Twister mutant mice are defective for otogelin, a ... - Springer Link

Original Contributions Incorporating Mouse Genome

Mammalian Genome 11, 961–966 (2000). DOI: 10.1007/s003350010197

© Springer-Verlag New York Inc. 2000

Twister mutant mice are defective for otogelin, a component specific to inner ear acellular membranes Marie-Christine Simmler,1 Ingrid Zwaenepoel,2 Elisabeth Verpy,2 Laurent Guillaud,1 Colette Elbaz,1 Christine Petit,2 Jean-Jacques Panthier1 1

UMR 955 INRA de Geńe´tique Molećulaire et Cellulaire, Ecole Nationale Ve´te´rinaire d’Alfort, 7 avenue du Geńe´ral de Gaulle, 94704 Maisons-Alfort Ce´dex, France 2 Unite´ de Geńe´tique des De´ficits Sensoriels, CNRS URA1968, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris Ce´dex 15, France Received: 31 May 2000 / Accepted: 7 July 2000

Abstract. Deafness is a common sensory defect in human. Our understanding of the molecular bases of this pathology comes from the study of a few genes that have been identified in human and/or in mice. Indeed, deaf mouse mutants are good models for studying and identifying genes involved in human hereditary hearing loss. Among these mouse mutants, twister was initially reported to have abnormal behavior and thereafter to be deaf. The recessive twister (twt) mutation has been mapped on mouse Chromosome (Chr) 7, homologous to the long arm of human Chr 15 (15q11). Otog, the gene encoding otogelin, a glycoprotein specific to all the acellular membranes of the inner ear, is also localized to mouse Chr 7, but in a region more proximal to the twister mutation, homologous to the short arm of human Chr 11 (11p15) carrying the two deafness loci, DFNB18 and USH1C. Mutant mice resulting from the knockout of Otog, the Otogtm1Prs mice, present deafness and severe imbalance. Although twt had been mapped distally to Otog, these data prompted us to test whether twt could be due to a mutation in the Otog locus. Here, we demonstrate by genetic analysis that twt is actually allelic to Otogtm1Prs. We further extend the phenotypical analysis of twister mice, documenting the association of a severe vestibular phenotype and moderate to severe form of deafness. Molecular analysis of the Otog gene revealed the absence of detectable expression of Otog in the twister mutant. The molecular and phenotypical description of the twt mouse mutation, Otogtwt, reported herein, highlights the importance of the acellular membranes in the inner ear mechanotransduction process.

Introduction In developed countries, it has been estimated that two-thirds of the cases of nonsyndromic (isolated) deafness in childhood have genetic causes, most of them being associated with inner ear defects (Denoyelle et al. 1999). It has also been shown that deafness in human and several mouse mutants are underlined by defects in orthologous genes (Steel 1995; Avraham 1998; Petit et al. 2000). As a result, deaf mouse mutants are helpful models to investigate the pathology of the various human gene defects. The mammalian inner ear is composed of two organs, the cochlea for hearing and the vestibule for balance. It contains six sensory areas, the organ of Corti in the cochlea, the saccular and utricular maculae and the cristae ampullares of the three semicir-

Correspondence to: M.-C. Simmler; E-mail: [email protected]

cular canals in the vestibule. Each sensory patch is composed of a sensory epithelium overlaid by an acellular gelatinous membrane, i.e., the tectorial membrane, in the cochlea; the otoconial membranes, in the saccule and utricle and the cupulae, in the semicircular canals, in the vestibule. In the saccule and utricle, otoconial membranes are loaded with biominerals, named the otoconia. When sound vibration reaches the cochlea and gravity or movements of the head exert forces on the vestibule, the displacement of these gelatinous membranes relative to the sensory epithelium leads to the deflection of the stereocilia of the sensory hair cell, which in turn opens the transduction channels, inducing the hair cell depolarization and the synaptic activity (Hudspeth 1989). Acellular membranes of the mammalian inner ear are composed of collagenous and non-collagenous glycoproteins and proteoglycans. Otogelin is one of the non-collagenous glycoproteins that is specifically expressed in all of the inner ear acellular membranes (Cohen-Salmon et al. 1997). We have recently shown that homozygous mice lacking otogelin, Otogtm1Prs mice, display severe imbalance and progressive hearing loss. Based on the histopathological analysis of the Otogtm1Prs mice, a role for otogelin in the resistance of the tectorial membrane to sound stimuli and in the anchoring of the vestibular acellular membranes to the underneath sensory epithelium was assigned (Simmler et al. 2000). Twt (twister) is a recessive mutation that arose spontaneously in a C57BL/6xC3Heb/Fe hybrid stock (Lane 1981). Twister mouse mutant was harvested because of its abnormal behavior, typical of a vestibular dysfunction. It was at first not reported as deaf and is still not described as a deaf mutant, in the Mouse Locus Catalog (http://www.informatics.jax.org). However, in 1988, deafness in twister was mentioned, and the data were later quoted (Steel 1995). The twt mutation is linked to the pink-eyed dilution locus, p, with a 95% upper confidence limit for recombination between p and twt, at 4.08% (Lane 1981). This would map twt to mouse Chr 7, at 28 cM from the centromere. Because of the phenotypic similarities, associating vestibular and hearing defects between twister and Otogtm1Prs mice, we suspected that twt could be allelic to Otogelin (Otog). Indeed, owing to the low resolution of the mapping genetic data, twt could actually lie more proximally than reported, that is, in the region carrying the Otog gene (Cohen-Salmon et al. 1999). We report here the full description of the phenotype of twister mice and document their inner ear histopathology, which was compared with that of Otogtm1Prs. We provide genetic evidence showing that twt is allelic to Otog. Finally, we analyzed the expression of Otog and of the neighboring Ush1C gene, the homologous gene for the recently discovered USH1C human deafness gene (Verpy et al. 2000), in twister mouse mutants.

962

M.-C. Simmler et al.: twt is allelic to Otog

Table 1. PCR primer pairs used to study the otog locus. Position 1 refers to the start of the Otog sequence. The translation start point is at position 43, and the stop codon at position 8773. Series number 1 2 3 4 5 6 7 8 9

Primer

Sequence (5⬘-3⬘)

Location (5⬘ position)

BordG-B Muc4-lo Muc1-up L235 L68-8-lo Otog9-lo-s2 BordD-A L117Muc Otog8-up Otog8-lo Otog7-up Otog7-lo Otog5-up Otog5-lo Otog4-up Otog4-lo Otog1-up-s2 Otog1-lo

GATGTATGCCTGGGGAAATAACTT CCCTAAGTAGACAAAAGCAAGAGAGAAGAAGCAAGC CCCTGGGCCAGGCTCTCTGATTGC CTCCTGGCCTTCCATAACTGTTGTG AATGCCACTGGACCCCGTTGC GGCAGGTGTAGTGGGCAGAGCC GGCATCCGTCAGCCTTCACCC TCACACCGCTCATACACACCCTGCATGG GACAGGTGACCCTGACCCAGGCAGGA GTGCTGGCATGCTCAGGTTCAGCAGC GCCGAGCCCCAGCTGCTGCATGTCCA GCACTCACAGTCGCCACCCCGGCTGC GTCCCACGGAGACCCTTGGCAATGAG CCGGTCTGTGAGCACCTGTGGATGGA AAGCCCAGCCCCCATCGACCTGCCTC TCGGCAGGCACATTCCCATTGTGGGC TATGACCACCCACGGGACCTGGCG AGCTGGAAGGAGAACATGAAAGTGAC

-1956 103 8 257 395 622 764 1055 1742 2761 2633 3639 4412 5371 5250 6330 8233 8834

Materials and methods Phenotypical analysis. All protocols are fully described by Steel and Harvisty (1996). Briefly, the reaching response is considered as abnormal if the mice curl up towards their tail, when held by the tail above a flat surface. The elevated platform test is used by recording the time for the mouse to fall off a platform, on a soft surface. The contact-righting reflex consists of placing the mouse in a supine position in a long transparent tube. It is noted whether the animal returns to its normal upright position immediately or steps or walks along the upper surface of the tube. In the swimming test, the behavior of the mice, placed in a tank filled with warm and tepid water, is observed. The Preyer Reflex, which is a startle reflex of the pinna under a suprathreshold sound stimulus, is used to detect mice with severe/profound hearing impairment. Responses were recorded in seven twister mutants (−/−), eight heterozygotes (+/−), and ten C57BL/ 6xC3Heb/Fe wild-type (+/+) mice, at the same age. Responses were recorded twice.

Auditory-evoked brainstem responses (ABR). ABR recording was as described by Simmler et al. (2000). In brief, stimuli (test employed a broad-band click stimulus or pure tone pips at 8, 16, and 32 kHz each (Erway et al. 1993) were produced by computer (Tucker-Davies) and sound delivered into the ears of mice that were anesthetized with avertin and kept warm. For recording ABR, subdermal electrodes were inserted behind the ear pinna of the side to be recorded (active), on the top of the head (vertex), and at the front of the head (ground). The minimum and maximum intensities were 30 and 100 dB SPL (sound pressure level) respectively, with steps of 5 dB. The degree of hearing loss in Otgn−/− mice was defined according to ABR thresholds in mice (Erway et al. 1993): mild, intermediate, and severe/profound correspond to 20–40 dB SPL, 41–60 dB SPL, and 61–99 dB SPL above the mean threshold levels in Otgn+/+, respectively. ABR responses were recorded from five twister mice (−/−), four heterozygotes (+/−), and three C57BL/6xC3Heb/Fe wildtype (+/+) mice.

Expression analysis. Expression of Otog was assayed by RT-PCR on cochlear RNA (RNeasy kit, Qiagen) from 4-day-old twister and C57BL/ 6xC3He mice. Primers exon 1 (5⬘-TTGAGTCCACTGCAGAGGTGCAGCACCAG-3⬘) and L68.8.200 (5⬘-CTCCACGTGGTGCTGCCC-3⬘) were used to amplify the 5⬘ end of the cDNA (383 bp amplicon) (cDNA position 104–487). Primers otog6-up (5⬘-GGTGTTCGAGACCTGCCACCCAGTGG-3⬘) and otog6-lo-s1 (5⬘-TGCTGCTCGGCACTGGTGGTGG3⬘) were used to test the central part of the cDNA (944 bp amplicon) (cDNA position 3540–4484; Genbank accession n° U96411; CohenSalmon et al. 1997). Amplification of the inner ear specific Otoconin-95 (Ocn-95) transcript with primers 5⬘-GTGTGACAAGGCTGCTGTGGAGTGC-3⬘ and 5⬘-CTCCGTCTGGTGACTTGAGGCTTTG-3⬘ was carried out as a RT-PCR positive control (788 bp amplicon; cDNA position 695–1483; Genbank accession No. AF093591; Verpy et al. 1999). The

Otogelin cDNA was detected after transfer and hybridization of the PCR product with either an internal cDNA probe (cDNA position 8–258) or a specific internal oligonucleotide (5⬘-GCACTCACAGTCGCCACCCCGGCTGC-3⬘). Ocn-95 cDNA was detected directly by ethidium bromide staining on the agarose gel. The expression of Ush1C was tested by RTPCR on cochlear RNA from homozygous 4-day-old twister or wild-type inner ears with primers 5⬘-GAAGGCTGCCGAGGAGAATGAG-3⬘ and 5⬘-CTGCGATCTGCTCTGGCGAGAA-3⬘ (296 bp amplicon; Verpy et al. 2000). Amplicons were detected by ethidium bromide staining on the agarose gel.

Examination of the Otog locus. To grossly verify the structure of the Otog locus, we tested twister and C57BL/6xC3He DNA by a long-range PCR (Roche) approach. The 9 pairs of primers (out of 12 pairs, see below) that successfully amplified both DNA are presented in Table 1 and Figure 2B. The resulting amplicons were visualized by ethidium bromide staining on an agarose gel. Three regions were not amplifiable under the experimental conditions used in this study. Indeed, one region covered by the primer pair series Otog2 (positions 7092–8036 within the cDNA) was not amplified from either DNA, most probably owing to the presence of a very large intron. Two other regions covered by the two following pairs of primers, the Otog9 (positions 870–1874) and the Otog3 (positions 6223– 7220) series, were amplified only from wild-type DNA and not from twister DNA. Following Southern blot hybridization analysis (see below), we suggest that the lack of twister PCR products is owing to a technical failure of the long-range PCR, known to be highly sensitive to the quality of the DNA. The genomic region for Otog was thus further investigated by Southern blot analysis of EcoRI- and BstEII-restricted fragments from twister and C57BL/6xC3He DNA, hybridized with nine probes (see Figure 2C). Seven of them were purified Otog cDNA fragments. The nucleotide positions within the Otog gene sequence were the following: probe 1, nucleotides (nt) 8–1006; probe 2, nt 1006–1464; probe 3, nt 1440–1954; probe 4, nt 2008–3340; probe 6, nt 3317–7184; probe 8, nt 7145–7970; and probe 9, nt 7904–8833. The two other probes were PCR-amplified C57BL/ 6xC3He genomic DNA fragments using exonic primers: probe 5 is a 2.8-kb fragment obtained with otog6-up-s2 (5⬘-CCCATGACCCTGACGTGGTGTC-3⬘) (position 3911) and otog6-lo-s1 (5⬘-TGCTGCTCGGCACTGGTGGTGG-3⬘) (position 4463), and probe 7 is a 12-kb fragment obtained using Otog3-up (5⬘-ACCCAGCACTGTCCTCAGGGTGCCGT-3⬘) (position 6223) and Otog3-lo (5⬘-AGCACCTGGCACTGCTCTGGGTCCAG3⬘) (position 7220) (see Fig. 2C). Probes 5 and 7 were used with an excess of unlabeled total mouse DNA as a competitor. Hybridizations were carried out in 0.5 M sodium phosphate, pH 7.2, 1mM EDTA, and 7% SDS at 65°C, modified from Church and Gilbert (1984). Blots were washed in 2 × SSC, 0.1% SDS or 0.5 × SSC, 0.1% SDS at 65°C.

Histology. Dissected inner ears from P4 mice were fixed in 4% paraformaldehyde/PBS buffered solution for 48 h, at 4°C. For adult stages (P170–P200), dissected inner ears were decalcified for 3 days in a 16.8% EDTA solution, after a 1-h fixation step, then post-fixed for 24 h. Inner ears


963

Table 2. ABR thresholds for wild-type C57BL/6xC3Heb/Fe (+/+), heterozygotes (+/−), and twister (−/−) mice. Stimulus frequency click

8 kHz

16 kHz

32 kHz

Genotype

Number of ears recorded

Mean ABR (dB SPL)

Standard deviation

+/+ +/− −/− +/+ +/− −/− +/+ +/− −/− +/+ +/− −/−

6 8 10 5 8 10 5 8 10 5 8 10

30 31.9 79 31.7 35.6 93.1 38 40 100 40 35 98

0 3.7 9.2 4.1 1.8 9.6 5.7 2.5 0 7.9 2.7 5.3

were embedded in paraffin and sectioned at 5␮m. Sections were mounted on microscope slides (SuperFrost, Fisher) and counterstained with hematoxylin-eosin and safran.

Results Twister mice have severe imbalance. Lane reported in 1981 a new recessive mutation, named twister (twt), that induced an abnormal behavior in all affected mice, such that when picked up by the tail, mutant mice tuck their head as if to somersault. Some had a tilted head, some showed a circling behavior, and none could swim. Deafness remained undetected at that time (Lane 1981). Later, twister mice were described as deaf mutants (Steel 1995). Here, we fully describe the phenotype of twister mutant mice. Twister mice showed sign of postural abnormal control as early as 4 days of age (P4) by exhibiting head and trunk curling when picked up by their tail. P4 twister mice also tended to stay on their back when placed upside down. This extremely slow, and even uncontrolled, postural response was never observed in wild-type pups. After 2 weeks of age, the majority of twister mice had tilted head on one side. They also showed an abnormal response in elevated platform, air righting reflex, and contact righting reflex tests (see Materials and methods). In some cases, twister mice showed hyperactivity and circling behavior. All mutants adopted a swimming, circling style with tilted head, but they did not sink. Heterozygotes (twt/+) behave normally in all tests. Mutant mice appeared otherwise healthy after birth. Both sexes were fertile, although twister mice were poor breeders compared with wild-type mice. Twister mice are deaf. To assess gross hearing defects, Preyer reflex was tested in twister mice and compared with that of wildtype animals or heterozygotes at the same age. At 3–4 weeks, a normal response was observed in all three genotypes. All mice also had a body startle reflex showing that the acoustic fibers influencing motoneurones responded. At 10–20 weeks, a weak Preyer reflex was detected in five out of seven twister animals. Two animals that did not react were the two oldest mutants. None of the twister mice exhibited a body startle reflex. In contrast, wild-type and heterozygous mice exhibited a normal Preyer reflex. We carried out click- and pure tone-evoked auditory brainstem response (ABR) in five twister mutants, four heterozygous mice, and three wild-type animals, at 10–20 weeks. Both ears were tested individually. In twister mice, moderate and severe/profound hearing loss was recorded in eight and two ears, respectively (n ⳱ 10 ears recorded). The two most affected ears were from two different mutants. The threshold sound pressure levels (SPL) required for detection of an ABR were elevated in twister mutants both for a click stimulus and for all pure tone tested (Table 2). The decrease in hearing sensitivity for a click stimulus averaged 50 dB (79 ± 9.2 dB, range of 65–95 dB) in twister mice when compared with wild-type mice.

Fig. 1. RT-PCR analysis of the inner ear showing the absence of the Otog transcript in twister mice. PCR products were blotted and hybridized with a radiolabeled probe or oligonucleotide specific to Otog (Cohen-Salmon et al. 1997): lanes 1 and 2, twister cDNA; lanes 3 and 4, wild-type cDNA; lanes 1 and 3, without RT; lanes 2 and 4, with RT. (A) PCR with primers in the 5⬘ end of the gene. (B) PCR with primers in the central part of the gene. In both cases, the lack of expression was confirmed by a long-term exposure. The lower bands in A and B correspond to migration of the primers. (C) PCR with primers in the Ocn-95 gene (Verpy et al. 1999). (D) PCR with primers in the Harmo gene (Verpy et al. 2000).

twt is allelic to Otogtm1Prs. The twt mutation was localized on mouse Chr 7 in the vicinity of Otog. The phenotype of the twister mice clearly resembled that observed in mice homozygous for the null mutation for Otog, Otogtm1Prs. Based on this consideration, we suspected that the twt mutation could disrupt the Otog gene. To test this hypothesis, we generated intercrosses of twister and Otogtm1Prs mice. All members of their offspring exhibited abnormal behavioral responses and a weak Preyer reflex (n ⳱ 13 offspring in two crosses). Furthermore, out of 33 offspring of crosses between twt/+ and Otogtm1Prs/+, 7 mice exhibited a mutant phenotype, while 26 mice behaved normally. We thus concluded that twt is allelic to the null mutation at the Otog locus. Otog gene expression analysis in twister mice. We tested the Otog gene expression by RT-PCR analysis of mRNA isolated from twister inner ears, using primers corresponding to the 5⬘ end (Fig. 1A), and the central part of the cDNA (Fig. 1B) (Cohen-Salmon et al. 1997). RT-PCR controls by using an inner ear-specific gene, Ocn-95 (Verpy et al. 1999), are presented on Fig. 1C. In twister mice, the Otog gene expression was abolished (see Fig. 1A and 1B). We have also studied the expression of the recently identified Ush1C deafness gene, which is located in close proximity to the Otog gene (Verpy et al. 2000). As in Otogtm1Prs mice (data not shown), Ush1C is normally expressed in twister mice inner ears (Fig. 1D). Genomic study of the Otog locus in twister mice. A long-range PCR-based (LR-PCR) approach with a set of nine primer pairs was undertaken to evaluate the integrity of the Otog locus. This series of primer pairs is covering most of the Otog gene (see Table 1 and Fig. 2B). PCR amplification of a 2-kb fragment located upstream of the first exon and covering the first 103 bp of the Otog gene showed that the absence of the mRNA was not due to any large deletion in the putative promoter region. Similarly, DNA from wild-type and twister mice was amplified, with the set of primer pairs described on Table 1 and Fig. 2B (data not shown). Because three regions were refractory to amplification (see Fig. 2B), we further investigated the Otog locus by a Southern blot hybridization analysis. We used BstEII and EcoRI DNA restrictions and

964


Fig. 2. Positions of PCR primers and probes used in the study of the Otog locus. (A) Schematic representation of the Otog cDNA sequence. (B) Distribution of the series of Otog primers used in the long-range amplification analysis of twister and control DNA (see details for positions in Table 1). Stars indicate the three different unamplifiable regions. (C) Distribution of the probes used for Southern blot analysis. The seven cDNA probes (filled boxes) are cDNA amplified fragments with Otog primers located at the following positions: probe 1, nt 8–1006; probe 2, nt 1006–1464; probe 3, nt 1440–1954; probe 4, nt 2008–3340; probe 6, nt 3317–7184; probe 8, 7145–7970; and probe 9, nt 7904–8833. The two genomic probes (lines with two arrows), i.e., probes 5 (a 2.8-kb fragment) and 7 (a 12-kb fragment), are genomic amplified DNA fragments with Otog primers located at positions 3911–4463 and 6223–7220 of the Otog cDNA sequence, respectively.

nine probes, fully covering the Otog gene (see Fig. 2C). Southern blot analysis of EcoRI-digested genomic DNA allowed us to estimate the size of the genomic region containing the Otog locus, to approximately 65 kb. Moreover, no size differences of the EcoRI restriction fragments were detected between the mutant and the wild-type controls, suggesting no large deletions within the genomic region (data not shown). Interestingly, BstEII digestions revealed a difference in banding pattern between twister and wildtype DNA. Indeed, a 8.9-kb fragment detected by probes 6 and 7 was absent, specifically in twister mice DNA (Fig. 3). The loss of the 8.9-kb BstEII fragment was associated with the appearance of two denser bands, i.e., of two duplicated fragments, one of 2.6 kb (see Fig. 3B) and one of 6.3 kb (see Fig. 3C). Additional data using a series of primers (7, 8, and 9; see Table 1 and Fig. 2B) confirmed that no other rearrangements exist within the 3⬘ end of the Otog gene. Altogether, Southern blot analysis data of all Otog probes, including the two adjacent probes 5 and 8 (see Fig. 3A and 3D), suggests that a point mutation or a discrete rearrangement, creating a new BstEII site and mapping to an approximately 12-kb genomic region of the Otog gene, is present within the 3⬘ part of the Otog locus. Data from the orthologous OTOG human locus (see Genbank accession No. AC005137, human Chr 11p14.3 PAC clone 6-106f23) are indeed concordant with the presence of small exons and large introns within this same region. This suggests that the 12-kb region contains a small amount of exonic sequences (corresponding to positions 6223–7145 of the Otog cDNA) and approximately 11 kb of intronic sequences. The twister mutant shares similar inner ear histopathological features with the Otogtm1Prs mice. We have studied the effects of the twt mutation on the morphology of the inner ear, by histological examination of 4-day-old (P4) twister mice. P4 mice have been chosen, because at this age evident morphological defects are already seen in Otogtm1Prs mice. On Fig. 4A is presented the morphology of the vestibule of a C57BL/6xC3Heb/Fe wild-type inner ear. All acellular membranes are normally covering the corresponding sensory epithelium. By contrast, the twister mutant (see Fig. 4B) exhibited abnormal position of the two otoconial membranes, particularly in the utricle, where the membranes are completely detached. In the saccule, the otolithic membrane is still partly attached to the sensory epithelium (Fig. 4B). In the cristae ampullares, the cupulae are normally present, as in the control inner ear (compare Fig. 4A and 3B). However, cupulae were detached in older twister mutant mice (P170; data not shown). Figure 4C and 4D present the morphology of the cochlea of wild-type and twister inner ear: in the normal cochlea (Fig. 4C), the tectorial membrane is attached to the underneath sensory epithelium, showing several sites of attachment, along the developing organ of Corti. An apparent similar structure has also been observed in the twister inner ear (Fig. 4D). Besides, as in the control inner ear,

Fig. 3. Southern blot analysis of the Otog locus. Genomic DNA from wild-type (+/+) and twister (twt) mice were digested with BstEII and hybridized with the following probes: (A) genomic probe 5 (a 2.8-kb fragment); (B) cDNA probe 6; (C) genomic probe 7 (a 12-kb fragment); (D) cDNA probe 8. The two asterisks point to the two duplicated 2.6 and 6.3 kb fragments, in B and C, respectively. Molecular fragment sizes are from HindIII-digested lambda DNA.

twister inner ear contains in all turns of the cochlea a normally differentiated sensory epithelium. Finally, even in older animals, the tectorial membrane was apparently normal, as observed in Otogtm1Prs inner ears (data not shown). Discussion Our interest in the twt mutation has been stimulated both because of its localization near the Otog gene on mouse Chr 7, and its phenotype, associating vestibular and hearing defects. The full


965

Fig. 4. Comparative sections of the vestibule and the cochlea in wild-type (A, C) and twister (B, D) mice, at P4: Hematoxylin-eosin and safran colorations. (A) The section shows the sensory epithelium of the utricle (U) and the saccule (S), as well as a crista ampullaris (CA) in a wild-type animal, normally covered by an otoconial membrane (arrow). (B) The section shows that in the saccule (S), the otoconial membrane is slightly displaced (arrow); in the utricle (U), the otoconial membrane is present but

fully detached and not visible in the presented field; however, it is clearly present, as shown in the inset; the two cristae ampullares (CA) are normally covered by a cupula (arrow). (C) and (D) These two figures show that, despite a slight difference in the plane of the sections, both the tectorial membrane (arrowhead) and the sensory epithelium in the cochlea of P4 twister inner ear have a development similar to the wild-type inner ear, at the same age.

phenotypical analysis of twister led us to consider that twt could be allelic to Otogtm1Prs. Breeding studies between twister and Otogtm1Prs mice revealed that twt and Otogtm1Prs fail to complement each other. Expression data showed the absence of the Otog transcript in twister inner ears. Taken together, our observations demonstrate that the twt mutation is allelic to the Otog locus. The absence of the Otog gene expression is not owing to a large genomic rearrangement, since long-range PCR or Southern blot hybridization analysis did not reveal gross differences in the amplicons or restriction fragments pattern between twister and wild-type C57BL/6xC3Heb/Fe DNA. On the contrary, on the basis of the distinct genomic DNA restriction pattern modification, we suggest that a point mutation or the presence of a non-sizable rearrangement, possibly involving part of a retroviral-like insertion (Johnson et al. 1999), could affect the transcription process and/or the stability of the transcript in twister mice DNA. Alternatively, the mutation underlying the difference in banding pattern could be fully unrelevant to the transcription of the Otog gene. Nevertheless, we cannot rule out the possibility that a point mutation within the putative promoter abolishes the transcription or that a mutation within a splice site or a nonsense mutation or a frameshift bringing a premature stop codon into register within the 8730 bp of the Otog

gene could be responsible for the absence of Otog transcript in twister mice [see Culbertson (1999)]. Twister phenotype is very similar to that of Otogtm1Prs mice. Firstly, twister and Otogtm1Prs mutant mice are both severely imbalanced. In both, a defect in the anchoring of the gelatinous structure to the underlying sensory epithelium may account for the balance dysfunction, since in the two mutants the otoconial membranes are detached or are starting to detach from the underlying sensory epithelium a few days after birth. Similarly, in both mutants, the cupulae are also detached, although, in both cases, detachment occurs later than for the otoconial membranes. Secondly, twister and Otogtm1Prs mutant mice both have a progressive, moderate to severe/profound hearing loss. As in Otogtm1Prs mice, the tectorial membrane of twister inner ear appears morphologically normal at any stage of development, even when deafness is established. The deafness is probably due, as proposed for the Otogtm1Prs mice, to ultrastructural defects in the fibrillar network organization, which affects the resistance of the tectorial membrane to sound stimuli. Despite this great similarity, the phenotypes of twister and Otogtm1Prs mice are not identical. Some of the few differences noticeable are minor variations in the behavior of twister and Otogtm1Prs mice. Firstly, typical waltzer behavior cor-

966

responding to circling can be observed in twister mice, whereas it is not observed in Otogtm1Prs mice. Secondly, twister mutants in the swimming test adopt a circling style with tilted head; they do not exhibit an underwater style nor do they sink, as Otogtm1Prs mice always do. Furthermore, a slight variation in the degree of hearing loss, between twister and Otogtm1Prs mice, is noted. Twister mice (n ⳱ 10) show moderate hearing loss in the majority of the cases, while most of the Otogtm1Prs mice (n ⳱ 20) show severe/profound deafness at the same age (p ⳱ 0.001). In order to test whether this slight phenotypical variability was influenced by a defect in the expression of the neighboring deafness gene, Ush1C, which is responsible for Usher syndrome type 1C in human (Verpy et al. 2000), we tested Ush1C gene transcription in twister mice. Expression of Ush1C gene is normal in twister mutants. As Otogtm1Prs mice were obtained by a targeting gene strategy for the Otog locus implicating homologous recombination events within the putative promoter region common to these two head-to-head genes, Otog and Ush1C, the transcription of Ush1C, in Otogtm1Prs mice has been examined. Expression of Ush1C is normal in Otogtm1Prs mutants. Therefore, a defect in the neighboring Ush1C gene is not responsible for the higher severity of the phenotype in Otogtm1Prs mice. Since twister is kept in a C57BL/ 6xC3Heb/Fe hybrid stock while Otogtm1Prs was established in the 129/SvPas substrain, it is possible that the slight differences in the phenotypes of twister as compared with Otogtm1Prs mice may be accounted for by hearing modifier genes that may interact with the twt mutation. Indeed, genes such as mdfw, a modifier of deaf waddler (dfw; Noben-Trauth et al. 1997), or moth1, a modifier of tubby (tub; Ikeda et al. 1999) have been previously described and found interacting and affecting hearing in mice. In conclusion, description of forms of deafness implicating mutations in genes coding for components of the inner ear acellular membranes, both in human and mice (Verhoeven et al. 1998; Alloisio et al. 1999; McGuirt et al. 1999; Mustapha et al. 1999; Simmler et al. 2000; and herein) emphasize the importance of genes involved in the normal architecture of acellular membranes, for hearing and ear-balance functions. Acknowledgments. This work was supported by the Centre National de la Recherche Scientifique (CNRS) (Action Concerteé Biologie cellulaire, grant 96114) and European Community (EC) (grant QLG2-CT-199900988). M.-C. Simmler is CR at the CNRS. We are grateful to Sylvie Compain and Michel Leibovici for gifts of Otog cDNA fragments, Ste´phane Blanchard for sequencing, and Corinne Koenen for care of the animals. We thank Martine Cohen-Salmon, Jacqueline Levilliers, and Evie Melanitou for critical reading of the manuscript.

References Alloisio N, Morle´ L, Bozon M, Godet J, Verhoeven K et al. (1999) Mutation in the zonadhesin-like domain of ␣-tectorin associated with autosomal dominant non-syndromic hearing loss. Eur J Hum Genet 7, 255–258 Avraham KB (1998) Hear come more genes! Nat Med 4, 1238–1239 Church GM, Gilbert W (1984) Genomic sequencing. Proc Natl Acad Sci USA 81, 1991–1995

M.-C. Simmler et al.: twt is allelic to Otog Cohen-Salmon M, El-Amraoui A, Leibovici M, Petit C (1997) Otogelin: a glycoprotein specific to the acellular membranes of the inner ear. Proc Natl Acad Sci USA 94, 14450–14455 Cohen-Salmon M, Mattei M-G, Petit C (1999) Mapping of the otogelin gene (OTGN) to mouse chromosome 7 and human chromosome 11p14.3: a candidate for human autosomal recessive nonsyndromic deafness DFNB18. Mamm Genome 10, 520–522 Culbertson MR (1999) RNA surveillance. Unforeseen consequences for gene expression, inherited genetic disorders and cancer. Trends Genet 15, 74–80 Denoyelle F, Marlin S, Weil D, Moatti L, Chauvin P et al. (1999) Clinical features of the prevalent form of childhood deafness, DFNB1, due to a connexin26 gene defect: implications for genetic counseling. Lancet 353, 1298–1303 Erway LC, Willott JF, Archer JR, Harrison DE (1993) Genetics of agerelated hearing loss in mice: I. Inbred and F1 hybrid strains. Hear Res 65, 125–132 Hudspeth AJ (1989) How the ear’s works work. Nature 341, 397–404 Ikeda A, Zheng QY, Rosenstiel P, Maddatu T, Zuberi AR et al. (1999) Genetic modification of hearing in tubby mice: evidence for the existence of a major gene (moth1) which protects from hearing loss. Hum Mol Genet 8, 1761–1767 Johnson KR, Cook SA, Erway LC, Matthews AN, Sanford LP et al. (1999) Inner ear and kidney anomalies caused by IAP insertion in an intron of the Eya1 gene in a mouse model of BOR syndrome. Hum Mol Genet 8, 645–653 Lane PW (1981) Twister (twt). Mouse News Lett 64, 5 McGuirt WT, Prasad SD, Griffith AJ, Kunst HP, Green GE et al. (1999) Mutations in COL11A2 cause non-syndromic hearing loss (DFNA13). Nat Genet 23, 413–419 Mustapha M, Weil D, Chardenoux S, Elias S, El-Zir E et al. (1999) An ␣-tectorin gene defect causes a newly identified autosomal recessive form of sensorineural pre-lingual non-syndromic deafness, DFNB21. Hum Mol Genet 8, 409–412 Noben-Trauth K, Zheng QY, Johnson KR, Nishina PM (1997) mdfw: a deafness susceptibility locus that interacts with deaf waddler (dfw). Genomics 44, 266–272 Petit C, Levilliers J, Marlin S, Hardelin J-P (2000) Hereditary hearing loss. In The Metabolic and Molecular Bases of Inherited Disease, Scriver CR, Beaudet AL, Sly WS, Valle D, eds. Chapter 254 (Montreal, McGrawHill) in press Simmler M-C, Cohen-Salmon M, El-Amraoui A, Guillaud L, Benichou J-C et al. (2000) Targeted disruption of Otog results in deafness and severe imbalance. Nat Genet 24, 139–143 Steel KP (1995) Inherited hearing defects in mice. In Annu Rev, Genet 29, 675–701 Steel KP, Harvisty R (1996) Assessing hearing, vision and balance in mice. In What’s Wrong With My Mouse? New Interplays Between Mouse Genes and Behavior, Soc. Neurosci. short course syllabus No. 1 Washington DC, pp 26–38 Verhoeven K, Van Laer L, Kirschhofer K, Legan PK, Hughes DC et al. (1998) Mutations in the human ␣-tectorin gene cause autosomal dominant non-syndromic hearing impairment. Nature Genet 19, 60–62 Verpy E, Leibovici M, Petit C (1999) Characterization of otoconin-95, the major protein of murine otoconia, provides insights into the formation of these inner ear biominerals. Proc Natl Acad Sci USA 96, 529–534 Verpy E, Leibovici M, Zwaenepoel I, Keats B, Liu X-Z et al. (2000) A defect in harmonin, a PDZ domain-containing protein expressed in the inner ear sensory hair cells, underlies Usher syndrome type 1C. Nat Genet 26, 51–55