Organization of the Human IMPG2 Gene and Its Evaluation As a Candidate Gene in Age-Related Macular Degeneration and Other Retinal Degenerative Disorders

Organization of the Human IMPG2 Gene and Its Evaluation as a Candidate Gene in Age-Related Macular Degeneration and Other Retinal Degenerative Disorders

Markus H. Kuehn, Edwin M. Stone, and Gregory S. Hageman

1 PURPOSE. To characterize the genomic organization of human retinal photoreceptor outer segments and ellipsoids. Other IMPG2, the gene encoding the retinal interphotoreceptor ma- reports suggest that IPM 200 may also be expressed in non- trix (IPM) proteoglycan IPM 200, to evaluate its relationship to ocular tissues, including the brain.6,7 IPM 150, and to evaluate its involvement in inherited retinop- The IPM is crucial for normal function and viability of retinal athies, such as age-related macular degeneration, retinitis pig- photoreceptor cells. It is a likely participant in the exchange of mentosa, and Leber congenital amaurosis. metabolites and catabolic byproducts between the retinal pig- METHODS. After isolation of human genomic clones, the structure ment epithelium and photoreceptor cells, the regulation of the subretinal ionic milieu, and the orientation, polarization, and of IMPG2 was determined by sequence analysis. Mutational anal- 8–12 yses were conducted on genomic DNA isolated from 316 pro- turnover of photoreceptor outer segments. IPM proteogly- bands using single-strand conformation polymorphism analysis. cans have been shown to mediate photoreceptor cell adhesion.13–16 IPM 200 and IPM 150 may also mediate photorecep- RESULTS. The IMPG2 gene is organized into 19 exons, and the tor cell survival by sequestration of growth factors10,17 or structure of the gene is highly similar to that of the IMPG1 through the epidermal growth factor (EGF)-like domains con- gene, which encodes another retinal proteoglycan, IPM 150. tained within their core proteins.2,3,5 Mutational analyses indicate that the observed sequence Photoreceptor cells are highly vulnerable to dysfunction changes are present at approximately equal rates in donors and/or death in various heritable retinal dystrophies and de- with and without retinal disease. Additional data derived from generations (reviewed in Ref. 18). Nucleotide sequence varia- RT-PCR and Northern blot analysis show that IMPG2 is pro- tions in the genes encoding a number of retinal proteins are cessed in the human retina into multiple alternatively sized associated with the etiologies of various forms of retinal de- transcripts that may represent splicing isoforms. generation. For example, mutations in the genes encoding CONCLUSIONS. Analysis of the overall relationship of human retinal rhodopsin, ␤-phosphodiesterase, rab geranylgeranyl IMPG2 (located on chromosome 3q12.2-12.3) to human transferase, rim protein, and the RP1 gene product cause reti- IMPG1 (located on chromosome 6q14) suggests that these nal degeneration.19–23 genes have evolved from a common ancestral gene. Although Because IPM 200 is expressed at high levels by retinal this is an excellent candidate gene for hereditary retinopathies, photoreceptor cells and probably plays a critical role in the single-strand conformation polymorphism analyses provided maintenance of the interphotoreceptor space, it is reasonable no evidence that variations in IMPG2 coding region are respon- to postulate that sequence variations within its gene, IMPG2, sible for the inherited retinopathies examined. (Invest Oph- may cause photoreceptor cell dysfunction and/or retinal de- thalmol Vis Sci. 2001;42:3123–3129) generation. The IMPG2 gene has been mapped to chromosome 3q12.2-12.3 between markers WI3277 and NIB1880.2 Although no human hereditary diseases have yet been mapped In previous studies, we have identified and characterized two to this interval, IMPG2 is a strong candidate gene for un- novel human interphotoreceptor matrix (IPM) proteoglycans, mapped inherited ocular, or neuronal, disorders, including designated IPM 2001,2 and IPM 1501,3–5 that, together with age-related macular degeneration (AMD). their associated isoforms, comprise a unique family of extra- AMD is a significant cause of irreversible blindness world- cellular proteins. These molecules were initially identified as wide. There is strong evidence that a significant proportion of constituents of the IPM, an extracellular matrix that surrounds AMD has a genetic foundation,24–27 and several AMD loci have been identified.28,29 In addition, the ApoE4 allele has been shown to be protective for the disease.30 In this study we screened DNA from patients with abnormalities on both sides From the Department of Ophthalmology and Visual Sciences, The of the photoreceptor cell–retinal pigment epithelium interface University of Iowa Center for Macular Degeneration, Iowa City. Supported in part by National Eye Institute Grants EY06463 with a wide range in age of onset, from birth to the ninth (GSH), EY11515 (GSH), and EY10539 (EMS); The Roy C. Carver En- decade of life, for mutations in IMPG2. These afflictions in- dowment for Molecular Ophthalmology (EMS); a Research to Prevent clude AMD, retinitis pigmentosa (RP), and Leber congenital Blindness Lew R. Wasserman Merit Award (GSH); a National Research amaurosis (LCA)—three genetically heterogeneous retinal dis- Service Award (MHK); and an unrestricted grant to the Department of eases characterized by photoreceptor cell death.31–33 Ophthalmology and Visual Sciences, University of Iowa, from Research to Prevent Blindness, Inc. Submitted for publication April 6, 2001; revised July 30, 2001; accepted August 14, 2001. MATERIALS AND METHODS Commercial relationships policy: P. The publication costs of this article were defrayed in part by page Identification and Characterization charge payment. This article must therefore be marked “advertise- of Genomic Subclones ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact. Corresponding author: Gregory S. Hageman, The University of To identify bacterial artificial chromosome (BAC) clones that contain Iowa Center for Macular Degeneration, Department of Ophthalmology portions of the IMPG2 gene, several PCR primer pairs were generated and Visual Sciences, The University of Iowa, 11190E PFP, 200 Hawkins based on the IPM 200 cDNA sequence (GenBank accession no. Drive, Iowa City, IA 52242. [email protected] AF173155; GenBank is provided by the National Center for Biotech-

Downloaded from iovs.arvojournals.org on 10/01/2021 3124 Kuehn et al. IOVS, December 2001, Vol. 42, No. 13

NJ) without further purification. After transformation by electropora- tion, Escherichia coli TOP 10 cells were grown overnight, at 37°C, on Luria-Bertani (LB) broth–based agar plates containing carbenicillin (50 ␮g/ml). From each restriction digest, 48 subclones were randomly selected and arrayed into 96-well microtiter plates. To identify subclones containing specific regions of IMPG2, nitrocellulose membranes were placed on LB-agar plates containing carbenicillin, and a small number of cells from each subclone were transferred to FIGURE 1. Organization of the IMPG2 gene and the BAC contig used establish colonies directly on the membranes. Colonies were grown to determine it. Broken lines: intronic regions that were not com- overnight at 37°C. DNA bound to the filter was then denatured by pletely sequenced. incubation in 0.5 M NaOH and 1.5 M NaCl for 5 minutes, neutralized in 1 M Tris (pH 8.0) and 1.5 M NaCl for 5 minutes, briefly rinsed in nology Information and available in the public domain at http:// 2ϫ SSC, and cross-linked to the membranes using UV irradiation. www.ncbi.nlm.nih.gov/genbank). Two of these primer pairs (sense 1: The filters were then incubated overnight in hybridization buffer 32 5Ј-AAA AAG AAA CAG CCT CTG GAC CGC AG-3Ј and antisense 1: containing [P]-labeled oligonucleotides that were designed based 5Ј-CAG CCT CTG CAA CAC TTT CAT CTG GG-3Ј, spanning nucleotides on the human IPM 200 cDNA sequence. After removal of unbound 372 to 492 of the IPM 200 cDNA; sense 2: 5Ј-TCA TTC ACT CAA CCT probe, filters were exposed to x-ray film to identify subclones GTG C-3Ј and antisense 2: 5Ј-GAC CCT GAA CCT AAA CCA C-3Ј, yielding hybridization signals. spanning nucleotides 1794 to 2005 of the IPM 200 cDNA) consistently Plasmids were isolated from these colonies and sequenced. Exonic yielded PCR amplification products of the expected size when human domains were determined by comparison of the obtained genomic genomic DNA was used as a template. These primer pairs were used to sequences to that of the cDNA sequence. screen a commercially available human BAC library (Genome Systems, St. Louis, MO). BAC clones 340M10, 366H07, 325H22, and 493P03 Human Subjects appeared to contain portions of the IMPG2 gene. The human genomic DNA was isolated from these BAC clones and fragmented by digestion The study included 92 individuals with AMD, 92 with RP, 40 with LCA, with the restriction endonucleases HindIII, SacI, or EcoRI. The result- and 92 normal individuals (control subjects). The control subjects ant fragments were ligated into either dephosphorylated vector pBlue- were between the ages of 44 and 93 and were not afflicted with any script SK (Stratagene, La Jolla, CA) or pClonesure (CPG Inc., Fairfield, ocular diseases. All probands with AMD and all control subjects were

TABLE 1. PCR Primers used during SSCP Analyses of the IMPG2 Gene

Size (Exon Sense Primer (5؅ to 3؅) Antisense Primer (5؅ to 3؅) (bp

1-1 TTATCTCACCAGCTTTTATAGCA TGTCCAAATCCTTGAAACTTCC 192 1-2 GGTTGTTCATTCTCAAACATAGA TATGGAAAGATACAAACAAATG 256 2-1 TTTTGTACTGTATCTTCATTATCG TTTCAGTTTCTCTGCGGTCC 185 2-2 TCCTGCCTGAAGAATCAACA GGAGCTGAAGGATTTGGATG 201 3 CCCCAGAGCATGTTAGCTTT CCAGGAATCCCTTCCTTTGT 247 4 GGGCAGTAGTGGTCTCTATG TTAGGCTATGACACATCTGTG 179 5 CCTTTTTGGAAAACAACCCC GTGAGCCTGTCTTAAACC 197 6 TCATGCATATGTTTTTGCTTTTC TGTGTAGTCCAGCAATGGGA 199 7 ATTGAATGAATAAGCCTTGACA GGCATACTGCCTTGTTTGTT 249 8 CAGTTCACTTTTTATTCTACTCTT GGACATTCCATTCAGAATAAAG 218 9 AATAATAACTGTCTCAAACTCTG GGACCTACGGCCTGCTATATT 145 10-1 GCTCCTTTCTTTGTGCTTCC CAACAGTGGGTTTATCATCCAG 191 10-2 CCAACAAGGTGGAAAACCAT ACCAGAGCATACTGGAAAAGA 223 11 GTTGTCCCTGCACCTCAAAT GAGGGCCTGGTTCTAGCATA 190 12-1 GGAATATAGACAGATAGGTGG GAAAGGCTAATTTGTGTGTAGA 249 12-2 CAGGGAACTCTGGTCAGAAAG ATACAATAAGAAGTACGAAAA 246 13-1 GGAATACATTTTGGCAACTCTGT GAGGTCAGATATGGTGAAGATG 199 13-2 CTCAACCTGTGCCAAAAGAA TGAACCTAAACCACCGTCAA 199 13-3 AAAAGTGAGCCCTTTCCTGC ATCTCAGCTGGCAAAAGTGAA 191 13-4 ACTTGGCCATGGAGTGAGAC GTATCTGCGAAGATGGGCAC 241 13-5 CACTTTCCAGAGGAAGAGCC TCAAACCATTCATAGTTGGATGA 256 13-6 TGCCATCCTAAGGGAGGATA GGTTGGAGGCAATTTGGTAG 243 13-7 AGTGCTGACAGGCTCTGGTT GCCAAGCCACACTAACCATC 237 13-8 GGTTGGTAGTTATGTGGAAATG GTGCCCATGTTTCACTTTTT 249 14-1 TGAACAAAATGTGACACGCTG CCAGAATCATGTACACCGCA 199 14-2 AAGTTTGCCAATTCTGTCCC ATGAGGAAGAAAACACACAAG 199 15-1 TGTGTTTTGCCTTTTCTTGC TCAGGCTGTAGGTCACAGAGA 197 15-2 TGCTTCCCTGGATACCTGAG AGGCCCCTTTTCTTTAGAGG 186 16-1 TGTCAGTGTGAGCAGAGTGATT TGAAGAAGTAGATGATAGCAG 194 16-2 GCAAGCACTGTGAGGAATTT TGTGCTCTTTCTTATTTACCTGA 175 17-1 CCAAACAAGAGCCTGAATCC CCGCTAGCAGAGCTGTAGAA 178 17-2 ACAGGGCTGGATGTGAGAAG AGCTGCGACAGCAACATAAA 171 18 TGCAATGTGTGGCCTATTGT TGCTCACTCAGGTGTGACATT 195 19 GCTCTTTGTGTTTTCATGG ATCTCCATCTTCTCCAGGC 84

Several exons required the use of multiple primer pairs due to the length of the exon; these are designated as 1-1, 1-2, etc.

Downloaded from iovs.arvojournals.org on 10/01/2021 IOVS, December 2001, Vol. 42, No. 13 Characterization of the Human IMPG2 Gene 3125

TABLE 2. Exon–Intron Junctions of the Human IMPG2 Gene

Exon Length Intron Exon Acceptor Site (bp) Donor Site Length*

1 TAACAGGTATTTAAAA 454 bp 2 CACTCTGTAGCACAAA 249 TCCGAGGTAAGCGAAC Ͼ1.1kb 3 TATTTTCTAGTGTGTC 167 ATGAAGGTAAGTGTCA Ͼ1.2kb 4 TTTTCCTTAGAAACTG 32 AAGCAGGTGAGTGTCT Ͼ1.3kb 5 TTTTTTCTAGCTCTGA 50 TGGGAGGTATACTTTT 919 bp 6 TTCTTCTTAGACACTA 83 GAGAGTGTGAGTGATA Ͼ1.0kb 7 AAAAATTCAGATTAGC 162 TCAGAGGTGGGTGATT Ͼ730 bp 8 CTTTTAACAGGTTGAA 59 ATTTAGGTAAATAGAC Ͼ1.1kb 9 TTGTTTTCAGGTCCCC 21 TGACAGGTACTTTTTG Ͼ640 bp 10 CCATCCATAGTGGCGT 245 TCAATGGTGAGTTTGA Ͼ780 bp 11 CTTTCCTCAGTGAGAG 86 ATTCTGGTATGTTTTT Ͼ730 bp 12 TTCCCCCCAGGATAAT 304 AAGATGGTGAGAAACT Ͼ720 bp 13 TTATTTTTAGGATTAG 1259 GAATTGGTAAGCATAA Ͼ880 bp 14 TATATTGTAGCTGGTT 220 AATCAGGTATGATATT Ͼ790 bp 15 AATGTTTTAGGTGATG 211 TTGTAGGTATGTTGTA Ͼ1.0kb 16 TTGGTTGCAGGTGCCG 189 CTTCAGGTAAATAAGA 579 bp 17 TTTCTTTTAGTGGCTC 211 AGAGAGGTGGGAAACT Ͼ480 bp 18 TTTGTTTAAGGAAATT 80 ACAAGTGTAAGCTTTT Ͼ1.0kb 19 TTCATGGCAGGGAAGA

Residues in bold type represent exonic sequences * Base pairs or kilobites (kb).

ascertained at The University of Iowa Hospitals and Clinics, whereas RT-PCR Analyses portions of the RP and LCA groups were ascertained at other centers. Total retinal RNA was isolated from human donor tissue within 4 hours The protocol was in compliance with the tenets of the Declaration of after death using spin columns (RNeasy; Qiagen, Valencia, CA). The Helsinki. RNA was reverse transcribed using random hexamer primers and Single-Strand Conformation reverse transcriptase (SuperscriptII; Gibco BRL, Grand Island, NY). Polymorphism Analyses Fifty nanograms of the resultant single-stranded cDNA was PCR amplified using primers designed based on the human IPM 200 cDNA Genomic DNA from each study participant was screened for sequence sequence. The derived PCR fragments were analyzed by electro- variations in the IPM 200 coding sequence by single-strand conforma- phoretic separation on agarose gels. The primers used to amplify the tion polymorphism (SSCP) analysis. PCR amplification reactions (10 ␮l) open reading frame of human IPM 200 were sense, 5Ј-TTGGAAGTTT contained 5 ng human genomic DNA, 10 ng each PCR primer, 2.5 mM CAAGGATTTG-3Ј, and antisense, 5Ј-AACACAGCAT TCAGTCTTTA MgCl2, 0.25 U Taq polymerase, and 1ϫ Taq polymerase buffer. Occa- TAG-3Ј. The expected 4017-bp amplification product of spans between sionally, reaction conditions were varied slightly to achieve optimal bp 120 and bp 4135 (exons 1 and 19, respectively) of the previously amplification when using specific primer pairs (Table 1). These varia- published IPM 200 cDNA sequence (GenBank accession no. tions included the addition of 10% dimethyl sulfoxide (DMSO), the use AF173155). of 1.5 mM MgCl2 instead of 2.5 mM MgCl2, or the use of “touchdown” PCR. All samples generally underwent 35 cycles of amplification and were then diluted with one volume of sample buffer (95% formamide, RESULTS 20 mM EDTA, 0.05% xylene cyanol green and 0.05% bromophenol Genomic Organization of the IMPG2 Gene blue). Samples were heat denatured, electrophoresed on 6% nondena- turing polyacrylamide gels and visualized by silver staining, as de- The intron–exon boundaries of the IMPG2 gene and por- scribed previously.34 Samples of exons exhibiting band shifts were tions of the intronic sequences flanking the exons were amplified again, purified, and sequenced bidirectionally on an auto- determined by partially sequencing four overlapping BAC mated DNA sequencer (model 377; PE Applied Biosystems, Foster City, clones (Fig. 1). The IMPG2 gene comprises 19 exons rang- CA). ing in size between 21 and 1259 bp. The 5Ј and 3Ј ends of

FIGURE 2. Organization and comparison of the IMPG2 (top) and IMPG1 (bottom) genes and their encoded proteins, IPM 200 and IPM 150. Black boxes: regions of high sequence homology; darkly shaded boxes: regions of moderate homology; lightly shaded boxes: regions of low sequence homology. Vertical lines: borders between exons; thin horizontal lines: insertion of gaps; ovals: EGF-like domains; diamond: hydrophobic, putative transmembrane domain of IPM 200.

Downloaded from iovs.arvojournals.org on 10/01/2021 3126 Kuehn et al. IOVS, December 2001, Vol. 42, No. 13

FIGURE 3. Alignment of the IPM 200 (top) and IPM 150 (bottom) core proteins. Arrows: exon boundaries.

all exons exhibited sequences that were consistent with (accession numbers AF271363 through AF271379). Summa- consensus acceptor and donor splice sites, respectively (Ta- tion of all nonoverlapping sequences obtained in the course ble 2).3 These sequences have been deposited in GenBank of this study indicated that the gene is at least 31.0 kb in

Downloaded from iovs.arvojournals.org on 10/01/2021 IOVS, December 2001, Vol. 42, No. 13 Characterization of the Human IMPG2 Gene 3127

TABLE 3. Allelic Distribution of the Observed Polymorphic Markers in the Human IMPG2 Gene

Exon AMD RP LCA Control

6A3G intronic 46 42 44 51 13 Thr674Ile 41 35 40 38 15 Pro1013Leu 1.2 0 0 0.8 16 Leu1127Leu 31 27 20 18

Data are percentages.

size. In addition, approximately 700 bp of genomic se- Northern blot analysis of retinal RNA hybridized with IPM 200 quence located immediately upstream of the IMPG2 gene cDNA probes (Fig. 4B). These data indicate the presence of were determined. transcripts of several sizes, suggesting that various splicing isoforms of IPM 200 may exist. To confirm these data and to Comparison of the Organization of IMPG2 rule out that the additional observed bands are due to alterna- and IMPG1 tive initiation of transcripts or polyadenylation, RT-PCR ampli- As reported earlier, comparison of IPM 200 and IPM 150 fication of IPM 200 from human retinal cDNA was performed. cDNAs and their deduced amino acid sequences indicated RT-PCR of the open reading frame of human IPM 200 produced that the two proteins are closely related to one other and the 4.0-kb amplification product predicted from the previously constitute a novel family of glycoproteins.2,5 Analyses of the described cDNA sequence, as well as several smaller PCR genomic organization of their respective genes, IMPG2 and products of approximately 3.7, 3.6, 2.9, and 1.8 kb (Fig. 4A). IMPG1, support these data. Alignment of the human IPM These findings demonstrate that the observed transcripts differ 150 and IPM 200 amino acid sequences revealed that the within the coding region of the cDNA, resulting in all likeli- size, distribution and overall organization of exons is highly hood in the synthesis of several distinct protein isoforms. conserved between the two genes (Fig. 2). Closer analysis indicated that regions of high amino acid sequence conser- DISCUSSION vation correspond directly to regions in which the genomic organization is more stringently preserved. In regions of the In this report, we describe the organization of the human genes that encode the more highly conserved amino- and IMPG2 gene that encodes the prominent retinal proteoglycan carboxyl-terminal regions of the IPM 150 and IPM 200 pro- IPM 200.2,4 Analyses of the IMPG2 gene, located on chromo- teins, the intron–exon boundaries often occur precisely at some 3q12, indicate that it comprises 19 exons that span a the same amino acid (Fig. 3). The conservation of genomic minimum of 31.0 kb. A few sequence variations were identified structure is less stringent in the regions that encode the in the IMPG2 gene after genetic analyses of individuals with central domains of the IPM 150 and IPM 200 proteins. These RP, LCA, or AMD and unaffected control subjects. However, no are the same regions in which the primary structures of the significant partitioning of these polymorphisms between af- proteins are also less conserved. fected and unaffected individuals was detected. Hence, there are no indications that mutations in the coding region of Screening of IMPG2 in Patients with Retinal Disease To assess the potential involvement of the IMPG2 gene in the development of AMD, LCA, and RP, we screened genomic DNA obtained from 224 patients affected with these retinopathies for sequence changes within the exons of this gene. The data obtained were compared with those derived from a group of 92 patients for whom there was no clinical evidence of retinal disease. Three sequence changes were identified in the coding region of the gene (Table 3).6 A silent T3 C transition in the third position of the codon for Leu1127 was observed in approximately 25% of both affected and unaffected individuals. Approximately one third of all evaluated alleles display a C3 T change in exon 13, which induces a Thr674Ile change in the mucin-like domain of IPM 200. A rare C3 T change, resulting in a Pro1013Leu substitution immediately preceding the first EGF-like domain, was observed in approximately 1% of AMD-affected and control individuals. In addition, an intronic A3 G sequence change 10 bp downstream of exon 6 was observed in approximately 45% of all examined alleles. Thus, individuals from both the affected and control groups harbored all observed sequence variations at approximately equal rates, suggesting that the detected base changes rep- resented nondisease-causing polymorphisms. Identification of Alternative Transcripts of Human IPM 200 FIGURE 4. RT-PCR (A) and Northern blot (B) analyses of IPM 200 Several bands running at a molecular weight lower than 6.2 kb expression in the human retina and RPE-choroid (RPE/Ch). Arrows: were observed on extended autoradiographic exposure of alternatively sized transcripts of IPM 200.

Downloaded from iovs.arvojournals.org on 10/01/2021 3128 Kuehn et al. IOVS, December 2001, Vol. 42, No. 13

IMPG2 are involved in the development of these three retinop- a gene expressed in the pineal gland and the outer nuclear layer of athies. However, it is difficult to completely rule out the pos- the retina. Mol Brain Res. 1996;41:269–278. sibility that mutations in this gene may be involved in these 7. Acharya S, Foletta VC, Lee JW, et al. SPACRCAN, a novel human retinal diseases, because certain types of mutations, such as interphotoreceptor matrix hyaluronan-binding proteoglycan syn- genomic rearrangements, cannot be detected by SSCP. In ad- thesized by photoreceptors and pinealocytes. J Biol Chem. 2000; 275:6945–6955. dition, sequence changes in noncoding regions, such as 5Ј 8. Marmor MF, Yao XY. The metabolic dependency of retinal adhe- regulatory elements or splicing branch sites, may severely sion in rabbit and primate. Arch Ophthalmol. 1995;113:232–238. interfere with the functionality of the IMPG2 gene. 9. Bok D, Hageman GS, Steinberg RH. Repair and replacement to The exonic structures of IMPG2 and that of IMPG1, which restore sight: report from the panel on photoreceptor/retinal pig- 3,4,35,36 encodes the related proteoglycan IPM 150, are remark- ment epithelium. Arch Ophthalmol. 1993;111:463–471. ably well conserved based on alignment of the amino acid 10. Hageman GS, Kirchoff-Rempe MA, Lewis GP, Fisher SK, Anderson sequences and insertion of gaps to account for the difference DH. Sequestration of basic fibroblast growth factor in the primate in overall size of the proteins (Fig. 3). Many exons are either of retinal interphotoreceptor matrix. Proc Natl Acad Sci USA. 1991; identical length or terminate in similar locations. These obser- 88:6706–6710. vations indicate that IPM 200 and IPM 150 are members of a 11. Hewitt AT, Adler R,. The retinal pigment epithelium and interpho- single gene family. We speculate that the IMPG1 and IMPG2 toreceptor matrix: structure and specialized function. In: Ogden T, genes may have arisen from the duplication of a single ances- ed. Retina. St. Louis: CV Mosby Co.; 1989:57–64. 12. Lazarus HS, Hageman GS. Xyloside-induced disruption of interpho- tral gene. Thus, it is conceivable that IPM 200 and IPM 150, at toreceptor matrix proteoglycans result in retinal detachment. In- least in part, may be functionally redundant and that retinal vest Ophthalmol Vis Sci. 1992;33:364–375. dysfunction and/or degeneration occurs only if both proteins 13. Yao XY, Hageman GS, Marmor MF. Retinal adhesiveness is weak- are defective. ened by enzymatic modification of the interphotoreceptor matrix Previous reports suggested that there are splicing isoforms in vivo. Invest Ophthalmol Vis Sci. 1990;31:2051–2058. of IPM 150 in the human retina.3,35 Based on the data pre- 14. Hageman GS, Marmor MF, Yao XY, Johnson LV. The interphoto- sented herein, it appears that human retinal IPM 200 is also receptor matrix mediates primate retinal adhesion. Arch Ophthal- synthesized as several isoforms. The nature of these IPM 200 mol. 1995;113:655–660. variants has not been elucidated, although the properties of 15. Marmor M, Yao XF, Hageman G. Retinal adhesiveness in surgically individual isoforms of the same gene can be quite diverse. For enucleated human eyes. Retina. 1994;14:181–186. example, tenascin-C splice variants differ only in the number of 16. Hollyfield JG, Varner HH, Rayborn ME, Osterfeld AM. Retinal at- their fibronectin type III domains, yet some variants possess tachment to the pigment epithelium. Retina. 1989;9:59–68. adhesive properties, whereas others display antiadhesive char- 17. Alberdi E, Hyde CC, Becerra SP. Pigment epithelium-derived factor 37–39 (PEDF) binds to glycosaminoglycans: analysis of the binding site. acteristics. Furthermore, it appears that the functional Biochemistry. 1998;37:10643–10652. differences of the tenascin isoforms are in part dependent on 40 18. Stone J, Maslim J, Valter-Kocsi K, et al. Mechanisms of photore- their environment. By analogy, it is possible that isoforms of ceptor death and survival in mammalian retina. Prog Retinal Eye IPM 200 serve distinct functions. It is also conceivable that Res. 1999;18:689–735. these isoforms are synthesized by specific photoreceptor cell 19. Dryja TP, McGee TL, Reichel E, et al. A point mutation of the types. Future studies should provide additional insight into the rhodopsin gene in one form of retinitis pigmentosa. Nature. 1990; isoforms and functions of IPM 200. 343:364–366. 20. McLaughlin ME, Sandberg MA, Berson EL, Dryja TP. Recessive mutations in the gene encoding the beta-subunit of rod phospho- Acknowledgments diesterase in patients with retinitis pigmentosa. Nat Genet. 1993; 4:130–134. The authors thank Louisa Affatigato and Jake Roos for technical assis- 21. Guillonneau X, Piriev NI, Danciger M, et al. A nonsense mutation tance, Val Sheffield, MD, Ph.D., and his staff for performing sequence in a novel gene is associated with retinitis pigmentosa in a family analyses, and Robert Mullins, Ph.D., for helpful discussions throughout linked to the RP1 locus. Hum Mol Genet. 1999;8:1541–1546. the project. 22. Allikmets R, Singh N, Sun H, et al. A photoreceptor cell-specific ATP-binding transporter gene (ABCR) is mutated in recessive Star- gardt macular dystrophy. Nat Genet. 1997;15:236–246. References 23. Seabra MC, Brown MS, Goldstein JL. Retinal degeneration in choroideremia: deficiency of rab geranylgeranyl transferase. Sci- 1. Hageman GS, Johnson LV, Structure, composition and function of ence. 1993;259:377–381. the retinal interphotoreceptor matrix, In: Osborne N, Chader J, 24. Gorin MB, Breitner JC, De Jong PT, et al. The genetics of age- eds. Progress in Retinal Research. Oxford, UK: Pergamon Press; related macular degeneration. Mol Vis. 1999;5:29; available on-line 1991:207–249. at http//:www.molvis.org/molvis/v5/p29/. 2. Kuehn MH, Hageman GS. Molecular characterization and genomic 25. Heiba IM, Elston RC, Klein BE, Klein R. Sibling correlations and mapping of human IPM 200, a second member of a novel family of segregation analysis of age-related maculopathy: the Beaver Dam proteoglycans. Mol Cell Biol Res Commun. 1999;2:103–110. Eye Study. Genet Epidemiol. 1994;11:51–67. 3. Kuehn MH, Hageman GS. Expression and characterization of the 26. Seddon JM, Ajani UA, Mitchell BD. Familial aggregation of age- IPM 150 gene (IMPG1) product, a novel human photoreceptor related maculopathy. Am J Ophthalmol. 1997;123:199–206. cell-associated chondroitin-sulfate proteoglycan. Matrix Biol. 27. Klein ML, Mauldin WM, Stoumbos VD. Heredity and age-related 1999;18:509–518. macular degeneration: observations in monozygotic twins. Arch 4. Kuehn MH, Hageman GS. Characterization of a cDNA encoding Ophthalmol. 1994;112:932–937. IPM 150, a novel human interphotoreceptor matrix chondroitin 28. Weeks DE, Conley YP, Mah TS, et al. A full genome scan for 6-sulfate proteoglycan [ARVO Abstract]. Invest Ophthalmol Vis age-related maculopathy. Hum Mol Genet. 2000;9:1329–1349. Sci. 1995:36(4):S510. Abstract nr 2788. 29. Klein ML, Schultz DW, Edwards A, et al. Age-related macular 5. Kuehn MH, Wietecki DT, Hageman GS. Molecular characterization degeneration: clinical features in a large family and linkage to of the murine orthologue of the human retinal proteoglycan IPM chromosome 1q. Arch Ophthalmol. 1998;116:1082–1088. 150. Mol Vis. 2000;6:148–156; available on-line at http//:www. 30. Klaver CC, Kliffen M, van Duijn CM, et al. Genetic association of molvis.org/molvis/v6/p148/. apolipoprotein E with age-related macular degeneration. Am J 6. Wang X, Brownstein M, Young W III. Sequence analysis of PG10.2, Hum Genet. 1998;63:200–206.

Downloaded from iovs.arvojournals.org on 10/01/2021 IOVS, December 2001, Vol. 42, No. 13 Characterization of the Human IMPG2 Gene 3129

31. Perrault I, Rozet JM, Gerber S, et al. Leber congenital amaurosis. 36. Kuehn M, Stone E, Hageman G. Molecular analyses of IPM 150, a Mol Genet Metab. 1999;68:200–208. photoreceptor cell-specific proteoglycan. Invest Ophthalmol Vis 32. Lotery AJ, Namperumalsamy P, Jacobson SG, et al. Mutation anal- Sci. 1997;38(4):S599. Abstract nr 2788. ysis of 3 genes in patients with Leber congenital amaurosis. Arch 37. Gotz B, Scholze A, Clement A, et al. Tenascin-C contains distinct Ophthalmol. 2000;118:538–543. adhesive, anti-adhesive, and neurite outgrowth promoting sites for 33. Phelan JK, Bok D. A brief review of retinitis pigmentosa and the neurons. J Cell Biol. 1996;132:681–699. identified retinitis pigmentosa genes. Mol Vis. 2000;6:116–122; 38. Mackie EJ. Molecules in focus: tenascin-C. Int J Biochem Cell Biol. http//:www.molvis.org/molvis/v6/p116/. 1997;29:1133–1137. 34. Nishimura DY, Swiderski RE, Alward WL, et al. The forkhead transcription factor gene FKHL7 is responsible for glaucoma phe- 39. Ruoslahti E. Brain extracellular matrix. Glycobiology. 1996;6:489– notypes which map to 6p25. Nat Genet. 1998;19:140–147. 492. 35. Felbor U, Gehrig A, Sauer C, et al. Genomic organization and 40. Meiners S, Powell EM, Geller HM. Neurite outgrowth pro- chromosomal localization of the interphotoreceptor matrix pro- motion by the alternatively spliced region of tenascin-C is in- teoglycan-1 (IMPG1) gene: a candidate for 6q-linked retinopathies. fluenced by cell-type specific binding. Matrix Biol. 1999;18: Cytogenet Cell Genet. 1998;81:12–17. 75–87.

Downloaded from iovs.arvojournals.org on 10/01/2021