Comparative Medicine Vol 53, No 6 Copyright 2003 December 2003 by the American Association for Laboratory Animal Science Pages 642-648

Fine Mapping of the Circling (cir) on the Distal Portion of Mouse 9

Kyoung In Cho, DVM,1, 2,* Jeong Woong Lee, PhD,1,* Kil Soo Kim, PhD,3 Eun Ju Lee,1 Jun-Gyo Suh, PhD,2 Ho-Joon Lee, PhD,4 Hyun Taek Kim, PhD,5 Sung Hwa Hong, MD,6 Won Ho Chung, MD,6 Kyu Tae Chang, PhD,7 Byung Hwa Hyun, PhD,7 Yang-Seok Oh, PhD,2,† and Zae Young Ryoo, PhD,1,†

Circling mice manifest profound deafness, head-tossing, and bi-directional circling behavior, which they inherit in autosomal recessive manner. Histologic examination of the inner ear reveals abnormalities of the region around the organ of Corti, spiral ganglion neurons, and outer hair cells. A genetic linkage map was constructed for an intraspecific backcross between cir and C57BL/6J mice. The cir gene was mapped to a region between D9Mit116/ D9Mit15 and D9Mit38 on mouse chromosome (Chr) 9. Estimated distances between cir and D9Mit116, and between cir and D9Mit38 were 0.70 ± 0.40 and 0.23 ± 0.23 cM, respectively. Order of the markers was defined as follows: centromere - D9Mit182 - D9Mit51/D9Mit79/ D9Mit310 - D9Mit212/D9Mit184 - D9Mit116/D9Mit15 - cir - D9Mit38 - D9Mit20 - D9Mit243 - D9Mit16 - D9Mit55/D9Mit125 - D9Mit281. On the basis of genetic mapping, we constructed a yeast artifi- cial chromosome (YAC) contig across the cir region. The cir gene is located between the lactotransferrin (ltf) and microtubule-associated (map4) . The distal portion of mouse Chr 9 encompassing the cir region is homologous with human chromosome 3p21, which contains the Deafness, form B: Autosomal Recessive Deafness (DFNB6) locus. Therefore, the circling mouse is a potential animal model for DFNB6 deafness in humans.

Over the past several years, the mouse has been used as a Table 1. Cloned genes for human nonsyndromic hearing impairment * model and, as such, has made substantial contributions toward (DFNB) our understanding of human developmental disorders. Muta- Gene Locus tions at different loci in humans and mice are known to cause Connexin 26 gene (GJB2) DFNB1 (13, 15) hearing impairment (5, 8, 13, 19, 29, 35, 36). In 70% of human Myosin 7A gene (MYO7A) DFNB2 (13, 35) cases, deafness is the sole clinical feature and is not accompa- Myosin 15 gene (MYO15) DFNB3 (10, 34) Pendred gene (PDS) DFNB4 (5, 29) nied by the broad clinical signs of disease that characterize the Transmembrane serine protease gene (TMPRSS3) DFNB8, DFNB10 (27) remaining 30% of syndromic deafness (21, 33). In recent years, Otoferlin gene (OTOF) DFNB9 (36) Cadherin-like gene (CDH23) DFNB12 (8) there has been a marked increase in the localization of genes for Alpha-tectorin gene (TECTA) DFNB21 (23) autosomal dominant and autosomal recessive nonsyndromic Claudin-14 gene (CLDN14) DFNB29 (37) hearing impairment (31). *Full name of each gene and its abbreviation are given, along with related DFNB Approximately one in every 1,000 individuals becomes deaf form. before adulthood, and about half of these cases are attributable to genetic defects of autosomal recessive deafness (DFNB). Al- cally, circling mice have cochlea degeneration and reduced cellu- though more than 30 DFNB loci have been mapped, only nine larity in the spiral limbus (7, 18). Results of an auditory test genes have been identified (31, Table 1). clearly documented hearing loss in the circling mouse (18). The circling mouse is a spontaneous mutant, with abnormali- Thus, the circling mouse is a useful model for the study of inner ties of the inner ear, that was first reported in Korea (18). The ear abnormalities and deafness. In the study reported here, a mutation is transmitted as an autosomal recessive trait, with cir locus for deafness was finely mapped, using microsatellite 100% penetrance. These mice become hyperactive at about markers. This genetic mapping not only facilitates positional seven days of age, then manifest circling behavior. Microscopi- cloning, but also will help to determine which, if any, human re- cessive deafness loci are homologues of cir. The cir gene has Received: 12/20/02. Revision requested: 2/21/03. Accepted: 7/17/03. been mapped to the central region between the D9Mit116/ 1Catholic Research Institutes of Medical Science, Catholic Medical College, 505 D9Mit15 and D9Mit38 loci on mouse Chr 9. The distal portion of Banpo-Dong, Seocho-Ku, Seoul 137-701, Korea, 2Department of Medical Genet- ics and Experimental Animal Center, College of Medicine, Hallym University, the mouse Chr 9 is syntenic with human Chr 3p21 (25). The hu- Chunchon 200-702, Korea, 3Department of Laboratory Animal Sciences, College man Chr 3p21 region includes the DFNB6 deafness region (11). of Medicine, Hanyang University, SEOUL 100-799, Korea, 4Eulji University, Therefore, the circling mouse may be an animal model of hu- Seoul, Korea, 5Department of Psychology, Korea University, Seoul, Korea, 6Sungkyunkwan University, Seoul, 135-710, Korea, and 7Biological Resource man DFNB6 deafness. Center, Korea Research Institute of Bioscience and Biotechnology, Oun-dong 52, Yusong-ku, Daejeon 305-333, Republic of Korea. *Corresponding authors who equally contributed to this report. Materials and Methods †These corresponding authors also equally contributed to this report. Mice. Circling mice were first discovered in an ICR out-bred 642 Fine mapping of the circling (cir) gene in mice

strain, and have been maintained for 16 generations by matings murine cir gene. The mouse YAC clones were obtained from Re- between affected siblings in the Laboratory Animal Center, search Genetics Inc. and Bioneer Co. Extremities of YAC clones Catholic Research Institutes of Medical Science, Catholic Medi- were recovered by use of the vectorette PCR technique (26). Spe- cal College, Seoul, Korea. Circling C57BL/6J mice were used dur- cific bands generated by PCR analysis were cloned into pGEM-T ing this study. The C57BL/6J mice were obtained from the Korea Easy vector (Promega, Madison, Wis.) and sequenced by use of Research Institute of Bioscience and Biotechnology, Daejeon. SP6 primer in an automated DNA sequencer (Perkin Elmer). Mice were kept in a specific-pathogen-free conditioned animal Cloning of candidate genes. To identify potential genes for care facility. Genetic analyses were conducted with C57BL/6J deafness in the mouse, the conserved genes located within the cir mice as a normal control strain. Mice were free of the following region were amplified by use of primer pairs around the open microorganisms: Sendai virus, mouse hepatitis virus, Myco- reading frame of each gene. The primers used for the candidate plasma pulmonis, Tyzzer’s organism, Pasteurella pneumotropica, analysis were derived from GenBank. The sequence, size and Salmonella spp., Corynebacterium kutscheri, Pseudomonas BLAST accession number of tested genes are listed in Table 2. aeruginosa, and Bordetella bronchiseptica. Microbiological moni- Total RNA was extracted from tissues of C57BL/6J and cir/cir toring against the aforementioned microorganisms in circling mice, using Trizol reagent (Sigma Chemical Co., St. Louis Mo.). and C57BL/6J mice was conducted quarterly. Reverse transcriptase (RT)-PCR analysis was performed, using Mice were housed individually in plastic cages (18 × 30 × 15 the Superscript II pre-amplification system (GIBCO BRL. En- cm) with bed-o’cobs litter (The Andersons Inc. Maumee, Ohio), gland, UK ) according to supplied protocols. Samples were se- and were maintained in an air-conditioned (temperature, 22 to quenced only when bands of correct size were obtained. To 26°C, and humidity, 55 to 60%) and light-controlled (12 h light, isolate the PCR products, the fragments were cloned, using the 12 h dark: lights on/off; 7 a.m./7 p.m.) animal room. All mice pGEM-T Easy Vector System. The cDNA fragments were se- were allowed ad libitum access to rodent chow (No. 5001, quenced, and identity was established by use of BLAST. Purina Mills, Bethlehem, Pa.) and tap water. Chow, bedding, and tap water were sterilized by autoclaving prior to use. All Results animal studies were performed with the approval of the Experi- Clinical observations. All affected mice were clinically nor- mental Animal Care and Use Committee of Catholic University. mal at birth, but gradually manifested hyperactive behavior at The backcross progeny were generated by mating ICR females approximately seven days of age. At that age, shaking of the (+/+) with affected male (cir/cir) mice; then F1 female (cir/+) mice head was first observed, and became more conspicuous during were backcrossed to cir/cir males. Affected mutant mice could be the next three to four days. This pattern was easily noticed at 13 easily distinguished from normal phenotypic littermates (cir/+) or 14 days. The affected mice frequently ran in tight circles, es- by observation of head tossing, circling, and inability to swim at pecially when placed in strange surroundings or when other- postnatal day 10. About three weeks later, 428 backcross prog- wise disturbed. However, they did not have directional eny were collected for this study. preference when circling and were approximately twice as ac- Single-sequence length polymorphism (SSLP) analysis. tive as control mice. Linkage analyses were performed by using microsatellite mark- Genetic mapping of the cir gene. Genetic analyses were ers (purchased from Research Genetics, Inc., Huntsville, Ala.), conducted with C57BL/6J mice as a normal control strain. In using a modification of a described method (34). Polymerase test crosses with circling mice, mating of carrier with affected chain reaction (PCR) analysis for SSLP analysis was performed and affected with affected mice produced normal and affected for a subset of the 428 animals. All primer pairs were obtained offspring in the ratio of 1:1 (189:165) and 0:1 (0:24), respectively. from Research Genetics and Bioneer Co. (Daejeon, Korea). These results indicated that the trait is controlled by a single These microsatellites were amplified from 100 ng of template autosomal recessive gene with complete penetrance. Linkage DNA in a final reaction volume of 30 µl containing 10 mM Tris- analyses were performed by using microsatellite markers (Re- HCl (pH 8.3), 50 mM KCl, 1 to 2 mM MgCl2, 10 pmol of each search Genetics, Inc.) and a modification of a described method primer, and 1.75 U of Taq DNA polymerase (Bioneer Co.). The (32). In total, 114 Massachusetts Institute of Technology (MIT) PCR analysis was carried out in a thermocycler (GeneAmp PCR markers, including six markers on each chromosome, were used System 9600, Perkin-Elmer, Boston, Mass.), with denaturation at for the linkage test. Percentage recombination and standard er- 94°C for one minute, annealing at 50 to 60°C for one minute, and ror were calculated according to Green (12). extension at 72°C for one minute, followed by a final extension at In linkage studies of the 428 offspring collected from the in- 72°C for 10 min at the end of 35 cycles. To screen the size differ- × × traspecific backcross ([cir/cir C57BL/6J] F1 cir/cir), using ence in the gene between ICR and C57BL/6J mice, 15 µl of PCR several markers from all , 35 mice had recombina- product was loaded onto a 15% non-denaturing polyacrylamide tion between the proximal region of cir and D9Mit182, in the gel (18 × 18 cm). Electrophoresis was performed at 100 V for proximal region of the cir locus, yielding a genetic distance of four hours at 25°C. The DNA bands were separated, stained 8.20 ± 1.33 cM between these two loci. This suggests that the cir with ethidium bromide, and visualized by determination of UV gene is located on mouse chromosome 9. Therefore, further link- fluorescence. The χ2 test was used to compare recombination age analysis was done, using markers distal to D9Mit1820. The fractions. Fisher’s exact probability correction was used in cases gene order is defined as: centromere - D9Mit182 - D9Mit51/ where only small numbers were available for statistical analysis. D9Mit79/D9Mit310 - D9Mit212/D9Mit184 - 9Mit116/ Yeast artificial chromosome (YAC) contig. The White- D9Mit15 - cir - D9Mit38 - D9Mit20 - D9Mit243 - D9Mit16 - head Institute’s mouse physical map database at: D9Mit55/D9Mit125 - D9Mit281. Haplotype analysis revealed http://www-genome.wi.mit.edu/cgi-bin/mouse/index the recombination number for each marker (Fig. 1), and the ge- was used in the selection of YAC clones containing regions of the netic map indicates the distances between them (Fig. 2). These 643 Vol 53, No 6 Comparative Medicine December 2003

Table 2. Primer pairs that span the entire sequence of each candidate gene. Primer sequence, exact size of polymerase chain reaction (PCR) products and GenBank accession numbers are shown

Gene Sequence Size of product (bp) GenBank accession no.

Myl3 Forward, TCCACATGCCTCTCTGTCACT Reverse, CTGCCCTGGGCTTCCTGAGAGGC 660 NM_000258 cTnC Forward, GGCAACCCCAGTAGCCTGTCCTGT Reverse, GCGAACAGGCGCAGGCAACCGTGC 556 M_29793

Catnb (1) Forward, TGCTGTGACACCGCTGCGTGGA Reverse, GTGGCCTTCCATCCCTTAGT 1226 NM_007614 Cant (2) Forward, GCAGGAAGGGATGGAAGGCCTC Reverse, TAAAGTATTCACCCACACTG 1166

Tna Forward, CTGACCCAGGCAGCAGCAGCAG Reverse, AGGAGGCACTTCAAGTTCACCTT 270 XM_195798

Limd (1) Forward, TAGGCCCGGACACTGATGTACA Reverse, ACTTGGGCCCATCCTGCAGCC 1026 NM_013860 Limd (2) Forward, CTGCAGGATGGGCCCAGTCTTAC Reverse, CCACGTGCAAGCTGGAGCAT 1293

Pthr (1) Forward, ACGCGGCCCTAGGCGGTGG Reverse, CAGCCACGAAGACAGCCGGC 1033 Pthr (2) Forward, GGGGCTTCACCATCTTTGGC Reverse, TCCAGCCCCTTGATGCCCAG 831

Cspg5 Forward, AGCGCCATGGGGCGCGCCGG Reverse, CCCGCTTCCTCTCTTCTTGCTG 1652 NM_006574 Ptk9 Forward, GAGCCGAAAGCTGAGCCATG Reverse, CCCTGTTCCCAGTGGAGACA 1087 NM_011876

Mtap4 (1) Forward, TCTGCGGCTCCGGCTTGGCTC Reverse, TTCCATAAATGGTGCTACATCC 1056 NM_008633 Mtap4 (2) Forward, CAAATGTCACGCAGCCATCT Reverse, TTAGGCAGAGTTGTAGCTGGACTT 1563 Mtap4 (3) Forward, AGCCACATCTCCCTCAACTCT Reverse, GGTTCCCTGGCATTTGCCGGAA 1174

Zic4 Forward, GCACCGACTGCCAGAAACCTAG Reverse, GGGGTGGGGGCGCTGCAGAG 1103 NM_009576 Zic1 Forward, GGCTTGCCCCGTGCAGCCAC Reverse, GGTCGGCATGTTTTGTTTCTGA 1367 NM_009573

Camk Forward, CGGAGCGCGACGATTCCAAA Reverse, GCCATGTTACCCCCAATCACA 1126 NM_005508

Kif9 (1) Forward, CCACTGCCCCGCGAGCTAAA Reverse, TCTCGTCCAGCGTCCCCTCC 1248 XM_193105 Kif9 (2) Forward, TATTGATGAGACTATTTTCTGT Reverse; CTGAGATCAACTCTCAAGTA 1206

My13 = Myosin light chain; cTnC = Cardiac troponin C; Catnb = Catenin B; Tna = Tetranectin; Limd = LIM domain; Pthr = Parathyroid hormone receptor; Cspg5 = Chondroitin sulfate proteoglycan 5; Ptk9 = Protein tyrosin kinase 9; Mtap4 = Microtubule – associated protein4; Zic4 = Zinc finger protein 4; Zic1 = Zinc finger protein 1; Camk = Calcium/calmodulin – dependent protein; Kif9 = Kinesin family member 9.

results indicate that D9Mit116 and D9Mit38 provided useful and lactotransferrin (Ltf) genes, respectively. Therefore, this flanking anchors for isolation of the cir gene. analysis suggested that the cir gene was located between Matp4 The YAC contig spanning 1 Mb of the distal portion of and Ltf. mouse Chr 9. The analysis of genetic mapping defined a 1-cM Closure of YAC contigs was accomplished by further rounds of candidate interval for the cir gene located between D9Mit116/ chromosome walking, using novel markers derived from end se- D9Mit15 and D9Mit38. To construct the preliminary YAC contig quences of YAC clones. TheYAC ends were isolated by use of from several YAC libraries, three markers (D9Mit116, D9Mit15, vectorette PCR analysis (26). The YAC end sequences were and D9Mit38) known to be closely linked to the cir locus on the screened by searching BLAST (www.ncbi.nlm.nih.gov/BLAST/). distal portion of mouse Chr 9, were used to identify YAC clones By use of sequence-tagged site (STS) analysis, sequences de- (purchased from the MIT/Whitehead, Cambridge, Mass.). To rived from the right end of 457G1 (457G1T7) revealed striking confirm the correct chromosomal location, a chromosome walk identity of the Ltf gene with Map4. Consequently, it was discov- across the cir locus was completed in YACs, using the flanking ered that precise position of Ltf was located at 60 cM on mouse markers, D9Mit184, D9Mit212, D9Mit116, M72414, D9Mit277, Chr 9. On the basis of the sequence obtained from YAC ends, new D9Mit292, D9Mit37, D9Mit185, D9Mit350, and D9Mit309, as primer pairs were synthesized. Details of these YACs and their starting points. The physical order of the microsatellite markers positions relative to the YACs that map to the region are shown is in agreement with the order generated by the genetic map. in Fig. 3. These genes and additional markers established a YAC Interestingly, M72414 and D9Mit38 were the type-II markers contig across the 0.96-cM interval from D9Mit116/D9Mit15 to (intergenic marker) for microtubule-associated protein 4 (Matp4) D9Mit38, which included the entire cir core region. The contig 644 Fine mapping of the circling (cir) gene in mice

Figure 1. Haplotype analysis of the cir gene at mouse chromosome (Chr) 9. Segregation of cir and the flanking markers D9Mit182, D9Mit51, D9Mit79, D9Mit310, D9Mit212, D9Mit184, D9Mit116, D9Mit38, D9Mit20, D9Mit245, D9Mit16, D9Mit55, D9Mit125, and D9Mit281 in the progeny of an intraspecific backcross: (cir xC57BL/6J)F1 x cir. Each column represents the chromosome identified in the back- cross progeny. Bottom line indicates the number of recombinations between cir locus and marker. assembled by STS content mapping was composed of 14 YACs typed by use of 10 markers, including three genes. Mouse and human transcript maps are highly conserved. In the comparative map of the distal region of mouse Chr 9 and human Chr 3 (Fig. 4), the gene order established from mouse YAC contig mapping is Map4-Myl3-Ltf within the cir critical region. These three genes have also been mapped in humans. This region of the distal portion of mouse Chr 9 has conserved synteny with human Chr 3p21, excluding the opposite orienta- Figure 2. Genetic map of the cir locus according to the recombina- tion rate. Genetic map indicating flanking markers on the left, dis- tion of the regions with respect to the centromeres, and the or- tances between each marker, or with cir locus on the right; ± indi- der of human genes is identical to that in the mouse, indicating cates SEM. The filled boxes represent cir alleles, and the open boxes a high degree of conservation between the orthologous regions. represent the C57BL/6J alleles.

Figure 3. Schematic representation of the yeast artificial chromosome (YAC) contig across the cir region. The YACs depicted are a selection of the nonchimeric clones covering the region. Each YAC is drawn to scale and aligned according to the size and sequence-tagged site (STS) content. 645 Vol 53, No 6 Comparative Medicine December 2003

sensorineural components; and sensorineural deafness due to cochlear or retrocochlear defects. The nonsyndromic forms of he- reditary deafness are classified according to their mode of inher- itance: DFN, DFNA, and DFNB refer to forms of deafness inherited from X chromosome-linked, autosomal dominant, and autosomal recessive modes of transmission, respectively. Today, 53 loci have been identified: 21 associated with the DFNA, 26 with the DFNB, and four with the DFN forms, and 2 linked to the mitochondrial genome (17, 29). In previous studies (21, 30, 31), there were 50 to 100 deafness-causing loci, but only eight of these have been genetically mapped. The cir gene of the cir mouse has complete penetrance and is autosomal recessive. Thus, the cir mouse is a useful model for understanding the pathogenesis and molecular mechanism of DFNB nonsyndromic deafness. The cir gene was mapped to 60.1 cM on mouse Chr 9. Currently, it has been reported that two other deafness/vestibular genes are located on Chr 9: Myo6 (44 cM) and Tecta (25 cM) (22). The Myo6 was identified from Snell’s waltzer (sv) mutant mice that had a phenotype similar to that of cir mice (4, 18). In the case of Tecta, the a-Tectorin gene is inac- tivated by targeted deletion, and this null mouse has changes in the structure and orientation of the hair bundles (22). Com- pared with these two mutants, cir mice have loss of outer hair cells. In contrast, outer hair cells were normally orientated in these Snell’s waltzer and Tecta null mice. In addition to these mice, a recessive deafness mutation locus, spinner (sr), is lo- Figure 4. Comparative map of the distal region of mouse Chr 9 and human Chr 3. Mouse genes and markers are listed on the left; human cated at 64 cM on mouse Chr 9 (9). Since the sr gene of the spin- genes and the DFNB6 region are shown on the right. ner mouse is closely linked to the cir gene, it will be important to investigate the relationship between the sr and cir genes, us- Furthermore, on the basis of clone size, the genomic distance of ing an allelism test. approximately 1,980 kb encompassing these transcripts in the Five new STSs and 12 microsatellite markers were used to mouse and human also is strikingly similar. Therefore, genes in build a contig between D9Mit15 and D9Mit38, respectively. the Chr3p21 region are candidates for a human homolog of the However, these markers were genetically positioned 0.93 cM circling mouse. proximal to cir. The physical analysis indicated that the cir gene Cloning of candidate genes. Because hair cells of circling is between Matp4 and Ltf, which lies proximal to cir, concordant mice are severely defective, we selected candidate genes that with D9Mit38, and Matp4 lies distal to cir, concordant with may function in hair cell structure and stereocillia. These are M72414. Thus, assuming recombination events within this re- Mtap4, calcium/calmodulin dependent protein kinase-like gion, these findings strongly suggest that M72414/Matp4 and (Camkl), zinc finger protein of the cerebellum 1 (Zic 1), Zic 4), D9Mit38/Ltf are located near the cir gene and are good mark- laminin beta 2 (Lamb 2), and myosin light chain (Mylc) (1-3, 6, ers for chromosome walking to this gene. Furthermore, the or- 16, 20). The human transcripts place several known genes near der of the genetic markers obtained in our experiments is the cir region including Ltf, myosin light chain 3 (Mylc3), par- consistent with that derived from our physical map. In addition athyroid hormone receptor 1 (Pthr1), kinesin like protein 9 to matp4 and Ltf, several other candidate genes for cir, such as (Kif9), chondroitin sulfate proteoglycan 5 (Cspg5), and other Camkl, Zic 1, Zic 4, and Mylc map to this region (1-3, 6, 16, 20). genes (Fig. 3). The RT-PCR products spanning the open reading The human maps place several other known genes, including frames of each gene were then sequenced, but differences be- Ltf, Myl3, Pthr1, Kif9, Cspg5, and other genes (Fig. 4), near the tween wild-type mouse (C57BL6/J) genes and circling mouse cir region. genes were not observed. We examined all exons of each gene by use of sequencing analysis, with the possibility that one of the aforementioned Discussion candidates for cir was negative as the cause for cir disease. The current comparative map for mice and humans provides However, these findings suggest that the human 3p21 region evidence for many conserved linkages during mammalian evo- does not strictly correspond to the mouse 61 cM region on Chr 9 lution (24, 28). Mouse models of human genetic disorders could (10). Moreover, during divergence of the common ancestor of provide profound insight into basic biological mechanisms and a mouse and human, an insertion occurred in the mouse genome way of evaluating potential therapies. The various forms of or a deletion occurred in the , making it difficult deafness are classified into two categories, syndromic and to identify cir by homology with humans (24). Another reason- nonsyndromic forms (29). The syndromic form is distributed to able possibility is that the cir gene may be a novel gene, func- three classes: conductive deafness due to middle ear and exter- tion of which is not known. nal ear defects, with or without sensorineural components; con- The mouse map location of the cir locus appears to place it in ductive deafness due to middle ear defects, with or without a region homologous to human 3p21. Furthermore, the human 646 Fine mapping of the circling (cir) gene in mice

hearing loss locus, DFNB6, has been assigned to chromosome re- 12. Green, E. L. 1981. Linkage, recombination and mapping, p. 77- gion 3p21 (11, 14). On the basis of the comparative mapping 133. Genetics and probability in animal breeding experiments. data, the circling mouse is a likely mouse model for human Macmillan, New York. 13. Guilford, P., H. Ayadi, S. Blanchard, H. Chaib, D. Le Paslier, DFNB6. This high incidence of closely linked hearing-related loci J. Weissenbach, M. Drira, and C. Petit. 1994. A human gene on Chr 9 is striking, as it points to the existence of multiple responsible for neurosensory, non-syndromic recessive deafness clusted alleles of the hearing-controlling genes. Therefore, the cir- is a candidate homologue of the mouse sh-1 gene. Hum. Mol. cling mouse provides an experimental model for human deafness Genet. 3:989-993. 14. Hmani, M., A. Ghorbel, A. Boulila-Elgaied, Z. Ben Zina, W. and the development of ear structures. Identifying the gene en- Kammoun, M. Drira, M. Chaabouni, C. Petit, and H. Ayadi. coding circling, and elucidating its role in ear development, will 1999. A novel locus for Usher syndrome type II, USH2B, maps to provide a clearer understanding of DFNB6 hearing loss. chromosome 3 at p23-24.2. Eur. J. Hum. Genet. 7:363-367. 15. Kelsell, D. P., J. Dunlop, H. P. Stevens, N. J. Lench, J. N. Liang, G. Parry, R. F. Mueller, and I. M. Leigh. 1997. Connexin 26 mutations in hereditary non-syndromic sensorineural deaf- Acknowledgment ness. Nature 387:80-83. This work was supported by the Korea Research Foundation (2002- 16. Kiss, C., J. Li, A. Szeles, R. Z. Gizatullin, V. I. Kashuba, T. 015-EP0123). Lushnikova, A. I. Protopopov, M. Kelve, H. Kiss, I. D. Kholodnyuk, S. Imreh, G. Klein, and E. R. Zabarovsky. 1997. Assignment of the ARHA and GPX1 genes to human chromo- some bands 3p21.3 by in situ hybridization and with somatic cell References hybrids. Cytogenet. Cell Genet. 79:228-230. 1. Aberdam, D., M. F. Galliano, M. G. Mattei, J. P. Ortonne, 17. Lalwani, A. K. and C. M. Castelein. 1999. Cracking the audi- and G. Meneguzzi. 1994. S-laminin gene (Lams) maps to F1 tory genetic code: nonsyndromic hereditary hearing impairment. band of mouse chromosome 9. Mamm. Genome 5:393-394. Am. J. Otol. 20:115-132. 2. Aruga, J., T. Nagai, T. Tokuyama, Y. Hayashizaki, Y. Okazaki, 18. Lee, J. W., E. J. Lee, S. H. Hong, W. H. Chung, H. T. Lee, T. W. V. M. Chapman, and K. Mikoshiba. 1996. The mouse zic gene Lee, J. R. Lee, H. T. Kim, J. G. Suh, T. Y. Kim and Z. Y. Ryoo. family. Homologues of the Drosophila pair-rule gene odd-paired. 2001. Circling mouse, a possible animal model for deafness. Comp. J. Biol. Chem. 271:1043-1047. Med. 51(6):550-554. 3. Aruga, J., A. Yozu, Y. Hayashizaki, Y. Okazaki, V. M. 19. Li, X. C., L. A. Everett, A. K. Lalwani, D. Desmukh, T. B. Chapman and K. Mikoshiba. 1996. Identification and charac- Friedman, E. D. Green, and E. R. Wilcox. 1998. A mutation in terization of Zic4, a new member of the mouse Zic gene family. PDS causes non-syndromic recessive deafness. Nat. Genet. Gene 72:291-294 18:215-217. 4. Avraham, K. B., T. Hasson, K. P. Steel, D. M. Kingsley, L. B. 20. Mangan, M. E. and J. B. Olmsted. 1996. The gene for microtu- Russell, M. S. Mooseker, N. G. Copeland, and N. A. Jenkins. bule-associated protein 4 (Mtap4) maps to the distal region of 1995. The mouse Snell’s waltzer deafness gene encodes an un- mouse chromosome 9. Mamm. Genome 7:918-919. conventional myosin required for structural integrity of inner 21. Morton, N. E. 1991. Genetic epidemiology of hearing impair- ear hair cells. Nat. Genet. 11:369-375. ment. Ann. N.Y. Acad. Sci. 630:16-31. 5. Baldwin, C. T., S. Weiss, L. A. Farrer, A. L. De Stefano, R. 22. Mustapha, M., D. Weil, S. Chardenoux, S. Elias, E. El-Zir, J. Adair, B. Franklyn, K. K. Kidd, M. Korostishevsky, and B. S. Beckmann, J. Loiselet, and C. Petit. 1999. An alpha-tectorin Bonne-Tamir. 1995. Linkage of congenital, recessive deafness gene defect causes a newly identified autosomal recessive form (DFNB4) to chromosome 7q31 and evidence for genetic hetero- of sensorineural pre-lingual non-syndromic deafness, DFNB21. geneity in the Middle Eastern Druze population. Hum. Mol. Hum. Mol. Genet. 8:409-412. Genet. 4:1637-1642. 23. Nadeau, J. H., B. Herrmann, M. Bucan, D. Burkart, J. L. 6. Barton, P. J., A. Cohen, B. Robert, M. Y. Fiszman, F. Crosby, M. A. Erhart, M. Kosowsky, P. J. Kraus, F. Michiels, Bonhomme, J. L. Guenet, D. P. Leader, and M. E. A. Schnattinger, M. B. Tchetgen, D. Varnum, K, Willison, Buckingham. 1985. The myosin alkali light chains of mouse H. Lehrach, and D. Barlow. 1991. Genetic maps of mouse chro- ventricular and slow skeletal muscle are indistinguishable and mosome 17 including 12 new anonymous DNA loci and 25 an- are encoded by the same gene. J. Biol. Chem. 260:8578-8584 chor loci. Genomics 9:78-89. 7. Bergstrom, L., W. G. Hemenway, and I. Sando. 1972. Patho- 24. Oakey, R. J., M. L. Watson, and M. F. Seldin. 1992. Construc- logical changes in congenital deafness. Laryngoscope 82:1777- tion of a physical map on mouse and human chromosome 1: com- 1792. parison of 13 Mb of mouse and 11 Mb of human DNA. Hum. Mol. 8. Bork, J. M., L. M. Peters, S. Riazuddin, S. L. Bernstein, Z. Genet. 1:613-620. M. Ahmed, S. L. Ness, R. Polomeno, A. Ramesh, M. Schloss, 25. Petit, C. 1996. Genes responsible for human hereditary deaf- C. R. Srisailpathy, S. Wayne, S. Bellman, D. Desmukh, Z. ness: symphony of a thousand. Nat. Genet. 14:385-391. Ahmed, S. N. Khan, V. M. Kaloustian, X. C. Li, A. Lalwani, 26. Riley, J., R. Butler, D. Ogilvie, R. Finniear, D. Jenner, S. S. Riazuddin, M. Bitner-Glindzicz, W. E. Nance, X. Z. Liu, Powell, R. Anand, J. C. Smith, and A. F. Markham. 1990. A G. Wistow, R. J. Smith, A. J. Griffith, E. R. Wilcox, T. B. Fried- novel, rapid method for the isolation of terminal sequences from man, and R. J. Morell. 2001. Usher syndrome 1D and yeast artificial chromosome (YAC) clones. Nucleic Acids Res. nonsyndromic autosomal recessive deafness DFNB12 are caused 18:2887-2890. by allelic mutations of the novel cadherin-like gene CDH23. Am. 27. Scott, H. S., J. Kudoh, M. Wattenhofer, K. Shibuya, A. Berry, J. Hum. Genet. 68:26-37. R. Chrast, M. Guipponi, J. Wang, K. Kawasaki, S. Asakawa, 9. Deol, M. S. and M. W. Robins. 1962. The spinner mouse. J. S. Minoshima, F. Younus, S. Q. Mehdi, U. Radhakrishna, M. Hered. 53:133-136. P. Papasavvas, C. Gehrig, C. Rossier, M. Korostishevsky, A. 10. Friedman, T. B., Y. Liang, J. L. Weber, J. T. Hinnant, T. D. Gal, N. Shimizu, B. Bonne-Tamir, and S. E. Antonarakis. Barber, S. Winata, I. N. Arhya, and J. H. Asher, Jr. 1995. A 2001. Insertion of beta-satellite repeats identifies a transmem- gene for congenital, recessive deafness DFNB3 maps to the brane protease causing both congenital and childhood onset au- pericentromeric region of chromosome 17. Nat. Genet. 9:86-91. tosomal recessive deafness. Nat. Genet. 27:59-63. 11. Fukushima, K, A. Ramesh, C. R. Srisailapathy, L. Ni, S. 28. Searle, A. G., J. Peters, M. F. Lyon, J. G. Hall, E. P. Evans, J. Wayne, M. E. O’Neill, G. Van Camp, P. Coucke, P. Jain, and H. Edwards, and V. J. Buckle. 1989. Chromosome maps of man E. R. Wilcox. 1995. An autosomal recessive nonsyndromic form and mouse. IV. Ann. Hum. Genet. 53:89-140. of sensorineural hearing loss maps to 3p-DFNB6. Genome Res. 29. Smith, S. D. 1995. Overview of genetic auditory syndromes. J. 5:305-308. Am. Acad. Audiol. 6:1-14.

647 Vol 53, No 6 Comparative Medicine December 2003

30. Steel, K. P. and T. J. Bussoli. 1999. Deafness gene expression 35. Weil, D., P. Kussel, S. Blanchard, G. Levy, F. Levi-Acobas, of surprise. Trends Genet. 15:207-211. M. Drira, H. Ayadi, and C. Petit. 1997. The autosomal reces- 31. Sundstrom, R. A., L. Van Laer, G. Van Camp, and R. J. Smith. sive isolated deafness, DFNB2, and the Usher 1B syndrome are 1999. Autosomal recessive nonsyndromic hearing loss. Am. J. Med. allelic defects of the myosin-VIIA gene. Nat. Genet. 16:191-193. Genet. 89(3):123-129. 36. Wilcox, E. R., Q. L. Burton, S. Naz, S. Riazuddin, T. N. Smith, 32. Suh, J. G., T. Yamanishi, K. Matsui, K. Tanaka, and K. Wada. B. Ploplis, I. Belyantseva, T. Ben-Yosef, N. A. Liburd, R. J. 1995. Mapping of the gracile axonal dystrophy (gad) gene to a Morell, B. Kachar, D. K. Wu, A. J. Griffith, S. Riazuddin, region between D5Mit197 and D5Mit113 on proximal mouse chro- and T. B. Friedman. 2001. Mutations in the gene encoding tight mosome 5. Genomics 27:549-551. junction claudin-14 cause autosomal recessive deafness DFNB29. 33. Van Camp, G., P. J. Coucke, H. Kunst, I. Schatteman, D. Cell 104:165-172. Van Velzen, H. Marres, M. van Ewijk, F. Declau, P. Van 37. Yonezawa, S., A. Yoshiki, A. Hanai, T. Matsuzaki, T. Hauwe, J. Meyers, J. Kenyon, S. D. Smith, R. J. Smith, B. Matsuzaki, J. Matsushima, T. Kamada, and M. Kusakabe. Djelantik, C. W. Cremers, P. H. Van de Heyning, and P. J. 1999. Chromosomal localization of a gene responsible for Willems. 1997. Linkage analysis of progressive hearing loss in vestibulocochlear defects of BUS/Idr mice: identification as an five extended families maps the DFNA2 gene to a 1.25-Mb re- allele of waltzer. Hear. Res. 134:116-122. gion on chromosome 1p. Genomics 41:70-74. 34. Wakana, S. and K. Imai. 1998. Mouse chromosome 9. Mamm. Genome 8(S):180-199.

648