|Home | About | Journals | Submit | Contact Us | Français|
We have characterized a new allele of the protocadherin 15 gene (designatedPcdh15av-6J) that arose as a spontaneous, recessive mutation in the C57BL/6J inbred strain at Jackson Laboratory. Analysis revealed an inframe deletion in Pcdh15, which is predicted to result in partial deletion of cadherin domain (domain 9) in Pcdh15. Morphologic study revealed normal to moderately defective cochlear hair cell stereocilia in Pcdh15av-6J mutants at postnatal day 2 (P2). Stereocilia abnormalities were consistently present at P5 and P10. Degenerative changes including loss of inner and outer hair cells were seen at P20, with severe sensory cell loss in all cochlear turns occurring by P40. The hair cell phenotype observed in the 6J allele between P0 and P20 is the least severe phenotype yet observed in Pcdh15 alleles. However, young Pcdh15av-6J mice are unresponsive to auditory stimulation and show circling behavior indicative of vestibular dysfunction. Since these animals show severe functional deficits but have relatively mild stereocilia defects at a young age they may provide an appropriate model to test for a direct role of Pcdh15 in mechanotransduction.
The mouse Ames waltzer (av) is a recessive mutation, which causes deafness and vestibular dysfunction associated with degeneration of the inner ear neuroepithelia. The gene that harbors the av mutation is Protocadherin 15, Pcdh15 (Alagramam et al., 2001a). Mutation in PCDH15 causes Usher syndrome type 1F (Ahmed et al., 2001; Alagramam et al., 2001b) and non-syndromic deafness DFNB23 (Ahmed et al., 2003). Recently, the R245X mutation of PCDH15 was reported to account for 58% of USH1 cases in the Ashkenazi Jewish population (Ben-Yosef et al., 2003). The importance of studying the av mouse as a model for inner ear dysfunction in these patients has grown since the identification of the Pcdh15 gene in 2001 (Alagramam et al., 2001a).
Documenting the salient features of cochlear pathology in different alleles of av will help us understand (a) the function of Pcdh15 in hair cell development and (b) the cause of inner ear disorders in USH1F and DFNB23 patients. Reports on av mutants in the literature show that mutation in Pcdh15 affects hair bundle morphogenesis and polarity (Hampton et al., 2003; Pawlowski et al., 2006; Raphael et al., 2001; Washington et al., 2005) and mechanotransduction (Alagramam et al., 2005). More recently, a detailed study on the localization and function of Pcdh15 in hair cells by Senften et al. (2006) strongly supports the role of Pcdh15 in bundle morphogenesis and polarity. Specifically, the localization of Pcdh15 to the base of the stereocilia in young mice further strengthens the view that Pcdh15 plays a role in bundle polarization during early stages of development.
Little is known about the function of protocadherins. It has been suggested that Pcdh15 is involved in hair bundle cohesion via its cadherin domains. Cadherins are known to link cell membranes together (Gumbiner, 2005; Patel et al., 2003). Pcdh15 has 11 cadherin domains that are thought to adhere to similar domains of a second molecule on an adjacent stereocillium in a homophylic manner, forming lateral links between stereocilia early in development. However, they can also act as signaling molecules through homophylic or heterophylic binding at the extracellular domains, resulting in interactions with other molecules at the cell membrane or with the intracellular cytoskeleton (Gumbiner, 2005; Patel et al., 2003). Cadherin domain 1, which is affected in the Pcdh15av-2J mutation, is thought to be key for adhesion, yet the effect of the mutation in this allele varies between subtle and severe hair cell pathology, with occasional animals showing some preservation of auditory function. The affected cadherin domain in the Pcdh15av-6J allele reported here is cadherin domain 9 and animals having this mutation are consistently unresponsive to auditory stimulation by young adulthood. How do mutations in the cadherin chain produce such functional deficits? Close analysis of the more subtle Pcdh15 mutations, such as the one described here, will be important for better understanding the function(s) of this class of cadherin.
The new allele described in this report arose as a spontaneous mutation at The Jackson Laboratory (TJL) and was maintained in a C57BL/6J (B6) background. The Animal Care and Use Committee at TJL approved the care and use of the mice included in part of the investigation. A total of 200 mice of both sexes were used at TJL. The Animal Care and Use Committee at Case Western Reserve University (CWRU) approved the care and use of the mice included in the remaining part of the investigation. A total of 50 mice of both sexes were used (25 deaf-circlers; 25 controls) at CWRU.
Auditory brainstem responses (ABRs) were conducted as previously described (Zheng et al., 1999). Briefly, mice were anesthetized and their body temperature was maintained at 37–38 °C by placing them on a heating pad in a soundproof chamber during testing. Intelligent Hearing System (Miami, FL) was used to generate acoustic stimuli and ABR recording. Platinum subdermal needle electrodes were inserted at the vertex (active), ventrolaterally to the right ear (reference) and the left ear (ground). Alternating click stimuli of 50 ms duration and tone bursts with 3 ms duration (1.5 ms rise-fall time with no plateau) of 8, 16, and 32 kHz were presented to both ears of the animals through plastic tubes (a closed system). ABR threshold was obtained for each animal by reducing the stimulus intensity from 100 dB SPL in 10 dB steps and finally in 5 dB steps until the lowest intensity that could evoke a reproducible ABR pattern was detected on the computer screen.
RT-PCR was used to screen for mutations in the Pcdh15 coding sequence. RNA was isolated from affected mice and reverse transcribed to complementary DNA (cDNA). RNA isolated from brain was used for initial screening. RNA isolated from cochlear tissues was used later to confirm results of preliminary screening from brain RNA. RNA was extracted from five av6J homozygous mutants and five unaffected siblings as described previously (Washington et al., 2005). Standard PCR technique was used to amplify specific fragments of cDNA that were resolved on a standard agarose gel to verify size of the PCR product. These products were cloned followed by DNA sequence analysis. Results from the affected mice were compared to results from the unaffected siblings. Multiple clones were sequenced (n = 3) to ensure that results were not errors introduced by PCR. RT reactions were performed with ~2 μg of total RNA. The Superscript First-Strand Synthesis System for RT-PCR (Invitrogen, CA) was used to generate and amplify Pcdh15 cDNA. Primers designed to amplify 0.5 kb or 1 kb overlapping fragments of the coding region were synthesized. Conditions used for amplification were similar to those used previously to amplify Pcdh15 (Washington et al., 2005). PCR products were analyzed on 2% agarose gels for large fragment products (~1 kb) or 3% agarose gels for smaller fragments (~0.5 kb). The following PCR conditions were used: 94 °C for 2 min followed by 34 cycles of 94 °C for 30 s, 55 °C for 30 s and 72 °C for 1 min (or 72 °C for 30 s with 0.5 kb fragment). Temperature for annealing was adjusted according to Tm of the set of primers used. RT-PCR product was cloned into pCR2.1-TOPO vector (Invitrogen, CA) prior to sequence analysis. The sequences of RT-PCR products were determined using BigDye Terminator Cycle sequencing reagents and protocols (Applied Biosystems, CA). The ABI Prism 377 DNA sequencer (Applied Biosystem, CA) was used to analyze and display the resulting sequence data. The primer pairs used to confirm abnormal splicing as a result of mutation and the PCR conditions used are described here: KA100 (5′TTT TTG CAC TGC ATC CAT TC 3′) and KA113 (5′GTG GGA TCT CTC CAG GAT GT 3′) were used to amplify a 510 bp fragment spanning the region from exon 20 to exon 23. KA684 (5′ACA TGA ATG ACT ACC CTC CA 3′) and KA685 (5′CTG CTG GAC ATC ACA GGT 3′) were used to amplify a smaller 304 bp fragment spanning the region from exon 21 to exon 23. cDNA obtained from either cochlear or brain tissues was amplified under standard RTPCR reaction conditions using Platinum Taq polymerase (Invitrogen, CA) for 35 cycles of 94 °C for 30 s, 55 °C for 30 s and 72 °C for 12 s. All PCR products were resolved on a 3% low range ultra agarose gel (BIO_RAD) and stained with 5% EtBr.
Genomic DNA isolated from tail biopsies from four mutants and two heterozygous mice were screened for fine mapping of a putative deletion site flanking exon 22 of Pcdh15. Genomic DNA (5 μg/ml) was amplified under standard reaction conditions using Platinum Taq DNA polymerase (Invitrogen, CA) for 35 cycles of 94 °C for 30 s, 55 °C for 30 s, 72 °C for 12 s; annealing temperatures and extension times were adjusted according to the suggested Tm of primer sets and the size of expected products.
The primers used to screen for mutation at the 3′ end of exon 21 were KA684F (5′ACA TGA ATG ACT ACC CTC CA 3′) and KA761R (5′GAA ACC CTA CAG GAC AGC AA 3′). These amplified a 2157 bp sequence. The primers used to screen the 5′ end of exon 22 were KA774F (5′AAC CCT GGC TTT GTC ATT TG 3′) and KA775R (5′GAA ACA CAC ACT TTG GCT GA 3′), which amplified a 2179 bp fragment 4154 bp upstream of exon 22. KA754F (5′TCA GCC AAA GTG TGT GTT T3′) and KA753F (5′ACA TGA CAG AAA GCA TGT GA 3′) with KA683R (5′TAA CCT TAG GCA GAA TGC TC 3′); KA754/KA683 amplified a 2201 bp sequence and KA753/KA683 amplified a 1710 bp sequence. The primers used to screen the 3′ end of exon 22 were KA755F (5′GAC GTG CAG TTT CCA TAC C 3′) and KA756F (5′GTA GTA ACC CGC GTC AAT C 3′) with KA757R (5′CAG CGG AAT GCC ATA TAG AC 3′). KA755/KA757 amplified a 431 bp sequence and KA756/KA757 a 374 bp sequence.
For morphological studies on the sequence of postnatal degenerative changes occurring in the inner ears of av6J homozygotes, mice were examined at each of five time points (2, 5, 10, 20, and 40 days of age). Five homozygous mice at each time point plus at least two heterozygous controls per time point were processed for histological analysis. In all cases, inner ear tissues were fixed by perilymphatic perfusion of phosphate-buffered 2.5% glutaraldehyde and immersed in fixative for 24 h at 4 °C and rinsed in 0.1 M sodium phosphate buffer, pH 7.4. For light microscopic study, the temporal bones were decalcified in 0.35 M EDTA and subsequently embedded in glycol methacrylate. Tissue sections were then cut at a thickness of 2–5 μm and stained with toluidine blue.
Methods used for transmission electron microscopy (TEM) have been described elsewhere (Pawlowski et al., 2006). Briefly, inner ears from two P10 homozygous and two age-matched heterozygous av6J mice were decalcified in 0.35 M EDTA, rinsed in buffer, dehydrated in a graded series of ethanols, and embedded in Spurrs’ resin. The basal turn of the organ of Corti was thin sectioned either in a plane parallel to the surface of the reticular lamina (cross section) or in a plane perpendicular to that surface and radial to the modiolus (longitudinal section). Sections were then stained with lead citrate and uranyl acetate and viewed using a JEOL 1200EX transmission electron microscope.
In addition, three homozygous av6J mice together with appropriate control specimens (heterozygous littermates) were studied by scanning electron microscopy (SEM) at each of four time points (P2, P5, P10, and P20). Following glutaraldehyde fixation the specimens were stained with 1% osmium tetroxide and the organs of Corti exposed by cochlear microdissection. They were then taken stepwise to 100% ethyl alcohol and critical-point dried using liquid carbon dioxide. The dried preparations were mounted on aluminum stubs and sputter-coated with gold–palladium before SEM study.
Mutant offspring are identifiable at approximately 2 weeks of age by their reduced size and unstable gait. At P25, mutant mice fail to show a startle response and display distinct head tossing and bidirectional circling behavior indicating lack of balance function. The circling behavior was observed as early as P10–14 and persists throughout life. The mutation was proven to be recessive by outcross to unrelated B6 mice with a production of 21/21 normal phenotype mice. F2 progeny (n = 102) were generated by intercrossing with CAST/Ei mice. The homozygous mutant F2 progeny (n = 20) were used for genetic mapping with a pooled DNA strategy (Taylor et al., 1994). This strategy was used to efficiently localize the mutations to Chr 10, and then individual DNAs from the 102 F2 progeny (including 32 affected mice) were typed to refine the map position with D10Mit31, D10Mit91, D10Mit42, D10Mit264, and D10Mit178. Gene order, determined by minimizing the number of obligate crossover events, and recombination frequency estimates were calculated. The mutation was mapped between markers D10Mit31 and D10Mit42 with non-recombinant to D10Mit91. The genetic distance of mutation was calculated 3.9 cM to D10Mit31, 8.9 cM to D10Mit42, and 0 cM to D10Mit91.
The genetic mapping results of non-recombination to D10Mit91, suggested this mutation might be allelic to av. To determine if the new mutation is allelic to av, mice homozygous for the av3J allele were mated to mice heterozygous for the new allele. Out of the 11 offspring, five displayed circling behavior by P20. Further heterozygous mating (+/av3J × +/new allele) produced eight mice, three of which displayed circling behavior. All offspring were also tested for startle response as a preliminary screen for hearing impairment. Offspring that displayed circling behavior failed the startle response test. The new allele did not complement mutation in Pcdh15. This linkage data, together with the complementation data, confirmed allelism of the new mutation with av.
The mice that showed circling behavior also failed to produce a startle response, which is typical of waltzing deaf mutants. Four circling mutants at P21, a stage of mature hearing sensitivity, were tested for the ability to generate the ABR (Fig. 1). Even at the highest intensities the mice demonstrating circling behavior showed no auditory brainstem response. Unaffected littermate controls showed normal waveforms and thresholds at all frequencies tested (Fig. 1).
The coding sequence of Pcdh15 was screened by RT-PCR using several sets of primers that covered the 33 coding exons of Pcdh15. RT-PCR primers designed to amplify exon 20 through 23 produced a smaller PCR product than expected (Fig. 2). Sequencing analysis confirmed that the coding sequence of exon 21 was spliced into exon 23 and the reading frame was maintained. The entire coding sequence of exon 22 was absent in mice homozygous for the av6J allele (Fig. 3). As a result of the inframe deletion, Pcdh15 protein expressed in av6J allele would affect the 9th extracellular cadherin domain due to the loss of 141 bp and subsequent loss of 47 amino acids (Fig. 3).
In +/6J mice, an additional band was observed (labeled ‘*’in Fig. 2B). The additional band was not detected in the wild-type (wt) or 6J/6J mice. Further, +/6J mice do not show a phenotype. These facts suggest that this band might be an artifact that can sometimes result from heteroduplex formation between closely related sequences following PCR. To verify this possibility, a ‘mixing’ experiment was conducted in which equal amounts of first strand cDNA from wt and 6J/6J mice were mixed and amplified with the primer sets flanking exon 22. The additional product was amplified as efficiently with this mixed template as was found in RNA from +/6J (data not shown), thus demonstrating that the additional product observed is likely to be an artifact from heteroduplex formation between the wt and 6J/6J transcript.
These findings lead us to believe that the mutation in av6J affects normal splicing of exon 22. Therefore, we used a series of PCR primer pairs to fine map the deletion of the av6J genomic sequence between exons 21 and 22 (Fig. 4). DNA isolated from tail biopsies of circling and non-circling siblings was used. These reactions were repeated using DNA isolated from independent animals (five circlers; five non-cirlers). Primer pairs 754/683, 753/683, and 755/757 failed to amplify a product from circling mice but amplified products of expected size from non-circler controls. Primer pairs 774/775 and 756/757 did amplify both mutants and controls. These results show that the 5′-end of the deletion lies between primers 754 and 753 and the 3′-end of the deletion lies between primers 755 and 756. In addition, primer 754 is complementary to 775. Mismatch at the 3′-end of a primer is likely to prevent amplification, however mismatch at 5-end does not have the same effect. Successful amplification with primer 775 (774/775) but not with 754 (754/683) suggests that primer 754 is located at the 5′-break point. The 3′ break point is between primer 755 and 756. This series of analyses shows that a ~2 kb of the intron upstream of exon 22 and at least 36 bp from exon 22 has been deleted in the av6J allele (Fig. 4A). This deletion includes the 5′ splice acceptor sequence of exon 22 and explains the skipping of exon 22 in the av6J transcript. The mutation in av6J is different from other mutations in Pcdh15 identified thus far (Fig. 5).
Morphologic abnormalities of outer hair cell stereocilia were observed by SEM in the youngest animals included in this study (2 days postnatal, P2). However, considerable variation in the severity of pathology between individual animals was seen at this age. Fig. 6 shows representative basal-turn specimens from a control animal and three different homozygous mutants at P2. One of the mutants had normal-appearing stereocilia. In the second animal a few, scattered outer hair cells with alterations in the arrangement of their stereocilia were observed and in the third mutant virtually all the OHC hair bundles on basal-turn OHC were clearly abnormal. Even in the most severely affected animal, there were no obvious defects of inner hair cells. Except for the variation in severity between individual animals seen in av6J mutants at P2, the abnormalities of outer hair cell stereocilia bundles observed here resemble those seen in the inframe deletion mutation Pcdh15av-J and Pcdh15av-2J mutants at P2 (Pawlowski et al., 2006).
Stereocilia defects were clearly present on outer hair cells from all Pcdh15av-6J animals examined at P5. At that time point, many OHCs throughout the cochlea had hair bundles in which the ‘V’ or ‘W’ pattern characteristic of normal outer hair cells was markedly distorted in shape as shown in Fig. 7. Occasional cells were also found in which the stereocilia bundles were not only abnormal in shape, but also rotated out of normal position on the apex of the cell (Fig. 7B). In addition to the outer hair cell defects, abnormalities of inner hair cell stereocilia were consistently seen in Pcdh15av-6J mutants examined at P5 and those defects were particularly obvious in the basal cochlear turn. Compared to control specimens, the inner hair cell stereocilia bundles showed an irregular configuration, with stereocilia arranged in a disorderly fashion in poorly defined rows (Fig. 7D).
As illustrated in Fig. 8, stereocilia defects were also observed in all mutants examined at P10. At that age, hair bundles on both inner and outer hair cells from all turns showed irregularities in configuration. On outer hair cells, loss of stereocilia within individual bundles was observed and this often involved loss of stereocilia near the center of the hair bundle as can be seen in Fig. 8B–D. The cuticular plates of affected cells also tended to show various surface irregularities, including localized swelling in the areas where stereocilia were missing. Observations by TEM showed defects similar to those previously described in Pcdh15avJ alleles of this age, including irregular patterns of insertion of stereocilia rootlets into the outer hair cell cuticular plates (Pawlowski et al., 2006). An additional defect was detected in longitudinal sections in which the stereocilia rootlets were found not to run in parallel through the cuticular plate in several cells (Fig. 9). This defect was not observed in TEM cross sections from other Pcdh15 alleles at P10 and may be related to disruption of the cuticular plate near the center of the stereocilia bundle seen by SEM. This is a subtle defect in bundle structure that was rarely seen by SEM in the other alleles. No obvious defects in inner hair cells were seen in either cross sectional or longitudinal sections (data not shown). As was true of the P2 and P5 animals (as well as other Pcdh15 alleles), cochlear cross sections from P10 av6J homozygotes examined by light microscopy showed normal organ of Corti development without obvious alterations in the bodies of sensory or supporting cells and with normally developed fluid spaces within the cochlear neuroepithelium (data not shown).
Degenerative changes in the organ of Corti were, however, clearly apparent in P20 specimens. At that time point, loss of both inner and outer hair cells was observed throughout the cochlea as shown in Fig. 10. There were also scattered focal lesions where supporting cells as well as sensory cells were lost, leading to collapse of the neuroepithelium in sharply localized areas as illustrated in Fig. 10D.
Light microscopic study of cross sections demonstrated that by P40 there was complete loss of outer hair cells together with severe inner hair cell loss affecting all cochlear turns in av6J mutant animals. In addition, there were patchy areas scattered throughout the length of the cochlea in which the organ of Corti was completely collapsed due to degeneration of all sensory and supporting cells. As shown in Fig. 11, there were also reduced numbers of spiral ganglion cells in the most severely affected areas. Although there was severe neuroepithelial degeneration at P40, the cochlear duct retained its normal configuration in all parts of the cochlea. There was no evidence of collapse of Reissner’s membrane or obvious abnormalities of the stria vascularis.
Pcdh15av-6J adds to the growing series of Pcdh15 alleles (Fig. 5), some of which carry mutations similar to those reported in humans (Ahmed et al., 2001). Characterizing these alleles has helped us understand which cell types in the mouse inner ear require Pcdh15 and whether Pcdh15 is required for normal hair cell development. Findings from morphological studies and physiological testing have raised some important questions (as mentioned previously in the introduction) regarding the role for Pcdh15 in hair cell function and maintenance. Identification of new alleles of Pcdh15 may help to answer some of these questions.
The av6J allele is maintained in the B6 inbred genetic background. It is well known that the B6 mouse develops premature presbycusis due to the recessive Ahl (age-related hearing loss) allele (Erway et al., 1993; Keithley et al., 2004). B6 mice develop normal hearing function and show no inner ear pathology until they are several months old; mice homozygous for the av6J mutation fail to develop hearing function and show hair cell pathology several weeks after birth. Further, a previous report showed that mutation in Pcdh15 causes a similar ear phenotype in the FVB/N inbred background (Alagramam et al., 1999). Therefore, the early hearing loss and associated hair cell pathology in the homozygotes are due to the recessive, loss-of-function Pcdh15 mutation and not due to the B6 background.
This study demonstrates that in the av6J mutant postnatal degeneration of the organ of Corti occurs over a time course similar to that observed in other av alleles, starting with alterations in outer hair cell stereocilia a few days after birth and progressing to severe neuroepithelial degeneration affecting all cochlear turns by P40. Alterations in configuration of outer hair cell stereocilia were observed at the earliest time point included in this study (2 days postnatal). At least up to P10, abnormalities of the organ of Corti appear to be limited largely to the stereocilia and cuticular plates of hair cells. Light microscopic examination of cochlear cross sections from animals aged P2 through P10 showed normal development of the sensory epithelium without apparent defects of the cell bodies of hair cells or supporting cells. This observation is consistent with the pattern of degeneration seen in other av alleles in which alterations of OHC stereocilia are the first observable abnormalities occurring in the organ of Corti. Although stereocilia defects are a prominent feature during the initial stages of neuroepithelial degeneration in the av6J allele, alterations in the arrangement of stereocilia do not appear to be as severe as those previously observed in av alleles with presumptive null mutations (Alagramam et al., 1999, 2001a; Hampton et al., 2003; Raphael et al., 2001). Loss of outer hair cell stereocilia, most often seen in the central portion of individual hair bundles and first observed at P10, is likely to be indicative of early-stage degenerative changes in the sensory cells. By P20 those changes advanced to degeneration and loss of both inner and outer hair cells, which occurred in a scattered, patchy fashion throughout the organ of Corti. In addition to sensory cell loss, scattered, focal lesions were seen at P20 in which there was also degeneration of supporting cells, leading to collapse of the entire sensory epithelium in the affected areas. At P40 there was more widespread loss of sensory and supporting cells, together with a reduction in the spiral ganglion cell population in the most severely affected areas. Although there were severe degenerative changes at P40, the overall configuration of the cochlear duct remained normal. Reissner’s membrane was in normal position and no abnormalities were apparent in the stria vascularis. This finding is consistent with all other av alleles examined to date and is characteristic of cochlear degeneration of the neuroepithelial type.
The av6J allele, like the avJ and av2J alleles described earlier (Pawlowski et al., 2006), is an inframe deletion mutation affecting the cadherin domains of the Pcdh15 molecule. All three inframe deletion mutants show hair cell phenotypes that are mild compared to presumptive null mutants av3J or avTg at a young age (P0–P10). Slight differences in phenotype can be detected between the three inframe deletions; avJ mutants tend to have more abnormal cells which demonstrate rotated or clumped stereocilia bundles and only a few abnormal cells that have missing stereocilia at the center of the bundle, whereas the latter anomaly is more commonly seen in av6J’s. The av2J phenotype varies from a phenotype similar to the avJ and av6J mutants to one similar to that of the functional nulls, where most of the hair cells have abnormal stereocilia bundles (Alagramam et al., 1999, 2001a; Hampton et al., 2003; Raphael et al., 2001). Therefore, the mutations affecting various cadherin domains have similar but slightly different effects on the development and integrity of the stereocilia bundle.
The av6J allele harbors an intragenic deletion that disrupts the expression of exon 22. The genotype–phenotype correlation was 100% since all mice with auditory and vestibular deficiency failed to express exon 22. Splicing of exon 21 into 23 is predicted to result in the deletion of most of the amino acid coding for the nineth cadherin domain, but the rest of the Pcdh15 protein is predicted to be inframe. The hair cell defect observed in av6J during the first two weeks after birth is less severe compared to hair cells from presumptive null alleles at the same time point (ex. Pcdh15av-3J or Pcdh15av-nmf19). The hair cell phenotype observed in av6J is similar to, but slightly less severe than, those observed in the Pcdh15av-J allele, which is consistent with the nature of the mutations present in the Pcdh15av-J and Pcdh15av-6J alleles. Mice homozygous for the 6J mutation fail to develop normal hearing or balance function, similar to that observed for Pcdh15alleles reported previously. These results suggest that the cadherin domains of Pcdh15 are not dispensable and that even small changes in the Pcdh15 protein could lead to hair cell dysfunction.
Six mouse mutants carrying various mutations in Pcdh15 have been reported thus far, all demonstrating loss of auditory function early in life. The absence of hair cell function, as measured by lack of transduction currents and dye-uptake studies is thought to be a consequence of defects in stereocilia bundle morphogenesis in Pcdh15-deficient mice (Alagramam et al., 2005; Senften et al., 2006). However, since hair bundle morphogenesis is dramatically affected in most Pcdh15 mutants, none of the studies discussed above excludes the possibility of a direct role for Pcdh15 in mechanotransduction. Pcdh15 mutants demonstating hearing and balance dysfunction but with stereocilia relatively intact at a young age may provide useful models to test for a direct role for Pcdh15 in mechanotransduction.
This work was supported by NIDCD Grant DC007392 to QYZ and DC05385 to KA. We thank the Mouse Mutant Resource at The Jackson Laboratory, Bar Harbor, ME, for their mouse colony management and the initial discovery of the mouse mutant.