Characterization of unconventional MYO6, the human homologue of the gene responsible for deafness in Snell's waltzer mice
Characterization of unconventional MYO6 , the human homologue of the gene responsible for deafness in Snell's waltzer miceKaren B. Avraham1,*, Tama Hasson2, Tama Sobe1, Binaifer Balsara3, Joseph R. Testa3, Anne B. Skvorak4, Cynthia C. Morton4,5, Neal G. Copeland6 and Nancy A. Jenkins6
1Department of Human Genetics, Sackler School of Medicine, Tel Aviv University, Tel Aviv 69978, Israel, 2Departments of Biology and Pathology, Yale University, New Haven, CT 06511, USA, 3Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA 19111, USA, 4Department of Pathology and 5Department of Obstetrics, Gynecology and Reproductive Biology, Brigham and Women's Hospital and Harvard Medical School, Boston, MA 02115, USA and 6Mammalian Genetics Laboratory, ABL-Basic Research Program, NCI-Frederick Cancer Research and Development Center, Frederick, MD 21702, USA
Received February 25, 1997;Revised and Accepted May 8, 1997
DDBJ/EMBL/GenBank accession no. U90236
Deafness is the most common form of sensory impairment in humans. Mutations in unconventional myosins have been found to cause deafness in humans and mice. The mouse recessive deafness mutation, Snell's waltzer, contains an intragenic deletion in an unconventional myosin, myosin VI (locus designation, Myo6). The requirement for Myo6 for proper hearing in mice makes this gene an excellent candidate for a human deafness disorder. Here we report the cloning and characterization of the human unconventional myosin VI (locus designation, MYO6) cDNA. The MYO6 gene maps to human chromosome 6q13. The isolation of the human gene makes it now possible to determine if mutations in MYO6 contribute to the pathogenesis of deafness in the human population.
Deafness is the most common form of sensory impairment in humans. Approximately 1 out of 1000 individuals is born deaf, while about 1 out of 10 individuals become deaf or hard of hearing during their lifetime (1 ). Hearing impairment can result from a variety of injuries or diseases, and occurs both in association with other symptoms, or as an isolated finding. Hearing loss can be genetic or acquired, and occurs both prelingually due to genetic defects or childhood diseases, or postlingually due to inherited gene mutations, environmental noise pollution, infection or presbycusis. Despite its prevalence in the human population, very little is known about the molecular events leading to deafness and about the genes that control normal hearing. The study of genes involved in hearing will provide us with clues regarding the molecular basis of hearing transduction.
As many as 100 genes may be involved in non-syndromic deafness with an autosomal recessive mode of inheritance (2 ). It has been very difficult to identify these genes because of the absence of characteristic clinical signs in deaf individuals, genetic heterogeneity of deafness, and small family populations. To date, only 13 autosomal recessive deafness loci have been described (3 ). Additional genes for autosomal deafness exist, but the lack of suitable families for linkage analysis makes discovering these genes difficult. Therefore, genes responsible for hearing loss in the mouse provide another method to identify potential human deafness loci. The structure and development of the inner ear, and the pathology leading to hearing impairment is very similar between mice and humans (4 ).
Over 60 mouse mutations affecting the inner ear have been mapped in the mouse genome (5 ). Two of the most recently identified mouse deafness genes are unconventional myosins, demonstrating the important role in hearing of this family of proteins (6 ,7 ). Unconventional myosins are molecular motors defined by their conserved head or motor domain, light-chain binding neck or regulatory domain, and a unique tail domain. Upon interacting with actin, they convert energy from ATP hydrolysis to mechanical force and have been found to play a role in sensory hair cells, melanocytes, retinal epithelia and neurons (8 ,9 ). Myosins are crucial for the proper function of the hair cells, the sensory cells of the inner ear, since mutations in two of these proteins cause deafness in mice, and in one case, in humans. The mouse unconventional Myo6 was discovered during a positional cloning approach to clone the gene responsible for deafness in the Snell's waltzer (sv) mutation (6 ). Myosin VIIA was found to be defective in shaker-1 (sh1) mice, an autosomal recessive mutation characterized by deafness and vestibular dysfunction and in human Usher syndrome type 1B, an inherited disease characterized by deafness, vestibular dysfunction and retinitis pigmentosa (7 ,10 ). The involvement of myosin VIIA in both human and mouse deafness, the similarity of the human and mouse auditory systems, and the existence of additional unidentified human deafness loci suggest that Myo6 will also be involved in human deafness. Here we report the cloning and characterization of the human MYO6 cDNA, which maps to human chromosome 6q13. Cloning the human homologue of the mouse gene provides a method for discovering an additional human deafness locus, circumventing the need for conventional linkage analysis. The requirement for Myo6 in hearing in mice makes this gene an excellent candidate for a human deafness disorder.
The human homologue of the Snell's waltzer locus, MYO6, was isolated by screening a human cerebral cDNA library with the mouse Myo6 cDNA (6 ). Four clones, HVI-13 (from nucleotides -120 to 1045), HVI-11 (nt 1955-2917), HVI-16 (nt 2394-3079) and HVI-18 (nt 2977 to end), were isolated and sequenced (Fig. 1 a). Comparison of the sequence of HVI-13 with that of the mouse cDNA indicated that this clone was homologous to the 5' end of mouse Myo6. HVI-13 contained 120 bp of the 5' untranslated region (UTR) preceding the same ATG start site identified in the mouse. The remaining clones were derived from the central and 3' end of the cDNA, including 208 bp from the 3' UTR (HVI-18). The poly(A+) tail was not recovered during the screening. Alignment of the human cDNA clones to the full-length cDNA sequence revealed a gap of ~1000 nt (between HVI-13 and HVI-11). RT-PCR was performed on human cerebral RNA using primers derived from the human cDNA clones on either side of the gap. The sequences of the cDNA clones and RT-PCR products were assembled and revealed a predicted open reading frame (ORF) of 3789 bp (Fig. 1 a and b). cDNA nucleotide position +1 was assigned to the first nucleotide of the start codon (ATG). The human MYO6 cDNA is predicted to encode a protein of 1263 amino acids with a relative molecular weight of 142 kDa (Fig. 1 b). The protein contains a head/motor region of 776 amino acids, containing an ATP-binding domain (GESGAGKT, amino acids 151-158) and actin-binding sequences (11 ,12 ). The motor domain contains a conserved threonine residue at amino acid 405, which may serve as a site for phosphorylation (Figs 1 b and 2 a). Following the motor domain is a 52 amino acid region that is unique to myosin VI, whereas in other unconventional myosins, the light chain binding site immediately follows the motor region. The light chain binding site that follows is a single IQ motif, serving as a site for calmodulin (RAEACIKMQKTIRMWLCKRR, amino acids 829-848) (11 ). The tail, like that of the mouse myosin VI, contains a region predicted to be a coiled-coil of 192 amino acids (13 ), followed by a distal globular region of 232 amino acids (Fig. 1 b).
Myosin VI was first isolated in Drosophila in the process of isolating novel actin-binding proteins from embryonic extracts (14 ). This molecule was subsequently found to play an important role in the organization of the syncytial blastoderm and transportation of cytoplasmic particles (15 ,16 ). The pig myosin VI was isolated during a characterization of myosin isoforms expressed in kidney proximal tubule cells (11 ). Recently, a partial sequence of nematode myosin VI was isolated as well (J.P.Baker and M.A.Titus, personal communication). Comparison of the cDNA and predicted amino acid sequences from Drosophila, Caenorhabditiselegans, pig, mouse and human myosin VI shows strong evolutionary conservation not only in the head or motor domain, but also in the distal tail region (Table 1 ; Fig. 2 a), whose structure and binding interactions may determine subcellular localization and function (9 ). The threonine site at amino acid 405 in human myosin VI is conserved between all myosin VI species examined (Fig. 2 a).
Phylogenetic sequence comparisons of the conserved head domains of unconventional myosins have been used to demonstrate that the myosins are grouped into several distinct classes (9 ). An unrooted phylogenetic tree of myosins VI and other vertebrate myosins based on the amino acid sequence of their head domains is shown in Figure 2 b. Included for the first time in this analysis is the sequence of the human and C.elegans myosin VI.
The expression of human MYO6 was examined in both fetal and adult human tissues. The human MYO6 cDNA clone HVI-18, containing a portion of the tail region, was hybridized to northern blots containing RNA from a series of adult human tissues (Fig. 3 a). The major MYO6 transcript is ~6.0 kb and was observed in most tissues examined. A less abundant 8.0 kb transcript was also detected (Fig. 3 a) The two transcripts were also observed in the mouse (6 ). Highest levels of expression of MYO6 in human tissues were seen in brain, pancreas, prostate, testis and small intestine.
The human chromosomal location of MYO6 was determined by fluorescence insitu hybridization (FISH) analysis. Hybridization of the HVI-13 probe to metaphase spreads revealed specific labeling on human chromosome 6. Fluorescent signals were detected on chromosome 6 in 21 of 35 metaphase spreads. Among a total of 93 signals observed, 33 (35.4%) were located on 6q. The vast majority (29 of 33) of signals on chromosome 6 were located at band 6q13 (Fig. 5 a). Specific signals on 6q were distributed as follows: one chromatid (11 cells), two chromatids (9 cells), three chromatids (0 cells), four chromatids (1 cell).
We have cloned and characterized the human cDNA homologous to the mouse unconventional Myo6, a gene crucial for sensorineural hearing in the mouse (6 ). The structure and function of the mouse and human inner ear is remarkably similar, and thus the mouse has proved to be an invaluable model for studying human hearing loss. The first genes responsible for autosomal recessive deafness were initially identified in the mouse (6 ,7 ), demonstrating the value of the mouse for identifying and studying human deafness genes.The isolation of the mouse unconventional Myo6 gene has been instrumental in cloning the human MYO6 gene, and facilitates the discovery of mutations in the human deaf population.
The function and regulation of myosin VI remains largely unknown, although homology of a potentially phosphorylated amino acid between all myosins-VI and myosins-I of Acanthamoeba suggests a similar form of regulation (9 ). Actin- activated Mg-ATPase and motility of Acanthamoeba myosins-I require phosphorylation of a serine or threonine residue by a heavy-chain kinase (18 ). No other vertebrate myosins of any class share this site, except for myosins-VI. The conservation of this residue in this class of myosins has led to the postulation that myosin VI requires heavy chain phosphorylation for enzymatic and mechanochemical activity (19 ).
Gene-tree phylogenetic analyses are crucial to determine the evolutionary relatedness or homology between proteins (20 ). Protein sequence comparison between different myosin VI species shows a remarkably high degree of conservation, considering the diversity of the organisms, suggesting they retain a similar and vital function. From studies done in Drosophila, we can predict that myosin VI has an important role in development, and from work in the mouse, that myosin VI is crucial for proper hearing.
The expression of MYO6 in human fetal cochlea demonstrates the importance of myosin VI in the mammalian inner ear and supports its potential role in human inner ear pathology. Within the hair cells, myosin VI is enriched in the cuticular plate, an actin-rich structure at the base of the stereocilia. To carry out mechanical transduction, the hair cell utilizes an actin-rich hair bundle made up of the stereocilia, which is inserted rigidly into the cuticular plate. The arrangement and number of stereocilia at the surface of the hair cell is very precise and is formed in a series of steps to generate the complex actin cytoskeleton (21 ). Therefore, proper hair bundle formation is crucial for hearing. Myosin VI may play an important role in this formation due to its high level of expression in the cuticular plate. Other unconventional myosins are also expressed in the hair cells, namely myosin-I[beta] and myosin VIIa (locus designation, MYO7A) (22 ,23 ). Genetic analysis of MYO6 and MYO7A reveals differences in the role of these motors in the sensory hair cell, since mutations in both cause deafness in mice (6 ,7 ). Myosin-I[beta] is located at the stereociliary tips (22 ), whereas myosin VIIa is not located at the tips, but distributed along the length of the stereocilia (23 ). The different localization in the hair cell of these three unconventional myosins suggests that they have different functions in the sensory hair cell.
Comparative genome mapping between human and mouse has revealed very high conservation of linkage organization between the two organisms (24 ). Approximately 140-180 conserved chromosomal segments of linked homologous genes have been identified between the human and mouse genome maps (25 ). The mouse Myo6 gene maps to the central region of chromosome 9, in a region homologous to human chromosomes 6 and 15 (26 ). In the mouse, Myo6 maps between Gsta and Htr1b; the human homologues, GSTA2 and HTR1B, both map to human chromosome 6 (27 ,28 ).
Two loci for non-syndromic autosomal dominant deafness have been localized to human chromosome 6, one at 6q22-q23 (DFNA10) (29 ) and the second at 6q21 (DFNA13; R.Smith, personal communication), though no recessive loci have been identified on this chromosome to date. Microsatellite markers localized to 6q13 can be used to identify linkage in potential deaf individuals. Mutations in MYO6 can then be identified by sequencing the MYO6 coding region and/or by hybridizing RNA from normal and deaf individuals. Discovering mutations in MYO6 in the human deaf population may ultimately lead to better diagnosis of inherited deafness and help improve the quality of life of deaf individuals by suggesting new means of treating or preventing hearing loss.
An oligo(dT) and random primed human cerebral brain cDNA library in a [lambda]gt10 vector (Clontech) was screened with the mouse Myo6 cDNA (6 ). 106 p.f.u. were plated at a density of 5 * 104 p.f.u./15 cm plate and replica filters were hybridized with [[alpha]-32P]dCTP using a random priming labeling kit (Amersham). [lambda] DNA was isolated using a Qiagen DNA preparation kit.
Poly(A)+ RNA, isolated from human brain (Clontech), was reverse-transcribed using oligo(dT) and random hexanucleotide primers (Invitrogen) and SuperScript II reverse transcriptase (Gibco BRL). PCR primers 23617 (5'-gct gca cta gat act ttg cta ac-3'; nt 878-901) and 22779 (5'-gtg gtg gct tgt cat ctt taa g-3'; nt 2064-2085) were denatured at 94oC for 1 min, annealed at 60oC for 1 min, and extended at 72oC for 1 min for 30 cycles.
Sequencing of plasmid DNA was done using the PRISMTM Ready Reaction DyeDeoxyTM Terminator Cycle Sequencing Kit (Perkin Elmer). RT-PCR products were purified on columns to remove primers (Centricep columns, Princeton Separations). Taq-based cycle sequencing reactions were done using fluorescent dideoxy terminators. Reactions were read on an ABI Model 373A DNA Sequencer (Applied Biosystems).
Northern blots containing 2 [mu]g of poly(A)+ RNA from 16 different human tissues were hybridized with a portion of the human MYO6 cDNA (Multiple Tissue Northern I and II, Clontech) according to the manufacturer's protocol. The probe was a PCR-generated fragment of the tail (nt 2799-3677) derived from cDNA clone HVI-18. Uniformity of loading was checked by hybridization to a chicken glycerol-3-aldehyde phosphate dehydrogenase (GAPDH) probe.
Total RNA from human fetal cochlea and brain was extracted using the guanidine isothiocyanate method (30 ). Total RNA (2.5 [mu]g) was reverse transcribed using the SuperScript II kit of Gibco/BRL. Parallel reactions without reverse transcriptase were performed to control for DNA contamination of RNA samples. Ten percent of the reverse transcriptase (+- RT) reaction was used as template for PCR. PCR reagents including Taq polymerase were from Perkin Elmer and were used according to the manufacturer's instructions. PCR primers 211253 (5'-ggatctgtgttcaaggcaaag-3'; mouse cDNA nt 1073-1092; corresponding to human cDNA nt 701-721) and 202592 (5'-cagattgcagccacctgaag-3'; nt 1073-1092) were denatured at 94oC for 45 s, annealed at 58oC for 45 s, and extended at 72oC for 1 min for 30 cycles.
The mice cochlea were treated as described previously (6 ). Briefly, the dissected cochlea from 3 week old wild-type mice were fixed then treated with affinity purified rabbit anti-myosin VI-tail or control non-immune rabbit IgG. Samples were incubated with fluorescein-conjugated anti-rabbit secondary antibody (Cappel) and rhodamine-conjugated phalloidin (Molecular Probes), before being placed in PBS-glycerol mounting media (Citifluor). Samples were observed using a BioRad MRC600 laser scanning confocal microscope, captured using the COMOS program, and contrast enhanced in Photoshop 3.0 before compilation into a two-panelled figure in Canvas 3.52.
Metaphase spreads from normal human female lymphocytes were prepared as described (31 ). The MYO6 cDNA clone, HVI-13 (from nt -120 to 1045), was labeled with biotin-16-dUTP by nick translation. FISH and detection of immunofluorescence were performed as previously described (32 ). The chromosome preparations were stained with diamidino-2-phenylindole (DAPI) and observed with a Zeiss Axiophot fluorescence microscope. Images were captured with a cooled CCD camera connected to a Power Macintosh 8500 work station. Digitized images of DAPI staining and fluorescent signals were merged using Oncor Image software.
Research sponsored in part by the National Cancer Institute, DHHS, with ABL; by NIH grant CA-06927 and an appropriation from the Commonwealth of Pennsylvania to B.B. and J.R.T.; by NIH grant DK38979 and a grant from the Deafness Research Foundation to T.H.; by NIH grant DC00038-05 to A.B.S. and NIH grant DC00871 to C.C.M.; and by Tel Aviv University.
16 Mermall, V. and Miller, K.G. (1995) J. CellBiol. 129, 1575-1588.
17 Dib, C., Fauré, S., Fizames, C., Samson, D., Drouot, N., Vignal, A., Millasseau, P., Marc, S., Hazan, J., Seboun, E., Lathrop, M., Gyapay, G., Morissette, J. and Weissenbach, J. (1996) Nature380, 152-154.MEDLINE Abstract
18 Brzeska, H., Lynch, T.J. and Korn, E.D. (1990) J. Biol. Chem. 265, 3591-3594.
19 Bement, W.M. and Mooseker, M.S. (1995) CellMotil. Cytoskeleton31, 87-92.
33 Devereux, J., Haeberli, P. and Smithies, O. (1984) NucleicAcidsRes. 12, 387-395.
*To whom correspondence should be addressed. Tel: +972 3 640 7030; Fax: +972 3 640 9900; Email: karena@post.tau.ac.il
-->
This page is maintained by OUP admin. Last updated Tue Jul 15 11:13:42 BST 1997. Part of the OUP Journals World Wide Web service.
Copyright
Oxford University Press, 1996
