Cloning of the Genes Encoding Two Murine and ... - Science Direct

GENOMICS

40, 332–341 (1997) GE964526

ARTICLE NO.

Cloning of the Genes Encoding Two Murine and Human Cochlear Unconventional Type I Myosins FABIEN CROZET,* AZIZ EL AMRAOUI,* STE´PHANE BLANCHARD,* MARC LENOIR,† CHANTAL RIPOLL,† PHILIPPE VAGO,† CHRISTIAN HAMEL,† CEĆILE FIZAMES,‡ FABIENNE LEVI-ACOBAS,* DANIE`LE DEPE´TRIS,§ MARIE-GENEVIÈVE MATTEI,§ DOMINIQUE WEIL,* RE´MY PUJOL,† AND CHRISTINE PETIT*,1 *Unite´ de Geńe´tique Molećulaire Humaine, Centre National de la Recherche Scientifique, Unite´ de Recherche Associeé 1968, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris Cedex 15, France; †Neurobiologie de l’Audition-Plasticite´ Synaptique, Institut National de la Sante´ et de la Recherche Me´dicale, Unite´ 254, Hoˆpital St-Charles, 300 rue Auguste Broussonnet, 34295 Montpellier Cedex 05, France; ‡Geńe´thon, Centre National de la Recherche Scientifique, Unite´ de Recherche Associeé 1922, 1 rue de l’Internationale, 91000 Evry, France; and Geńe´tique Me´dicale et De´veloppement, Institut National de la Sante´ et de la Recherche Me´dicale Unite´ 406, Faculte´ de Me´decine de la Timone, 27 Bd Jean Moulin, 13385 Marseille cedex 05, France Received August 27, 1996; accepted November 20, 1996

Several lines of evidence indicate a crucial role for unconventional myosins in the function of the sensory hair cells of the inner ear. We report here the characterization of the cDNAs encoding two unconventional type I myosins from a mouse cochlear cDNA library. The first cDNA encodes a putative protein named Myo1c, which is likely to be the murine orthologue of the bullfrog myosin Ib and which may be involved in the gating of the mechanotransduction channel of the sensory hair cells. This myosin belongs to the group of short-tailed myosins I, with its tail ending shortly after a polybasic, TH-1-like domain. The second cDNA encodes a novel type I myosin Myo1f which displays three regions: a head domain with the conserved ATPand actin-binding sites, a neck domain with a single IQ motif, and a tail domain with the tripartite structure initially described in protozoan myosins I. The tail of Myo1f includes (1) a TH-1 region rich in basic residues, which may interact with anionic membrane phospholipids; (2) a TH-2 proline-rich region, expected to contain an ATP-insensitive actin-binding site; and (3) a SH-3 domain found in a variety of cytoskeletal and signaling proteins. Northern blot analysis indicated that the genes encoding Myo1c and Myo1f display a widespread tissue expression in the adult mouse. Myo1c and Myo1f were mapped by in situ hybridization to the chromosomal regions 11D-11E and 17B-17C, respectively. The human orthologuous genes MYO1C and MYO1F were also characterized, and mapped to the human chromosomal regions 17p13 and 19p13.2– 19p13.3, respectively. q 1997 Academic Press INTRODUCTION

Unconventional myosins are actin-based motor proteins. They share a well-conserved head domain that 1 To whom correspondence should be addressed. Telephone: (33) 1 45 68 88 90/50. Fax: (33) 1 45 67 69 78. E-mail: [email protected].

0888-7543/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

GENO 4526

/

6r26$$$201

contains the ATP- and actin-binding sites. This motor domain converts chemical energy, liberated by the hydrolysis of ATP, into mechanical force to move along actin filaments. The head domain is followed by a neck regulatory domain, containing at least one IQ motif that binds calmodulin or calmodulin-like light chains. The tail domain is highly divergent from one unconventional myosin to the other; it does not form a coiledcoil region allowing self-assembly into filaments as observed with conventional myosins. Tail domains are expected to determine the function of each individual myosin by mediating specific interactions with different cellular proteins (Mooseker and Cheney, 1996; Hasson and Mooseker, 1996). Myosins are distributed into at least 12 classes, a classification based on phylogenic analysis of the head domain-deduced amino acid sequences (the only sequences available for a number of myosins) (Cheney et al., 1993). In recent years, considerable attention has been drawn to unconventional myosins in the inner ear sensory hair cells (Ashmore, 1995; Gillespie, 1996). The auditory organ, the cochlea, is characterized by its highly ordered architecture, relying on the remarkable structural and spatial organization of sensory hair cells themselves. Furthermore, the actin-rich hair bundle of hair cells inserts rigidly into another structure formed from actin filaments, the cuticular plate (Gillespie, 1996). Thus, cytoskeletal defects are expected to be particularly deleterious for the auditory function. So far, two inner ear defects have been ascribed to the dysfunctions of two different unconventional myosins. The gene encoding myosin VIIA, MYO7A, has been shown to be responsible for the human Usher syndrome type IB (Weil et al., 1995), which associates sensorineural hearing loss, vestibular defect, and retinitis pigmentosa (Usher, 1913/1914). Interestingly, defects in the orthologous gene in the mouse result in the shaker-

332

02-11-97 13:05:47

gnma

TWO COCHLEAR UNCONVENTIONAL TYPE I MYOSINS

1 phenotype (Gibson et al., 1995), which consists of hearing impairment and vestibular dysfunction, without any retinal defect (Steel, 1995). It has also been shown that the gene encoding myosin VI, Myob, is responsible for the Snell’s waltzer mouse mutants, characterized by deafness and vestibular dysfunction (Avraham et al., 1995). Although these two proteins are expressed in various tissues (Avraham et al., 1995; Hasson et al., 1995; Weil et al., 1996; Sahly et al., unpublished results for myosin VIIA), the mutant phenotypes suggest that they play critical roles only in sensory cells. In the inner ear, both proteins are expressed in the sensory hair cells (Avraham et al., 1995; Hasson et al., 1995; El-Amraoui et al., 1996). Myosin VI is supposed to maintain the structural integrity of the cuticular plate in hair cells (Avraham et al., 1995). Myosin VIIA seems to contribute both to the formation of the cytoskeletal network, including the formation of stereocilia, and to the intracellular vesicular traffic (ElAmraoui et al., 1996). In addition, there is growing evidence that the transduction channel located at the tip of the stereocilia of the cochlear sensory hair cells is being gated via a mechanism implicating unconventional myosins (Gillespie et al., 1993; Metcalf et al., 1994; Solc et al., 1994). Indeed, the prevailing model of the adaptation process in hair cells postulates adjustment of the tension of the tip links joining the mechanotransduction channels of two adjacent stereocilia (Hudspeth and Gillespie, 1994). A putative molecular motor protein localized at the upper end of each tip link might regulate the tip link tension by mediating tiny displacements along actin filaments. Since the bullfrog myosin Ib has been localized to the tip of the stereocilia isolated from the sacculus (Gillespie et al., 1993), it has been proposed that this protein is the molecular motor responsible for the adaptation process (Metcalf et al., 1994; Solc et al., 1994). Nevertheless, no direct evidence supporting this hypothesis is yet available (Gillespie, 1995). Also, the possibility that several myosins are involved cannot be ruled out. To investigate the presence of other unconventional myosins in the inner ear, we generated a mouse cochlear cDNA library, which was screened for myosin clones. The cDNAs encoding two unconventional type I myosins were isolated. The first cDNA is likely to encode the murine orthologue of the rat and bullfrog myosins Ib. The other cDNA encodes a novel type I myosin. The human orthologous cDNAs were also cloned, sequenced, and the corresponding genes were mapped to human chromosomes. Human genes were given the prefix MYO, in accordance with the Human Nomenclature Committee, and given a corresponding number (for the class) and letter (for the locus) according to the mouse locus name, as recently proposed in Hasson et al., 1996. MATERIALS AND METHODS Construction of a cDNA library from the mouse cochlea and screening of clones. Four hundred cochleae from 16-day-old BALB/c mice

AID

GENO 4526

/

6r26$$$202

02-11-97 13:05:47

333

were dissected in phosphate-buffered saline (PBS). Careful microscopic dissection was performed to keep only the membranous portions of the cochlea and discard as much as possible of surrounding bony and cartilaginous elements. RNAs were extracted by the guanidium isothiocyanate method (Chomczynski and Sacchi, 1987). Poly(A)/ RNAs were purified from approximately 200 mg of total RNA using magnetic beads coupled to oligo(dT)25 (Dynabead) according to the manufacturer’s instructions. To construct this cDNA library, first-strand cDNAs were synthesized using avian myeloblastosis virus reverse transcriptase with both oligo(dT) and random primers. The RNA–cDNA hybrids were then converted to double-stranded cDNAs by DNA polymerase I in combination with RNase H and Escherichia coli DNA ligase. The cDNAs were blunt-ended using T4 DNA polymerase, and BstxI/EcoRI adaptators were added. cDNAs were size-selected at 800 bp on an agarose gel, ligated into a BstxIcut pcDNAI phagemid, a mammalian expression vector, and the recombinant vectors were used to transform MC1061/P3 E. coli. The PCR probes used in this study are located in the human MYO7A cDNA region encoding the highly conserved ATP- and actinbinding domains (positions 1 to 500 and 1830 to 2280) (Weil et al., 1996). PCR products were purified by agarose gel electrophoresis, labeled with [32P]dCTP by random priming, and used to screen the mouse cochlear cDNA library. Six different clones were isolated after three rounds of hybridization screening and were sequenced. To obtain the cDNA sequence encoding the C-terminal part of Myo1c, 3* terminal rapid amplification of cDNA ends (3* RACE) was performed on cochlear poly(A)/ RNAs according to the supplier’s instructions (cDNA Marathon amplification kit; Clontech). A human kidney cDNA library constructed in lgt10 (Clontech, Reference No. HL3001a), was used to isolate the human orthologues of the mouse cDNAs encoding Myo1c and Myo1f. The probes used were the Myo1c cDNA fragment from nucleotide 1647 to 2568, to isolate the MYO1C cDNA, and the Myo1f cDNA fragment from nucleotide 2205 to 2933, to isolate the MYO1F cDNA. DNA sequencing and sequence analysis. DNA sequencing was carried out using the PRISM Ready Reaction DyeDeoxy Terminator Cycle Sequencing Kit (Perkin–Elmer) on the ABI Model 377 DNA Sequencer (Applied Biosystems). Sequence analysis was performed using the University of Wisconsin Genetics Computer Group software (Devereux et al., 1984). Comparisons of nucleotide and deduced amino acid sequences were performed on the GenBank and EMBL databases, and on the PIR release 47 and SWISS-PROT release 33 databases, respectively, using the BLAST Network Service of the NCBI (Altschul et al., 1990). Northern blot analysis. Total RNAs from mouse cochlea, eye, brain, liver, kidney, intestine, lung, spleen, heart, and testis were prepared by the guanidium isothiocyanate procedure (Chomczynski and Sacchi, 1987). Three micrograms of poly(A)/ RNA/tissue were electrophoresed in 1% agarose, 2.2 M formaldehyde gels in 3-(Nmorpholino)propanesulfonic acid (Mops)–acetate–EDTA buffer. RNAs were transferred to Hybond-N membranes (Amersham). Membranes were prehybridized in 50% formamide, 51 SSC, 51 Denhardt’s solution, 1% SDS, and 100 mg/ml sonicated herring sperm DNA for 4 h at 427C and hybridized in the same buffer with the 32P-labeled probes at 427C overnight. Membranes were washed in 0.1% SDS, 0.11 SSC at 427C and were autoradiographed using either X-ray film RP2 (Agfa Curix) with intensifying screens at 0807C or PhosphorImager 445 SI and ImageQuantNT software Version 2.00 (Molecular Dynamics). Gene mapping. In situ hybridization experiments were carried out using metaphase spreads from a WMP male mouse, in which all the autosomes except 19 were in the form of metacentric robertsonian translocations. Concanavalin A-stimulated lymphocytes were cultured at 377C for 72 h with 5-bromodeoxyuridine added for the final 6 h of culture (60 mg/ml of medium), to ensure a chromosomal Rbanding of good quality. The recombinant PCR2.1 vectors (Invitrogen) containing the Myo1c and Myo1f cDNA fragments, 800 and 1000 bp, respectively, were tritium-labeled by nick-translocation to a specific activity of 108 dpm/mg. The radiolabeled probes were hybridized to metaphase spreads at final concentration of 100 ng/ml as

gnma

334

CROZET ET AL.

TABLE 1 Fifty Clones Isolated at Random from the Mouse Cochlear cDNA Library Clone

Accession No.

% Identity

Clone

Accession No.

319D3 182G3 272C6

(Z78163) (L22143) (L32836)

25H2 29C9 24G3

(P04634) (Z78141) (A45988)

104H7

(Z78142)

19B8 17G8 20A4 170A9 32B12 20H2

(Z78143) (U28151) (Z78144) (Z78145) (Z78147) (S81177)

None Mouse insulin-like growth factor Mouse S-adenosyl-L-homocysteine hydrolase Rat triacylglycerol lipase precursor None Rat dentin matrix acidic phosphoprotein AG1 Human Coch-5B2 fetal cochlear cDNA product None Mouse Crk-associated substrate None None Pig aconitate hydratase precursor Mouse retinoic acid regulated gene

95% 100%

301A6 103G3 65H4

(Z78150) (Z78151) (L34049)

None None Rat megalin

92%

32B8 28D4 27G8

(A54662) (A27671) (P29995)

90%

302E7

(Z78152)

Mouse myelin PO protein Rat spectrin a chain Rat inositol 1,4,5-trisphosphate receptor None

85% 98%

339B6 367G6 49F5 25A9 26C7 19D5

(Z78153) (Z78154) (Z78155) (X00525) (Z78156) (P16646)

15G2

(L22143)

Mouse insulin-like growth factor

95%

28G6

(PO2463)

17E10 148C6 31B4 29E9 30E12 31B5

(X95466) (X00686) (D14043) (U40952) (Z78148) (Z78149)

Rat CPG2 protein Mouse 18S ribosomal protein Human mRNA for MGC-4 C. elegans C03B1 gene product None None

96% 100% 80% 80%

25F9 22E1 32E4 272D3 28D2 19C12

(Z78157) (Z78158) (Z78159) (P30427) (Z78160) (L27841)

22A3

(X99668)

C. elegans W06E11.4 gene product

75%

25C10

(P48556)

292D9 95E10 169C9

(X80159) (Z49858) (X86691)

Mouse CW17 protein Rat plasmolipin Human Mi-2 protein

95% 90% 95%

20F5 260G4 28F3

(Z78161) (Z78162) (U34960)

29A7

(L13103)

Mouse MEK kinase

98%

189C6

(U20282)

Protein homology

90%

98%

Protein homology

None None None Mouse ribosomal 28S protein None Mouse peripheral myelin protein 22 Mouse procollagen a1(IV) chain precursor None None None Rat plectin None Human autoantigen pericentriol material 1 Human 26S proteasome regulatory subunit C. elegans F08F8.4 gene product None Mouse G protein b 2 subunit protein Mouse stromelysin PDGF protein

% Identity

92% 94% 95% 100%

100% 95% 95%

95% 85% 95% 65% 93% 100%

Note. Homologies of the deduced amino acid sequences with known proteins are indicated. In many cases of homology with nonrodent proteins, the high percentages of amino acid identity suggest that the corresponding cDNAs are expressed from orthologous genes. previously described (Mattei et al., 1985). After being coated with nuclear track emulsion (Kodak NTB2), the slides were exposed for 20 days at 47C and then developed. To avoid any slipping of silver grains during the banding procedure, chromosome spreads were first stained with buffered Giemsa solution and metaphases were photographed. R-banding was then performed with the fluochrome–photolysis–Giemsa (F.P.G.) method, and metaphases were rephotographed before analysis. One hundred metaphase cells were examined for quantification of silver grains associated with chromosomes. Gene localizations on human chromosomes were determined by a PCRbased method, using a pool of 168 whole-genome radiation hybrids, as described (Gyapay et al., 1996). The selected fragments to be amplified were from position 3084 to 3290 and from position 3504 to 3759 of the MYO1C and MYO1F cDNA sequences, respectively. Validation of chromosome assignment was carried out by calculating scores of pairwise linkage between each PCR fragment and the entire set of 404 AFM framework markers (Gyapay et al., 1996).

RESULTS AND DISCUSSION

Generation of a cDNA Library from the Mouse Cochlea A cochlear cDNA library from human has been reported, which was constructed from human fetal tis-

AID

GENO 4526

/

6r26$$$202

02-11-97 13:05:47

sues at 16 – 22 weeks of gestation (Robertson et al., 1994). In the mouse cochlea, the maturation of sensory hair cells is completed by 3 weeks postnatal (Lim and Anniko, 1985). We generated a mouse cochlear cDNA library at Day 16 postnatal, which provides a source for isolating genes that may serve crucial functions in the last steps of inner ear development and in hearing processes (see Materials and Methods). The estimated number of recombinant clones was 2.1 1 106. The analysis of 24 clones randomly selected revealed an average size of the inserts at 1.4 kb. To evaluate this library further, 50 clones were sequenced and compared to the genome databanks (see Materials and Methods). These clones in Table 1 are presented, along with the homologies of their deduced amino acid sequences with known proteins. Eighteen clones corresponded to already known sequences, 13 clones shared similarities with known sequences from either rodent or nonrodent species, and 19 clones exhibited no homology at all. To isolate cDNAs encoding unconventional myosins from this mouse cochlear cDNA library, two cDNA frag-

gnma


335

FIG. 1. (A) Schematic representation of the structure of Myo1c/MYO1C. (B) Nucleotide and derived amino acid sequences of the MYO1C cDNA (GenBank Accession No. X98507). Nucleotide numbering starts from the initiation codon. The conserved ATP- and actin-binding sites are in boldface and shaded. The three IQ motifs are boxed. The open reading frame is terminated by two successive stop codons. Numbers on the left refer to the nucleotide sequence; numbers on the right refer to the amino acid sequence.

ments corresponding to the highly conserved ATP- and actin-binding domains head of the human MYO7A (Weil et al., 1996, and see Materials and Methods) were used to screen 106 clones. Six positive clones were obtained. Upon sequencing, two clones turned out to encode a conventional myosin, the mouse orthologue of

AID

GENO 4526

/

6r26$$$202

02-11-97 13:05:47

the human nonmuscle myosin (NMMHC-A) (data not shown). Two other clones encoded a putative protein, Myo1c, which is likely to be the murine orthologue of the rat and bullfrog myosins Ib. The two remaining clones encoded a novel unconventional type I myosin that was named Myo1f.

gnma

336

CROZET ET AL.

tity, respectively, with the rat myr 2 and bullfrog myosin Ib, which are putative orthologues (Hasson and Mooseker, 1996), and only 49.6% identity with the single identified mouse myosin Ia, recently named Myo1b (Sherr et al., 1993; Hasson et al., 1996). The present mouse myosin was thus tentatively named Myo1c (Hasson et al., 1996). Tissue Expression of Myo1c

FIG. 2. Northern blot analysis of the Myo1c and Myo1f transcripts in adult mouse tissues. (A) Probe corresponding to a coding region of the Myo1c cDNA (nucleotides 1647 to 2568). A unique band of 4.9 kb is detected in all the tissues examined. (B) Probe consisting of 3* coding and noncoding regions of the Myo1f (nucleotides 2205 to 3453). A band (4.5 kb) is detected in all the tissues examined, with the highest levels of expression observed in kidney and liver. In lung, small intestine, and spleen, an additional band (4.4 kb) is observed. (C) Actin probe. Size standards are indicated on the left.

Deduced Primary Structure of the Mouse Myosin (Myo1c) As mentioned above, two overlapping cDNA clones covering 1.6 kb were isolated from the mouse cochlear cDNA library and sequenced. The reconstituted sequence was found to begin 34 nucleotides upstream of the initiation site of translation (ACCATGGG) (Kozak, 1989), but did not contain the region encoding the Cterminal tail domain. To obtain the missing coding sequence, we performed three successive 3* RACE procedures on cochlear RNA. Eight clones were isolated and sequenced. The final consensus nucleotide sequence, obtained from both the cDNA library and the 3* RACE experiments, covered a total of 3332 nucleotides including an open-reading frame of 3084 nucleotides (GenBank Accession No. X99638). Analysis of the deduced amino acid sequence indicated a structure typical of unconventional type I myosins (Fig. 1A). The motor domain sequence harbors the ATP-binding (GESGAGKTE) and actin-binding (PAYIRCIKPN) sites, two motifs present in all myosins. The head and tail domains are separated by three IQ motifs (Fig. 1A). These motifs consist of a basic 23-amino-acid unit, including a consensus IQX3RGX3R sequence thought to be a binding site for light chains of the calmodulin/EF hand superfamily (Espreafico et al., 1992). The tail domain is rich in basic residues and is expected to interact with negatively charged membrane phospholipids (Adams and Pollard, 1989; Doberstein and Pollard, 1992). This myosin exhibits 98.2 and 78.9% amino acid sequence iden-

AID

GENO 4526

/

6r26$$$202

02-11-97 13:05:47

Northern blot analysis of different adult mouse tissues was carried out with a cDNA probe corresponding to the region of the Myo1c tail exhibiting the highest divergence rate with the other myosins (see Materials and Methods). A single band (4.9 kb) was detected in all the tissues analyzed, although with different intensities. High levels of transcript were observed in the lung, kidney, spleen, heart, and testis. The mRNA was expressed at lower levels in the cochlea, eye, small intestine, heart, and spleen, and it was barely detectable in the brain (Fig. 2A). MYO1C, the Human Orthologue of Myo1c To identify the human myosin corresponding to mouse Myo1c, we screened a human kidney cDNA library with the Myo1c cDNA fragment from position 1647 to 2568. Ten clones were isolated and sequenced, among which three were found to contain the entire coding region. The reconstituted sequence of 3384 nucleotides (GenBank Accession No. X98507) contains a single open reading frame of 3084 nucleotides, followed by two successive stop codons. The deduced amino acid TABLE 2 Human Unconventional Myosins I and Their Rodent Orthologues Human Subclass 1 MYO1Da MYO1Ea MYO1F Subclass 2 MYO1Aa MYO1Ba Subclass 3 MYO1C Subclass 4 nd

Mouse

Rat

nd nd Myo1f (90.7%)

nd myr 3a (95.1%) nd

nd Myo1ba (100%)

nd myr 1a (98.9%)

Myo1c (95.5%)

myr 2a (96.4%)

nd

myr 4a

Note. The mouse and rat orthologues of the six human genes encoding unconventional myosins I identified so far are indicated, when known. The rat myosin myr 4 has no identified human equivalent. The complete deduced amino acid sequences of MYO1C, 1E, and 1F are available. For MYO1A, 1B, and 1D, only partial sequences have been determined thus far. Numbers in brackets indicate the percentage of amino acid sequence identity with the corresponding human myosin I. a References: MYO1A, 1B, and 1D (Bement et al., 1994a); MYO1E (Bement et al., 1994b); Myo1b (Sherr et al., 1993); myr 1 (Ruppert et al., 1993); myr 2 (Ruppert et al., 1995); myr 3 (Stoffler et al., 1995), myr 4 (Bahler et al., 1994).

gnma


337

FIG. 3. (A) Alignment of available deduced amino acid sequences from the head domains of the six identified human unconventional myosins I. Amino acids identical in the six myosins (boldface) or in myosins 1D, 1E and 1F (shaded) are indicated. Numbers on the right refer to the position of the last displayed amino acid in the corresponding myosins. The percentages of amino acid identity between these myosins, over the same stretch of sequence, are indicated in (B). Accession Nos.: MYO1A (L29137), MYO1B (L29138), MYO1C (X98507), MYO1D (L29140), MYO1E (U14391), and MYO1F (X98411).

sequence (Fig. 1B) exhibits 95.8% identity with the mouse Myo1c and 96.4% identity with the rat orthologue myr 2 (Table 2). The cDNAs encoding four human unconventional myosins I (MYO1A, B, D and E) have been previously identified (see Table 2), only one of which, encoding MYO1E, was completely sequenced (Bement et al., 1994b). For the other three myosins, MYO1A, B, and D, only short stretches of sequence from the head region, around the conserved ATP-binding site, are available, all of which contain the GESGAGKT and EAFGNAKT sequences (Figs. 3A and 3B) (Bement et al., 1994a). The present human myosin shares only 56% amino acid sequence identity with MYO1E, and comparison with the available sequences of the other human myosins I revealed 41.2% identity with MYO1A, 65.2% identity with MYO1B, and 55.1% identity with MYO1D (Figs. 3A and 3B). This myosin therefore represents a new human unconventional myosin I, which was named MYO1C. Deduced Primary Structure of the Mouse Myosin If (Myo1f) The 3775-nucleotide cDNA sequence obtained from two overlapping clones comprises the entire coding region as well as the 3* and part of the 5* noncoding regions (GenBank Accession No. X97650) (Fig. 4B). The deduced amino acid sequence predicted a protein of

AID

GENO 4526

/

6r26$$$202

02-11-97 13:05:47

1099 amino acids, with a molecular mass of 126 kDa. The initiation site of translation was identified by the presence of the start site consensus sequence (Kozak, 1989), preceded by an in-frame stop codon 60 nucleotides upstream. A typical polyadenylation signal (AATAA) at position 3666 preceded the poly(A) tail by 13 nucleotides. Analysis of the deduced amino acid sequence revealed that the encoded protein exhibits the typical structure of unconventional type I myosins (Fig. 4A). The neck regulatory domain contains a single IQ motif. Within the tail domain, comparison with other myosins I allowed us to distinguish three distinct subdomains, referred to as tail homology (TH) regions, initially described in protozoan myosins I (Pollard et al., 1991). The TH-1 subdomain spans 209 residues and is very basic (pI Å 10.82). This domain is expected to bind anionic membrane phospholipids (Adams and Pollard, 1989; Doberstein and Pollard, 1992). The TH-2 subdomain spans 122 residues, is rich in glycine, proline, and alanine residues (GPA-rich), and is also very basic (pI Å 12.15). This domain contains a putative ATPinsensitive actin-binding site that has been reported in Dictyostelium myosins IB (Rosenfeld and Rener, 1994) and IC (Jung and Hammer, 1994). The presence of two different actin-binding sites may allow these myosins to crosslink actin filaments. The TH-3 subdomain spans 51 residues and is referred to as Src homology region 3 (SH3) (Pawson, 1994). Based on their properties as mediators of protein–protein interactions (Bar

gnma

338

CROZET ET AL.

FIG. 4. (A) Schematic representation of the structure of Myo1f/MYO1F. (B) Nucleotide and derived amino acid sequences of the Myo1f cDNA (GenBank Accession No. X97650). Nucleotides are numbered starting from the initiation codon. The TGA termination codon is indicated by an asterisk. The conserved ATP- and actin-binding sites are in boldface and shaded. The unique IQ motif is boxed. Numbers on the left refer to the nucleotide sequence; numbers on the right refer to the amino acid sequence.

et al., 1993), the SH-3 domains are good candidates for driving the association of the long-tailed protozoan myosins I and of their vertebrate orthologues, with specific membrane compartments.

AID

GENO 4526

/

6r26$$$202

02-11-97 13:05:47

Analysis of the deduced amino acid sequence of this myosin revealed 45.9 and 46.1% identity with the mouse Myo1b and Myo1c, respectively. This suggests that this myosin represents a new identified member

gnma


of the mouse myosin I family, which was named Myo1f (Hasson et al., 1996). This myosin showed 74.5% amino acid sequence identity with the chicken brush border myosin IB, 70.5% with the human MYO1E, and 70.3% with the rat myr 3. Among protozoan myosins I, Myo1f most closely resembles Dictyostelium and Acanthamoeba myosins IB, with 45.5 and 46.7% sequence identity, respectively. However, Myo1f lacks the consensus sequence for phosphorylation in the head region (Brzeska et al., 1989); instead, it exhibits at the corresponding position a negatively charged residue, glutamic acid, also found in most other vertebrate myosins I (Sherr et al., 1993; Stoffler et al., 1995).

339

tails of both MYO1E and MYO1F exhibit the three TH subdomains initially described in protozoan myosins I (Pollard et al., 1991). Therefore, these myosins may represent a subclass of human myosins I (Table 2 and Hasson and Mooseker, 1996). Similarly, the identified 30 amino acids of MYO1A share 91% sequence identity with MYO1B (Figs. 3A and 3B). However, considering the general conservation of head sequences between myosins, especially around the ATP- and actin-binding sites, the identification of closely related myosins in the absence of their complete deduced sequences cannot be accurately ascertained. Upon partial amino acid sequence comparison, MYO1C appears to be the most divergent member of the human myosin I family (Figs. 3A and 3B).

Tissue Expression of Myo1f Northern blot analysis of poly(A)/ RNAs from different mouse tissues was carried out with a cDNA probe corresponding to the region of the Myo1f tail exhibiting the highest divergence rate with other myosins (see Materials and Methods). A unique band with an apparent size of 4.5 kb was observed in the liver, kidney, spleen, eye, brain, lung, small intestine, testis, and cochlea, but was barely detectable in the heart (Fig. 2B). The Myo1f transcript has a high level of expression compared to that of the Myo7A mRNA, which can be detected in these tissues by RT-PCR only (Weil et al., 1995). An additional band, just below the main band, was detected in the lung, small intestine, and spleen (Fig. 2B); this band could correspond either to the transcript from a highly homologous myosin gene or to an alternative splice product of the Myo1f transcript. Because a unique band was observed on Southern blots (data not shown), the latter hypothesis seems more probable. MYO1F, the Human Orthologue of Myo1f To identify the human orthologue of the mouse Myo1f, a human kidney cDNA library was screened with a cDNA probe corresponding to the poorly conserved region of the Myo1f tail (position 2205 to 2933). Three overlapping cDNAs were isolated and sequenced. These clones encompass 3799 bp (GenBank Accession No. X98411) and define a single open reading frame of 3297 nucleotides, encoding a putative protein with a predicted molecular mass of 125 kDa. This human myosin shares 90.7% sequence identity with the mouse Myo1f. Sequence comparison with other human myosins I revealed 69.9% identity with MYO1E and 45.8% identity with MYO1C, as well as 45.5, 63.3, and 96.7% identity with the available sequences of Myo1A, 1B, and 1D, respectively (Figs. 3A and 3B). This myosin thus represents a new human unconventional myosin I, which was named MYO1F. Although the complete sequence of MYO1D is not yet determined, MYO1D, MYO1E, and MYO1F share about 95% amino acid sequence identity over the same stretch of sequence (Figs. 3A and 3B). Moreover, the

AID

GENO 4526

/

6r26$$$202

02-11-97 13:05:47

Gene Mapping As previously discussed, the genes encoding unconventional myosins expressed in the auditory organ may represent candidate genes for hereditary deafness in mammals. Hereditary hearing loss may be divided into syndromic and nonsyndromic (isolated) forms. Several hundred syndromes have been described, in which hearing loss is associated with various anomalies. So far, only 60 genes responsible for such syndromes have been mapped to human chromosomes, half of which are cloned. For inherited isolated deafness, a total of 11 DFNA (autosomal dominant deafness), 10 DFNB (autosomal recessive deafness), and 3 DFN (X-linked deafness) genes have been mapped to human chromosomes at present (see Petit, 1996). MYO1C and MYO1F were mapped to the human chromosomes using PCR-amplified products derived from the 3* noncoding regions of the cDNAs on radiation hybrids (Gyapay et al., 1996; and see Materials and Methods). MYO1C was assigned to human chromosome 17p13, with a maximal lod score of 22.3 obtained for the marker D17S1866. For MYO1F, a maximal lod score of 21.3 was obtained with marker D19S413, and a multipoint analysis led to the mapping of MYO1F between this marker and D19S216, in the 19p13.2–p13.3 chromosomal region (data not shown). Myo1c and Myo1f were mapped by in situ hybridization to the C-E1 and B-C regions of murine chromosomes 11 and 17, respectively (Fig. 5). Taking into account the uncertainty of chromosomal mapping in pericentromeric regions, MYO1C might be considered a candidate gene for DFNB3, which has been mapped in the pericentromeric region of chromosome 17 (Friedman et al., 1995). At present, no human deafness locus has been localized to the same chromosomal region as MYO1F, nor was any deaf mutant mapped to the mouse Myo1f region. However, it is worth noting that in many families with either syndromic or isolated deafness, no cosegregation of the disease with any identified locus has been observed; therefore a significant number of yet unidentified deafness loci may still exist. Hence, the present data should focus the search for new deafness loci to the aforementioned regions.

gnma

340

CROZET ET AL.

FIG. 5. Chromosomal mapping of the mouse Myo1c and Myo1f by in situ hybridization. The labeled sites, indicated by black spots, appear on chromosomes 11 and 17, belonging, respectively, to the Robertsonian translocations ROB (1;11) and ROB (10;17) of the WMP mouse.

ACKNOWLEDGMENTS We thank J. P. Hardelin for helpful discussions and critically reading the manuscript. This work was supported by the A. and M. Such´ tudes sur les ert Foundation, by Groupement de Recherches et d’E Geńomes (GREG No. 30/95), by European Economic Community Grant BMH4-CT95-1324, and by the Association Francaise Retinitis Pigmentosa (AFRP).

REFERENCES Adams, R. J., and Pollard, T. D. (1989). Binding of myosin I to membrane lipids. Nature 340: 565–568. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. (1990). Basic local alignment search tool. J. Mol. Biol. 215: 403– 410. Ashmore, J. (1995). Hearing. A gene for deaf ears? Curr. Biol. 5: 716–718. Avraham, K. B., Hasson, T., Steel, K. P., Kingsley, D. M., Russell, L. B., Mooseker, M. S., Copeland, N. G., and Jenkins, N. A. (1995). The mouse Snell’s waltzer deafness gene encodes an unconventional myosin required for structural integrity of inner ear hair cells. Nature Genet. 11: 369–375. Bahler, M., Kroschewski, R., Stoffler, H. E., and Behrmann, T. (1994). Rat myr 4 defines a novel subclass of myosin I: Identification, distribution, localization, and mapping of calmodulin-binding sites with differential calcium sensitivity. J. Cell Biol. 126: 375– 389. Bar, S. D., Rotin, D., Batzer, A., Mandiyan, V., and Schlessinger, J. (1993). SH3 domains direct cellular localization of signaling molecules. Cell 74: 83–91. Bement, W. M., Hasson, T., Wirth, J. A., Cheney, R. E., and Mooseker, M. S. (1994a). Identification and overlapping expression of multiple unconventional myosin genes in vertebrate cell types. Proc. Natl. Acad. Sci. (USA) 91: 6549–6553. [Published erratum appears in Proc. Natl. Acad. Sci. USA (1994) 91: 11767] Bement, W. M., Wirth, J. A., and Mooseker, M. S. (1994b). Cloning and mRNA expression of human unconventional myosin-IC. A homologue of amoeboid myosins-I with a single IQ motif and an SH3 domain. J. Mol. Biol. 243: 356–363. Brzeska, H., Lynch, T. J., Martin, B., and Korn, E. D. (1989). The localization and sequence of the phosphorylation sites of Acanthamoeba myosins I. An improved method for locating the phosphorylated amino acid. J. Biol. Chem. 264: 19340–19348. Cheney, R. E., Riley, M. A., and Mooseker, M. S. (1993). Phylogenetic analysis of the myosin superfamily. Cell Motil. Cytoskeleton 24: 215–223.

AID

GENO 4526

/

6r26$$$203

02-11-97 13:05:47

Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162: 156–159. Devereux, J., Haeberli, P., and Smithies, O. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12: 387–395. Doberstein, S. K., and Pollard, T. D. (1992). Localization and specificity of the phospholipid and actin binding sites on the tail of Acanthamoeba myosin IC. J. Cell Biol. 117: 1241–1249. El-Amraoui, A., Sahly, I., Picaud, S., Sahel, J., Abitbol, M., and Petit, C. (1996). Human Usher 1B/shaker-1: The retinal phenotype discrepancy explained by the presence/absence of myosin VIIA in the photoreceptor cells. Hum. Mol. Genet. 5: 1171–1178. Espreafico, E. M., Cheney, R. E., Matteoli, M., Nascimento, A. A., De, C. P., Larson, R. E., and Mooseker, M. S. (1992). Primary structure and cellular localization of chicken brain myosin-V (p190), an unconventional myosin with calmodulin light chains. J. Cell Biol. 119: 1541–1557. Friedman, T. B., Liang, Y., Weber, J. L., Hinnant, J. T., Barber, T. D., Winata, S., Arhya, I. N., and Asher, J. H. (1995). A gene for congenital, recessive deafness DFNB3 maps to the pericentromeric region of chromosome 17. Nature Genet. 9: 86–91. Gibson, F., Walsh, J., Mburu, P., Valera, A., Brown, K. A., Antonio, M., Beisel, K. W., Steel, K. P., and Brown, S. D. M. (1995). A type VII myosin encoded by the mouse deafness gene shaker-1. Nature 374: 62–64. Gillespie, P. G. (1995). Molecular machinery of auditory and vestibular transduction. Curr. Opin. Neurobiol. 5: 449–455. Gillespie, P. G. (1996). Deaf and dizzy mice with mutated myosin motors. Nature Med. 2: 27–29. Gillespie, P. G., Wagner, M. C., and Hudspeth, A. J. (1993). Identification of a 120 kd hair-bundle myosin located near stereociliary tips. Neuron 11: 581–594. Gyapay, G., Schmitt, K., Fizames, C., Jones, H., Vega-Czarny, N., Spillett, D., Muselet, D., Prud’Homme, J.-F., Dib, C., Auffray, C., Morissette, J., Weissenbach, J., and Goodfellow, P. N. (1996). A radiation hybrid map of the human genome. Hum. Mol. Genet. 5: 339–346. Hasson, T., Heintzelman, M. B., Santos, S. J., Corey, D. P., and Mooseker, M. S. (1995). Expression in cochlea and retina of myosin VIIa, the gene product defective in Usher syndrome type 1B. Proc. Natl. Acad. Sci. USA 92: 9815–9819. Hasson, T., and Mooseker, M. S. (1996). Vertebrate unconventional myosins. J. Biol. Chem. 271: 16431–16434. Hasson, T., Skowron, J. F., Gilbert, D. J., Avraham, K. B., Perry, W. L., Bement, W. M., Anderson, B. L., Sherr, E. H., Chen, Z. Y.,

gnma

TWO COCHLEAR UNCONVENTIONAL TYPE I MYOSINS Greene, L. A., Ward, D. C., Corey, D. P., Mooseker, M. S., Copeland, N. G., and Jenkins, N. A. (1996). Mapping of unconventional myosins in mouse and human. Genomics 36: 431–439. Hudspeth, A. J., and Gillespie, P. G. (1994). Pulling springs to tune transduction: Adaptation by hair cells. Neuron 12: 1–9. Jung, G., and Hammer, Jr. (1994). The actin binding site in the tail domain of Dictyostelium myosin IC (myoC) resides within the glycine- and proline-rich sequence (tail homology region 2). FEBS Lett. 342: 197–202. Kozak, M. (1989). The scanning model for translation: An update. J. Cell Biol. 108: 229–241. Lim, D. J., and Anniko, M. (1985). Developmental morphology of the mouse inner ear. A scanning electron microscopic observation. Acta Otolaryngol. Suppl. (Stockholm) 422: 1–69. Mattei, M. G., Philip, N., Passage, E., Moisan, J. P., Mandel, J. L., and Mattei, J. F. (1985). DNA probe localization at 18p113 band by in situ hybridization and identification of a small supernumerary chromosome. Hum. Genet. 69: 268–271. Metcalf, A. B., Chelliah, Y., and Hudspeth, A. J. (1994). Molecular cloning of a myosin I beta isozyme that may mediate adaptation by hair cells of the bullfrog’s internal ear. Proc. Natl. Acad. Sci. USA 91: 11821–11825. Mooseker, M. S., and Cheney, R. E. (1996). Unconventional myosins. Annu. Rev. Cell Dev. Biol. 11: 633–675. Pawson, T. (1994). SH2 and SH3 domains in signal transduction. Adv. Cancer Res. 64: 87–110. Petit, C. (1996). Genes responsible for human hereditary deafness: symphony of a thousand. Nature Genet. 14: 385–391. Pollard, T. D., Doberstein, S. K., and Zot, H. G. (1991). Myosin-I. Annu. Rev. Physiol. 53: 653–681. Robertson, N. G., Khetarpal, U., Gutierrez, E. G., Bieber, F. R., and Morton, C. C. (1994). Isolation of novel and known genes from a human fetal cochlear cDNA library using subtractive hybridization and differential screening. Genomics 23: 42–50. Rosenfeld, S. S., and Rener, B. (1994). The GPQ-rich segment of Dic-

AID

GENO 4526

/

6r26$$$203

02-11-97 13:05:47

341

tyostelium myosin IB contains an actin binding site. Biochemistry 33: 2322–2328. Ruppert, C., Kroschewski, R., and Bahler, M. (1993). Identification, characterization and cloning of myr 1, a mammalian myosin-I. J. Cell Biol. 120: 1393–1403. Ruppert, C., Godel, J., Muller, R. T., Kroschewski, R., Reinhard, J., and Bahler, M. (1995). Localization of the rat myosin I molecules myr 1 and myr 2 and in vivo targeting of their tail domains. J. Cell Sci. 108: 3775–3786. Sherr, E. H., Joyce, M. P., and Greene, L. A. (1993). Mammalian myosin I alpha, I beta, and I gamma: New widely expressed genes of the myosin I family. J. Cell Biol. 120: 1405–1416. Solc, C. K., Derfler, B. H., Duyk, G. M., and Corey, D. P. (1994). Molecular cloning of myosins from the bullfrog saccular macula: A candidate for the hair cell adaptation motor. Auditory Neurosci. 1: 63–75. Steel, K. P. (1995). Inherited hearing defects in mice. Annu. Rev. Genet. 29: 675–701. Stoffler, H. E., Ruppert, C., Reinhard, J., and Bahler, M. (1995). A novel mammalian myosin I from rat with an SH3 domain localizes to Con A-inducible, F-actin-rich structures at cell–cell contacts. J. Cell Biol. 129: 819–830. Usher, C. (1913/1914). On the inheritance of retinitis pigmentosa with notes of cases. R. London Ophthalmol. Hosp. Rep. 19: 130– 236. Weil, D., Blanchard, S., Kaplan, J., Guilford, P., Gibson, F., Walsh, J., Mburu, P., Varela, A., Levilliers, J., Weston, M. D., Kelly, P. M., Kimberling, W. J., Wagenaar, M., Levi-Acobas, F., Larget-Piet, D., Munnich, A., Steel, K. P., Brown, S. D. M., and Petit, C. (1995). Defective myosin VIIA gene responsible for Usher syndrome type 1B. Nature 374: 60–61. Weil, D., Le´vy, G., Sahly, I., Levi-Acobas, F., Blanchard, S., El-Amraoui, A., Crozet, F., Phillipe, H., Abitbol, M., and Petit, C. (1996). Human myosin VIIA responsible for the Usher 1B syndrome: a predicted membrane-associated motor protein expressed in the developing sensory organ. Proc. Natl. Acad. Sci. USA 93: 3232–3237.

gnma