. C. A. M. M. D. S. C. H. S. K. A. M. K. B. M. J. S. J. D. C. J. J. L. C. A. M. J. D. J. S. M. T. M. S. J. M. R. J. M. T. A. T. Turner R. E. M. M. D. M. A. A. 2001; PubMed Scopus Google Scholar) in an to the genomic structures for each of the HAPLN genes. We identified all exon-intron for each HAPLN gene and the gene structures for and with the described structures for and the four large aggregating chondroitin sulfate proteoglycans versican brevican aggrecan and neurocan and are of and respectively. members of the HAPLN gene family are in a similar with the and two encoded The first of is two by a between the and the between the and and between and were in all HAPLN genes. Furthermore, these exon-intron were also and with the four CSPG genes This conservation of gene structure supports the of the HAPLN and the large aggregating CSPG core protein genes from a common ancestral gene. We gene structure of the HAPLN and CSPG genes to include all the members of the link module including and K. A. H. Res. 1998; PubMed Scopus Google Scholar). These members each a We identified the exon-intron the of each gene and the of the within the for each gene The exon-intron immediately is in all members of the link module Furthermore, this exon-intron is of the in each family The exon-intron immediately the is also of this and the of this is for and this is by in the family this encoding can be in or of each link module superfamily while the this be by genes and an creating a protein. results support the of the of the link module superfamily from a common also and also suggest that the link module superfamily share one common to the first of the HAPLN and CSPG genes and the of the family This support the hypothesis that all members of the link module superfamily were from a ancestral gene containing a link module sequence through a of gene duplication or and structure and amino acid sequence a of with the two that these genes share a The evolutionary between and other members of the not be with to the gene structures for the HAPLN that exons were predicted immediately of the gene. revealed that these exons to the exons for aggrecan The for the and the genes are only of the revealed that each HAPLN gene was with one of the four large aggregating chondroitin sulfate proteoglycan genes. The of and has been recently described H. Oohashi T. S. Y. H. Ninomiya Y. Med. Google Scholar). The four gene pairs and of the four pairs were in the with the two genes each other in a The gene was in a with pairs of the gene was by this not and it that this of the is by and most of the genes the four gene pairs that the gene pairs are of a of genes This suggests that the four gene pairs arose through a duplication one of the four genes was associated with each although the and of each gene was in to the gene two other and were located in the to and with genes present in the in to and This suggest that the and gene are related and arose through a The genes were also linked to the gene with and with In the is physically linked with the gene with an event from in The overall organization of each gene is with two large duplication by multiple This is with vertebrate S. Genet. Dev. PubMed Scopus Google Scholar, R. Res. 2001; PubMed Scopus Google Scholar). that two duplication early in the of the vertebrate and were by multiple and or that multiple smaller have by and upon the organization of the gene the most duplication are to have the and the pairs the expression of each gene in multiple to a in to the linked gene pairs of at the of In upon of the and gene pairs as these two gene pairs are by similar amounts of genomic while the of tissue most it many the results were used as a to of or and are not and versican were not in in adult not and neurocan, were restricted to the and and were although was not expressed as as and not and aggrecan were not in a aggrecan was restricted to the and the while was widely expressed. levels of expression for were in all of the the and and immediately the gene pairs were not with the HAPLN and/or CSPG genes not The functions of extracellular are dependent upon multiple interactions between of the interactions that is to the and support the functions of the ECM, is the binding between the large chondroitin sulfate proteoglycans and HA, to a This aggregate is dependent upon an protein, link protein. is the binding of link protein to both the proteoglycan and the HA that and the the cartilage proteoglycan aggregate, of HA, aggrecan, and cartilage link protein, is the most aggregate, HA and related proteoglycans are found in most This suggests that similar aggregates are to the ECM of many In the brain ECM is a HA of HA two in to aggrecan and versican Y. Cell. Mol. Scopus Google and A. F. Y. S. S. Biochem. Biophys. 2000; PubMed Scopus Google Scholar). of the proteins that to HA at the cell surface or within the ECM, do through a common the link or link module J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). This module is by two and is of two and two a large similar in structure to the a J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar, D. A.A. H. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar). The link module containing proteins the link module superfamily J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). date, members of this superfamily have only been identified within the vertebrate has also been that the link module superfamily from a common ancestral protein with a D. A.A. H. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar) within a exon. The link module within each superfamily is most often within one or two We have the link module superfamily through of a link protein gene family of four members, including cartilage link protein We have this gene HAPLN and proteoglycan link protein where Collectively, these four link proteins are expressed by most tissue types, and are in the brain/CNS. two of the four link proteins (HAPLN2 and HAPLN4) are is the most widely expressed link protein and to play an role in the organization and stability of the ECM of many We that in the and of and to the mouse phenotype to the developing The of sequence identity within the HAPLN family and species suggest that may be or in the of each HAPLN with one or of the The of sequence identity within the family to the and the is the of cartilage link protein that to the proteoglycan D. Hascall V.C. J. Biol. Chem. Full Text PDF PubMed Google Scholar). We that each link protein may stabilize aggregates of where one or CSPG or HAPLN are may be through the in between HAPLN proteins and all four HAPLN proteins may be to interact with all four CSPGs, Furthermore, are The of HAPLN proteins to with CSPGs and the of these be an of and may provide with an of the and functions of the encoding components of protein are in We have that the four HAPLN genes and the four CSPG genes are four gene pairs within the vertebrate with each HAPLN gene located adjacent to a CSPG gene. This is one of the first of a gene family of this where physically linked genes components of a protein We that this might be by of each gene the most HAPLN and CSPG This is the two restricted HAPLN genes are physically linked to the two restricted CSPG brevican and neurocan. and neurocan are in a in the adult and versican and and aggrecan are not in the adult be to the and of each HAPLN and CSPG protein within the embryo and adult at the for in and In this has been recently to with a of versican within the brain T. S. Y. Rauch U. D.R. Su W.D. A. Murakami T. Ninomiya Y. Mol. Cell. PubMed Scopus Google Scholar). The organization of the genes four gene the origin of the HAPLN genes and the other members of the link module superfamily We that the first HAPLN gene arose from an ancestral gene a gene duplication and this the that the ancestral gene was similar to the HAPLN gene to the We that the first link gene in an the common consisting of one or two exons is found in all the link module superfamily members We that the ancestral link module gene a within one exon. This ancestral gene to two The two to link module superfamily members containing a link module or proteoglycan tandem and members containing two tandem We to the as the and from the gene duplication and the through an gene duplication which in the of tandem link an duplication and event the first gene pair. We that these the vertebrate In this have recently identified a CSPG gene in the J. and A. in This gene exon-intron with its vertebrate and the amino acid sequence identity to most of the cell and tissue that with a HA matrix in and it be to the components of the ECM in this has been that two duplication early in vertebrate evolutionary S. Genet. Dev. PubMed Scopus Google Scholar). The hypothesis that multiple smaller have R. Res. 2001; PubMed Scopus Google Scholar), which have had the of many gene in the with In the origin of the the first duplication event to have been by of one of the gene such that its expression restricted to We that this event in an early The second gene duplication event the four gene as is that a event to This gene organization is in both and We that two similar duplication in the vertebrate although of the between and has not been to at this We that the first duplication event the CD44 and The second event CD44 and or and and a gene. duplication are to have in the two genes and the CD44 and genes. may a HA although a cDNA has not been reported to the is by HA while the is by ECM The of a HAPLN gene family and the expression of its members suggests that aggregates are present in most vertebrate The of each aggregate is not at this but it is to that each HAPLN protein may have in its to to CSPGs and to stabilize the We that each HAPLN protein functions in a to cartilage link protein, but In the of the of is not the expression and the in vitro and in vivo functions of each HAPLN protein, to the function of the ECM, and the importance of the ECM to normal development and We at the of for the sequence of link in to this
Spicer et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: