Human Molecular Genetics Advance Access originally published online on December 8, 2004
Human Molecular Genetics 2005 14(3):401-410; doi:10.1093/hmg/ddi036
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Human Molecular Genetics, Vol. 14, No. 3 © Oxford University Press 2005; all rights reserved
Myosin XVa and whirlin, two deafness gene products required for hair bundle growth, are located at the stereocilia tips and interact directly
1Unité de Génétique des Déficits Sensoriels, INSERM U587, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris, cedex 15, France, 2School of Biological Sciences, The University of Sussex, Falmer, Brighton BN1 9QG, UK, 3Laboratory for Neural Architecture, Brain Science Institute, RIKEN, Wako, Saitoma, Japan and 4Hybrigenics, 3–5 impasse Reille, 75014 Paris, France
* To whom correspondence should be addressed. Email: cpetit{at}pasteur.fr
Received September 27, 2004; Accepted December 1, 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTS AND DISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Defects in myosin XVa and the PDZ domain-containing protein, whirlin, underlie deafness in humans and mice. Hair bundles of mutant mice defective for either protein have abnormally short stereocilia. Here, we show that whirlin, like myosin XVa, is present at the very tip of each stereocilium in the developing and mature hair bundles of the cochlear and vestibular system. We found that myosin XVa SH3-MyTH4 region binds to the short isoform of whirlin (PR-PDZ3) that can rescue the stereocilia growth defect in whirlin defective mice. Moreover, the C-terminal MyTH4-FERM region of myosin XVa binds to the PDZ1 and PDZ2 domains of the long whirlin isoform. We conclude that a direct myosin XVa–whirlin interaction at the stereocilia tip is likely to control the elongation of stereocilia. Whirlin, unlike myosin XVa, is also transiently localized in the basal region of developing stereocilia in rat vestibular and cochlear hair cells until P4 and P12, respectively. Notably, whirlin also interacts with myosin VIIa that is present along the entire length of the stereocilia. Finally, we show that the transmembrane netrin-G1 ligand (NGL-1) binds to the PDZ1 and PDZ2 domains of whirlin and has an extracellular region that homophilically self-interacts in a Ca2+-dependent manner. The interaction between whirlin and NGL-1 might be involved in the stabilization of interstereociliar links.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTS AND DISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In the sensory organs of the inner ear, mechano-electrical transduction occurs in the hair bundle, a structure located at the apex of the sensory hair cells. The hair bundle consists of an array of 20–300 stiff microvilli known as stereocilia, each containing a rigid core of up to 1000 cross-linked actin filaments. Stereocilia are organized in rows of increasing height forming a characteristic staircase pattern and are held together by several sets of extracellular links that project laterally from the surface of each stereocilium in a symmetrical manner (1
). Deflection of the hair bundle toward the tallest stereocilia increases the opening probability of cation-selective mechano-electrical transduction channels, resulting in membrane depolarization. The tip link, a filament that extends from the tip of each stereocilium to the side of its adjacent, taller neighbour, is likely to play a crucial role in the channel gating process (2).
The development of the hair bundle is a complex process (reviewed in 3), involving molecular mechanisms that are still unknown. In particular, the mechanisms that control hair-bundle height, a parameter that varies continuously from base to apex of the cochlea, and the length of individual stereocilium as a function of its position within the hair bundle are not understood. Two recessive deaf mouse mutants with abnormally short stereocilia, shaker-2 (sh2) and whirler (wi), may help decipher the molecular mechanisms of stereociliary growth (4,5). In sh2 mice, the stereocilia are extremely short but are arranged in a nearly normal pattern (4). The causative gene, Myo15a, encodes myosin XVa, an unconventional myosin that localizes at the very tip of the differentiating stereocilia (6). In wi mice at postnatal day 15 (P15), the stereocilia of cochlear inner hair cells (IHCs), the primary sensory cells of the cochlea, are approximately half the normal length, whereas the stereocilia of the outer hair cells (OHCs) are almost normal in length, even though they are arranged in a rounded U shape instead of the usual V or W shape (5). Both IHCs and OHCs eventually degenerate in wi mutants. Morphological analysis of the IHCs in wi embryos has revealed that stereocilia are already significantly shorter than those in controls by embryonic day 18.5 (E18.5). We have identified the defective gene in these mutant mice as Whrn, a gene that encodes whirlin, PDZ domain-containing protein (7). Long and short transcripts resulting from alternative transcription start sites could be detected in the inner ear predicting two isoforms of whirlin. The long form contains a proline-rich (PR) domain and three PDZ domains, whereas the short C-terminal form, hereafter referred to as short whirlin, contains only the PR and the third PDZ domain. Wi mutant mice carry an intragenic deletion that creates a frameshift at codon 433, resulting in premature termination of the protein before the PR domain. In a BAC rescue experiment, the short whirlin isoform was able to correct both the height of IHC hair bundles and the degeneration of hair cells (7).
As both Myo15a and Whrn mutations produce stereociliary growth defects in the mouse, we studied the localization of whirlin in developing and mature hair cells. The results led us to investigate and characterize interactions between myosin XVa and whirlin. Because many PDZ domain-containing proteins have been shown to act as scaffolding proteins for the assembly of large protein complexes at specific sub-membranous locations (8), and to gain further insight into the role of whirlin, we also used yeast two-hybrids to search for additional whirlin-binding proteins. We thereby identified the netrin-G1 ligand (NGL-1) as another binding partner for whirlin.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTS AND DISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Whirlin and myosin XVa are co-located at the tip of the growing stereocilium
We used the previously described monoclonal anti-CIP98/whirlin antibody (9
) (Materials and Methods) for immunofluorescence analyses. To define the whirlin region recognized by the antibody, we used it to stain COS7 cells expressing different GFP-tagged whirlin constructs. We observed that the antibody recognizes both the long and the short whirlin isoforms by immunofluorescence. It recognizes the PR domain, but not the PDZ3 domain, when these are expressed as isolated GFP-tagged domains (data not shown). We studied whirlin expression in whole-mount preparations of rat cochlea and vestibule from E18 to P40, including a day-by-day analysis from P1 to P12. Actin filaments were labelled with phalloidin. Detailed confocal microscopy analysis was performed on all the preparations. At E18, strong labelling of the emerging hair bundle was observed in the vestibule (Fig. 1A), but no staining of IHCs and OHCs could be detected. At E20, whirlin was detected in IHC hair bundles. From P1 onwards, we observed whirlin at the very tips of the stereocilia in both IHCs and OHCs, in an area free of detectable actin (Fig. 1B). The intensity of the apical staining in the IHC hair bundle increased up to P10–P12. This apical staining was more intense in IHCs than in OHCs up to P12, whereas the reverse was observed in mature hair cells (Fig. 1B, P35). This staining was mainly detected at the tip of the highest row of stereocilia both in IHCs and in OHCs. Within the IHC hair bundles, this apical staining was barely detectable at the tip of the second row of stereocilia. Within the OHC hair bundles, the staining of the second row of stereocilia increased at P6–P8 and thereafter became weaker than that in the highest row (Fig. 1B). This observation suggests that the elongation of stereocilia involves the recruitment of whirlin to their tips. Additional strong labelling was observed in IHCs. Extensive confocal analysis led us to conclude that this staining is associated with the basal region of the hair bundle. The intensity of this basal staining peaked at P6 and became undetectable from P12 onwards (Fig. 1B). We observed the same apical and basal whirlin labelling of the stereocilia in whole-mount preparations of the mouse cochlea. As shown in Figure 1C, myosin XVa co-localizes with the apical but not the basal whirlin labelling. Neither the apical nor the basal whirlin labelling of stereocilia was detected in wi mutants (Fig. 1B). Whirlin labelling was not observed in other subcellular locations within the hair cells, including the synaptic region.
|
In the rat vestibular end organs, whirlin was also present both at the very tip and at the basal region of the stereocilia as early as E18 (Fig. 1A). The basal whirlin labelling was very strong but began to fade out at P3 and was no longer observed after P4. The apical labelling was detected at the tips of stereocilia of various lengths, with an intensity that seems to be roughly proportional to the length of the stereocilia in the adults (with the longer being brighter), which is reminiscent of what has been reported for myosin XVa (6
). Similar results were obtained in the mouse vestibular organs (data not shown).
Myosin XVa binds to whirlin
Because myosin XVa and whirlin are co-located at the tip of each growing stereocilium, we explored whether the two proteins could physically interact. As a first step, we examined the subcellular distribution of murine myosin XVa and both the long and short isoforms of whirlin in transfected COS7 cells. Two myosin XVa isoforms are predicted, which differ by the presence/absence of an N-terminal domain (10). Their tails contain two repeats, each composed of a MyTH4 (myosin tail homology 4) and a FERM (4.1, ezrin, radixin, moesin) domain, separated by a poorly conserved SH3 domain. We reconstructed the entire myosin XVa isoform devoid of the N-terminal domain and introduced a cMyc N-tag. In transfected cells producing myosin XVa, the protein was detected at the plasma membrane and at the tips of filopodia, as previously reported (11). In transfected cells producing either short or long whirlin isoform, the protein was detected at the plasma membrane but not in filopodia, even though some labelled filopodia were occasionally observed in cells producing long whirlin. Because some PDZ domains have been shown to bind to PIP2 (12), a possible interaction between whirlin and PIP2 may account for the location of whirlin at the plasma membrane. In co-transfected cells producing myosin XVa and either short or long whirlin, the proteins were co-localized at both the plasma membrane and the tips of filopodia (Fig. 2).
|
Because short whirlin is able to rescue the stereocilia growth defect in wi mutants, we used biochemical tests to explore a possible interaction between the myosin XVa tail and this whirlin isoform. In pull-down experiments, a soluble fraction from HEK293 cells transfected with the cMyc-tagged myosin XVa tail was passed over GST–short whirlin resin. As shown in Figure 3B (left panel), the myosin XVa tail binds to GST–short whirlin but not to GST alone. To test the possibility of a direct interaction between short whirlin and myosin XVa tail, we carried out in vitro binding assays. Because SH3 and PR domains are known to preferentially interact (reviewed in 13
), and as whirlin possesses two consensus PR sequences, a class I 581-KPPPPPP-587 (+-X-X-P-X-X-P) and a class II 600-PRKPGR-605 (P-X-X-P-X-+) sequence, we tested the possibility of a direct interaction between myosin XVa SH3-MyTH4 fragment and short whirlin. Indeed, we found that this in vitro translated myosin XVa fragment binds to a GST–short whirlin column (Fig. 3B, right panel).
|
The interaction between the SH3-MyTH4 domain of myosin XVa and short whirlin is likely to account for the rescue of the wi phenotype by this whirlin isoform. Although the binding was rather weak in vitro, it is in agreement with the known low affinity of SH3 domains for their ligands (13
). The rather low selectivity of these interactions has also been pointed out (13). However, selective interaction could be achieved in vivo by the compartmentalization of potential binding partners. Moreover, the low affinity of the interaction should be considered in the context of the stereocilium, a compartment with a small volume (
1 fl) filled with a very dense core of actin filaments that hampers protein diffusion.
As the long whirlin isoform differs from the short whirlin only by two additional N-terminal PDZ domains (PDZ1–PDZ2), it was expected to also interact with the myosin XVa tail. However, we were not able to show an interaction between the myosin XVa tail and the entire long whirlin isoform, either by pull-down experiments (Fig. 3C, left panel) or by immunoprecipitation (data not shown) of extracts from co-transfected HEK293 cells. Nevertheless, we pursued the search for a possible direct interaction as the myosin XVa sequence has a C-terminal, class I consensus PDZ-binding motif, ITLL* (X-S/T-X-
-*). For this purpose, we tested whether PDZ1 and PDZ2 whirlin domains are able to bind to the C-terminal MyTH4-FERM tail fragment of myosin XVa. Both in vitro translated PDZ1-PDZ2 (data not shown) and isolated PDZ1 and PDZ2 domains bind to GST-tagged C-terminal MyTH4-FERM tail fragment of myosin XVa but not to GST alone (Fig. 3C, right panel). Under the same experimental conditions, binding of short whirlin to this C-terminal myosin XVa fragment could not be observed. As this myosin XVa sequence has a C-terminal, class I consensus PDZ-binding motif, we examined whether this motif is required for the interaction with long whirlin. We found that the GST–myosin XVa (MyTH4-FERM) fragment lacking the ITLL motif still binds to whirlin. In contrast, whirlin was no longer able to bind to this myosin XVa fragment after its FERM domain and consensus PDZ-binding motif had been deleted (Fig. 3C, right panel). Therefore, we identified at least two whirlin regions, namely, the PR region and the PDZ1–PDZ2 region, which are able to interact with the myosin XVa tail. Whether the PDZ3 domain is also involved in the whirlin–myosin XVa interaction remains to be determined. Because we could not show an interaction between myosin XVa and whirlin long isoform produced as a whole protein either in bacterial or in eukaryotic cell systems, we suggest that in both cases, the protein was folded in a conformational state that masks the potential interaction sites with myosin XVa. It is therefore possible that in the hair cells, another interacting partner for whirlin induces a conformational change of whirlin, which in turn allows binding to myosin XVa.
Myosin XVa may convey and/or maintain whirlin at the tips of stereocilia (14). The accumulation of myosin XVa at the very tip of the stereocilium, however, suggests that this myosin locally exerts a tension force between the plasma membrane and the actin cytoskeleton. In addition, the long myosin XVa isoform is characterized by an additional large (about 1200 amino acid residues) N-terminal domain that precedes the motor head (10). On the basis of sequence similarities with elastomeric domains (15,16), it is tempting to speculate that this domain has elastic properties. Myosin XVa thus appears as a good candidate to sense the tension between the plasma membrane and the actin filaments, a role that is likely to be critical in the stereocilia growing process. The presence of short stereocilia at the apical pole of inner-ear hair cells in wi and sh2 mice suggests that in growing stereocilia, neither myosin XVa nor whirlin is required for the initial step of actin nucleation. Rather, the proteins are likely to play a role in the elongation of unbranched actin filaments. At the stereocilia tip, the interaction between whirlin and myosin XVa may be involved in two mechanisms that are not mutually exclusive. Firstly, the whirlin–myosin XVa interaction may bridge different components involved in actin polymerization at the stereocilia tips. Secondly, through its motor activity, myosin XVa may drive the full organization of the apical membrane domain of the stereocilia, which is initiated by whirlin. In support of the latter proposal, it is noteworthy that the tip links are not present in sh2 mutants (17).
Myosin VIIa binds to whirlin
Because we observed that whirlin is also transiently present in the basal region of stereocilia, i.e. in a region where myosin XVa has not been detected, we examined whether whirlin also interacts with myosin VIIa. This unconventional myosin, structurally closely related to myosin XVa, is present along the entire length of the hair cell stereocilium (18,20). Indeed, we found that myosin VIIa co-localizes with whirlin at the plasma membrane of co-transfected COS7 cells (Fig. 2), and the myosin VIIa tail binds to whirlin in a pull-down experiment (Fig. 3D). Phenotypic analysis of sh1 mouse mutants has revealed that myosin VIIa is not required for stereocilia growth but plays a key role in the cohesion of the growing hair bundle (19). Therefore, it is possible that the whirlin–myosin VIIa interaction that we show here accounts for the abnormal organization of the OHCs stereocilia in the wi mutants.
Whirlin can form oligomers
As the two whirlin isoforms possess a C-terminal class II consensus PDZ-binding motif, NVML* (X--X--*), we examined the possibility that the two proteins can form homodimers. As shown in Figure 3C, in vitro translated whirlin binds to GST–whirlin but not to GST alone. The same experiments performed with short whirlin showed that it can also form homomers (Fig. 3E).
Whirlin binds to NGL-1, a transmembrane protein containing LRR and Ig-like domains
We searched for other whirlin-binding proteins using the yeast two-hybrid technique. The whirlin PDZ1–PDZ2 fragment was used as the bait to screen a P2–P6 inner-ear cDNA library (20). One clone predicted to encode a 60 amino acid peptide was identified. This prey corresponds to the 3' translated end of a cDNA coding for the NGL-1, a predicted 640 amino acid (72 kDa) transmembrane protein expressed in the embryonic and adult brain (21). The deduced amino acid sequence contains a hydrophobic, 45 amino acid N-terminal region characteristic of a signal peptide, a single transmembrane domain (amino acids 527–548) and an intracytoplasmic domain of 92 amino acids that contains a C-terminal, class I PDZ-binding consensus motif (ETQI*). The extracellular part consists of nine contiguous leucine-rich repeats (LRRs) flanked by cysteine-rich, leucine-rich repeats (LRRNT and LRRCT) and a membrane proximal C2-type Ig-like domain (Fig. 5). In transfected MDCK cells, NGL-1 appears to be membrane-associated, whereas whirlin has a diffuse cytoplasmic distribution. However, in co-transfected MDCK cells expressing both proteins, whirlin distribution was similar to that of NGL-1 (Fig. 4A). To further test a possible interaction between whirlin and GST–NGL1, we performed in vitro binding assays. Whirlin was observed to bind to NGL-1. The isolated PDZ1 and PDZ2 domains of whirlin were both also observed to bind to GST–NGL1 (Fig. 4C). In contrast, neither short whirlin (Fig. 4C) nor either harmonin a or harmonin b (data not shown), isoforms of another PDZ domain-containing protein present in the growing hair bundle (22), binds to NGL-1.
|
|
As another LRR domain-containing protein, AMIGO, forms homomeric and heteromeric structures (22
), we tested the possibility that NGL-1 can interact homophilically. Indeed, homophilic interaction was detected when the NGL-1 protein was passed through a GST–NGL1 column. Binding was not detected between NGL-1 and an NGL-1 fragment encompassing the intracellular domain alone (Fig. 4D). Finally, we tested whether the NGL-1 homophilic interaction was dependent on Ca2+ by repeating the binding assay using varying concentrations of free Ca2+(from 0 µM to 1000 µM). As shown in Figure 4D, the interaction only occurred in the 5–250 µM Ca2+ concentration range.
RT–PCR analysis of P2 mouse tissues showed that the NGL-1 transcript is more abundant in the brain, eye and inner ear, although it was also detected in the heart, lung, kidney and intestine (Fig. 4B). By single cell RT–PCR at P6, the NGL-1 mRNA could be amplified in five out of the 32 cochlear hair cells (3/22 OHCs and 2/10 IHCs) analyzed (Fig. 4B), indicating that NGL-1 is expressed at a low level in both IHCs and OHCs at this stage. We could not detect the endogeneous protein in the developing hair cells by immunohistochemistry. However, this whirlin-binding transmembrane protein is an attractive candidate component of interstereociliary links. Indeed, we showed that homophilic interaction between the NGL-1 extracellular domains may occur in the presence of Ca2+ concentrations (5–250 µM) in the range of that of endolymph (20–100 µM), the extracellular fluid that bathes the hair bundles of inner-ear sensory cells, but not of perilymph (600–700 µM), the extracellular fluid that bathes the basolateral regions of the hair cells (reviewed in 23).
Unconventional myosins and PDZ domain-containing proteins in hair-bundle differentiation
The potential interactions among NGL-1, whirlin and myosin XVa revealed by this study are shown schematically in Figure 5. Differentiation of the hair cell's apical region into a functional hair bundle involves at least two different processes, namely, the growth of the stereocilia and their cohesion. Although these processes are far from being elucidated at the molecular level, in each case, a PDZ domain-containing protein (whirlin or harmonin) and an unconventional myosin (myosin XVa or myosin VIIa) are involved. Whirlin and harmonin, two proteins containing three PDZ domains that share the highest degree of sequence similarity, directly interact with myosin XVa (this study) and myosin VIIa (20), respectively. Notably, the tails of these myosins also have the highest degree of similarity among unconventional myosins. Moreover, in both cases, the binding of a PDZ domain to a myosin MyTH4-FERM domain is involved. In addition, harmonin interacts with the stereociliary transmembrane protein cadherin 23 (20,24) and whirlin with NGL-1 (this study). Therefore, various protein complexes based on the same molecular scheme underlie different and concomitantly occurring processes during hair-bundle differentiation. The two PDZ domain-containing proteins and the two myosins likely emerged from gene duplication and divergence; the latter process may have been driven by selection pressure during the evolution of the balance/auditory organs.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTS AND DISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Expression constructs
The cDNAs encoding the entire murine myosin XVa (1–3511) and the myosin XVa tail fragment (2536–3511) were obtained by RT–PCR on inner-ear total RNA and cloned in pCMV-tag 3b for expression in COS7 cells. The cDNAs encoding truncated forms of myosin XVa, i.e. C-terminal MyTH4-FERM (2950–3511), C-terminal MyTH4-FERM lacking the C-ter class I PDZ-binding consensus sequence (ITLL), C-terminal MyTH4 alone (2950–3185) and SH3-MyTH4 (2809–3185), were cloned in pGex-4T1 for protein production. A cDNA encoding the entire murine whirlin (1–906) was obtained by RT–PCR on mouse inner-ear total RNA and cloned into pcDNA. The cDNAs encoding truncated forms of whirlin, i.e. PDZ1 (1–263), PDZ2 (257–422), PDZ1–PDZ2 (1–422) and PR-PDZ3 (443–906), were subcloned into pcDNA and pGex-4T1. In order to map the anti-CIP98/whirlin epitope, GFP-tagged whirlin constructs, namely, long whirlin (1–906), short whirlin (443–906), PR (443–802) and PDZ3 (797–906), were expressed in COS7 cells. A cDNA encoding the full human myosin VIIa (1–2215) was cloned in pcDNA for expression in COS7 cells, and a cDNA encoding the human myosin VIIa tail fragment (847–2215) was cloned in pcDNA for expression in HEK293 cells. A cDNA encoding the entire murine NGL-1 protein (amino acids 1–640) was obtained by RT–PCR on inner-ear total RNA. PCR products were subcloned into pcDNA (No tag, Invitrogen) for expression in MDCK cells and into pGex-4T1 (GST tag, Amersham) for protein production. A cDNA fragment encoding the cytoplasmic region of NGL-1 (549–640) was obtained in the same way and subcloned into pGex-4T1 for protein production.
Single cell RT–PCR
In order to obtain single hair cells, the isolated organ of Corti was collected in 50 µl of PBS and 50 µl of 10% trypsine at room temperature and was slowly mechanically disassembled during 4 min. A total of 100 µl of fetal calf serum was added for the neutralization of the trypsin. Reverse transcription (RT) of single cell mRNAs was carried out using the SuperScript II Reverse Transcriptase kit (Invitrogen) according to the manufacturer's instructions, except that the RT reaction was carried out overnight at 37°C. PCR was initially carried out with NGL-1 sense (5'-GTTGGAGCTTCCATTGACACTC-3') and antisense (5'-ATCTGGTGTTGGTCCTTCTGGA-3') primers derived from exons 3 and 7, respectively. A second, nested PCR was carried out with NGL-1 sense (5'-CTCATGAATTACCGAAGGAACGAAG-3') and antisense (5'-TGGCTCTTCTCTTCCTGAGAGA-3') primers derived from exons 4 and 6, respectively, through 35 amplification cycles (40 s at 95°C, 45 s at 57°C and 50 s at 72°C). To confirm the cell type, i.e. IHC versus OHC, parallel amplification of myosin VIIa and prestin mRNAs was carried out using the following primer sets: myosin VIIa—first PCR (5'-TCTAATCCGGCAGGTCTCAC-3' and 5'-CCAGCAGATATGGTCAGTCC-3'), nested PCR (5'-TGCCATCAACAAGTACGGGG-3' and 5'-AGCGTCTCCTCTGCGGTTC-3'); prestin—first PCR (5'-CAACGTGGCCAATGCTACTG-3' and 5'-ATTCAGGAGCGGTGCACAAC-3'), nested PCR (5'-AACTCTGGCCGGGATTGTGA-3' and 5'-CATTGGGCTCCATATCCTCC-3'). The specificity of the PCR products was confirmed by DNA sequencing.
Immunofluorescence analysis of cells and tissues
Transient transfection of cells was carried out using Effectene (Qiagen). Immunofluorescence analysis was carried out on fixed cells and whole-mount cochleae, as described (25). Anti-Myc (clone 9E10) (Santa Cruz) and anti-CIP98/whirlin (20) mouse monoclonal antibodies were used. Cells and whole-mount cochleae were analyzed with a laser scanning confocal microscope (LSM-510, Zeiss). TF1 and PB48 antibodies (a kind gift from Thomas Friedman) were used to detect myosin XVa.
Antibody production
A guinea pig immune serum to the murine NGL-1 was produced (Covalab) against an epitope located in the putative cytoplasmic region (HRQNHHAPTRTVEI, amino acids 555–568, GenBank accession no. NP848840). The specificity of the affinity-purified antibodies was assayed by immunofluorescence analysis on transfected MDCK cells versus untransfected cells.
In vitro binding experiments
The in vitro binding assays were carried out using GST-tagged fusion proteins as follows: radiolabelled proteins were translated in vitro with the T7-coupled transcription–translation system (Promega), according to the manufacturer's instruction. To test interactions of NGL-1 and MyTH4-FERM domains of myosin XVa with whirlin constructs, a bacterial lysate containing GST constructs of either NGL-1 or MyTH4-FERM domains of myosin XVa, or GST alone, was incubated with pre-equilibrated glutathione–Sepharose beads (Pharmacia) for 1 h at 4°C on a rotating wheel. To verify that the same amounts of GST–proteins and GST alone were used, extracts were analyzed with Coomassie blue staining. The beads were washed three times with binding buffer (PBS with 5% glycerol, 5 mM MgCl2 and 0.1% Triton X-100) supplemented with a protease inhibitor cocktail (Roche), and then incubated with the same amount of 35S-labelled whirlin constructs for 3 h, at 4°C on a rotating wheel. The beads were then washed four times with binding buffer supplemented with 150 mM NaCl, and bound proteins were resuspended in 30 µl 2xSDS sample buffer, and then analyzed on a 4–12% SDS–PAGE.
To determine whether Ca2+ ions affect the homophilic interactions of NGL-1, we repeated the same binding assays with different free Ca2+ concentrations. EGTA (2 mM) was used to remove the free Ca2+ in solution, and CaCl2 at different concentration was used for the Ca2+-dependent binding reactions. The free Ca2+ concentration in buffers was estimated using WEBMAXCLITE version 1.15 software (http://www.stanford.edu).
GST pull-down assays
GST fusion proteins containing long whirlin isoform (1–906) and short whirlin isoform (443–906) were expressed in E. coli BL21 cells using pGEX-4T-1 and purified directly from bacterial extract on glutathione–Sepharose-4B beads. Pull-down assays were performed with recombinant myosin XVa tail (2536–3511) and myosin VIIa tail (847–2215), subcloned into pCMV-tag3B and pcDNA, respectively, and overexpressed in HEK293 cells. Immobilized GST or specific GST–whirlin fusion proteins were incubated with 250 µl of cell lysate expressing the myosin tail, overnight at 4°C. After extensive washes with binding buffer, the bead pellets were resuspended in SDS sample buffer and analyzed by SDS–PAGE and immunoblotting.
Immunoprecipitation
For testing interactions between long whirlin isoform and myosin XVa tail, co-transfected HEK293 cells were lysed and immunoprecipitated with anti-CIP98 antibody, previously coupled to protein-A Sepharose, and immunoprecipitated proteins were analyzed by western blotting for the myosin XVa tail using cMyc antibody (1 : 500). HEK293 cell lysates were prepared by extracting cells with lysis buffer (PBS p. 7.4, 0.5% Triton X-100, 0.1% DOC and a protease inhibitor cocktail) and clarifying the lysate by centrifugation (45 min, 13 000g). Aliquots of the extracts were immunoprecipitated for 6 h at 4°C. Lysates from HEK293 transfected with the myosin XVa tail alone were used as controls.
|
ACKNOWLEDGEMENTS |
|---|
We thank Thomas Friedman (Laboratory of Molecular Genetics, NIDCD, NIH, Rockville, USA) for providing myosin XVa antibody and Carine Houdon for technical help. This work was supported by Fondation pour la Recherche Médicale (ARS2000), European Community (QLG2-CT-1999-00988). B.D. has a fellowship from Letten F. Saugstad's Fund. R.G. and G.R. are supported by The Wellcome Trust (grant 071394/Z/03/Z).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTS AND DISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
- Pickles, J.O., Comis, S.D. and Osborne, M.P. (1984) Cross-links between stereocilia in the guinea-pig organ of Corti and their possible relation to sensory transduction. Hear. Res. 15, 103–112.[CrossRef][Web of Science][Medline]
- Hudspeth, A.J. (1989) How the ear's works work. Nature, 341, 397–404.[CrossRef][Medline]
- Tilney, L.G., Tilney, M.S. and De Rosier, D.J. (1992) Actin filaments, stereocilia, and hair cells: how cells count and measure. Annu. Rev. Cell Biol. 8, 257–274.[CrossRef][Web of Science][Medline]
-
Probst, F.J., Fridell, R.A., Raphael, Y., Saunders, T.L., Wang, A., Liang, Y., Morell, R.J., Touchman, J.W., Lyons, R.H., Noben-Trauth, K. et al. (1998) Correction of deafness in shaker-2 mice by an unconventional myosin in a BAC transgene. Science, 280, 1444–1447.
[Abstract/Free Full Text] - Holme, R.H., Kiernan, B.W., Brown, S.D.M. and Steel, K.P. (2002) Elongation of hair cell stereocilia is defective in the mouse mutant whirler. J. Comp. Neurol., 450, 94–102.[CrossRef][Web of Science][Medline]
-
Rzadzinska, A.K., Schneider, M.E., Davies, C., Riordan, G.P. and Kachar, B. (2004) An actin molecular treadmill and myosins maintain stereocilia functional architecture and self-renewal. J. Cell Biol., 164, 887–897.
[Abstract/Free Full Text] - Mburu, P., Mustapha, M., Varela, A., Weil, D., El-Amraoui, A., Holme, R.H., Rump, A., Hardisty, R.E., Blanchard, S., Coimbra, R.S. et al. (2003) Defects in whirlin, a PDZ domain molecule involved in stereocilia elongation, cause deafness in the whirler mouse and families with mutations in DFNB31. Nat. Genet., 34, 421–428.[CrossRef][Web of Science][Medline]
- Sheng, M. and Sala, C. (2001) PDZ domains and the organization of supramolecular complexes. Annu. Rev. Neurosci., 24, 1–29.[CrossRef][Web of Science][Medline]
- Yap, C.C., Liang, F., Yamazaki, Y., Muto, Y., Kishida, H., Hayashida, T., Hashikawa, T. and Yano, R. (2003) CIP98, a novel PDZ domain protein, is expressed in the central nervous system and interacts with calmodulin-dependent serine kinase. J. Neurochem., 85, 123–134.[Web of Science][Medline]
- Liang, Y., Wang, A., Belyantseva, I.A., Anderson, D.W., Probst, F.J., Barber, T.D., Miller, W., Touchman, J.W., Jin, L., Sullivan, S.L. et al. (1999) Characterization of the human and mouse unconventional myosin XV genes responsible for hereditary deafness DFNB3 and Shaker-2. Genomics, 61, 243–258.[CrossRef][Web of Science][Medline]
-
Belyantseva, I.A., Boger, E.T. and Friedman, T.B. (2003) Myosin XVa localizes to the tips of inner ear sensory cell stereocilia and is essential for staircase formation of the hair bundle. Proc. Natl Acad. Sci. USA, 100, 13958–13963.
[Abstract/Free Full Text] - Zimmermann, P., Meerschaert, K., Reekmans, G., Leenaerts, I., Small, J.V., Vandekerckhove, J., David, G. and Gettemans, J. (2002) PIP(2)-PDZ domain binding controls the association of syntenin with the plasma membrane. Mol. Cell, 9, 1215–1225.[CrossRef][Web of Science][Medline]
- Mayer, B.J. (2001) SH3 domains: Complexity in moderation. J. Cell Sci., 114, 1253–1263.[Abstract]
-
Kikkawa, Y., Mburu, P., Morse, S., Kominami, R., Townsend, A. and Brown, S.D.M. (2005) Mutant analysis reveals whirlin as a dynamic organizer in the growing hair cell stereocilium. Hum. Mol. Genet., 14, 391–400.
[Abstract/Free Full Text] - Tatham, A.S. and Shewry, P.R. (2000) Elastomeric proteins: biological roles, structures and mechanisms. Trends Biochem Sci., 25, 567–571.[CrossRef][Web of Science][Medline]
- Friedman, T.B. and Griffith, A.J. (2003) Human nonsyndromic sensorineural deafness. Annu. Rev. Genomics Hum. Genet., 4, 341–402.[CrossRef][Web of Science][Medline]
- Beyer, L.A., Odeh, H., Probst, F.J., Lambert, E.H., Dolan, D.F., Camper, S.A., Kohrman, D.C. and Raphael, Y. (2000) Hair cells in the inner ear of the pirouette and shaker 2 mutant mice. J. Neurocytol., 29, 227–240.[CrossRef][Web of Science][Medline]
-
Hasson, T., Gillespie, P.G., Garcia, J.A., MacDonald, R.B., Zhao, Y., Yee, A.G., Mooseker, M.S. and Corey, D.P. (1997) Unconventional myosins in inner-ear sensory epithelia. J. Cell Biol., 137, 1287–1307.
[Abstract/Free Full Text] - Self, T., Mahony, M., Fleming, J., Walsh, J., Brown, S.D.M. and Steel, K.P. (1998) Shaker-1 mutations reveal roles for myosin VIIA in both development and function of cochlear hair cells. Development, 125, 557–566.[Abstract]
- Boëda, B., El-Amraoui, A., Bahloul, A., Goodyear, R., Daviet, L., Blanchard, S., Perfettini, I., Fath, K.R., Shorte, S., Reiners, J. et al. (2002) Myosin VIIa, harmonin, and cadherin 23, three Usher I gene products, cooperate to shape the sensory hair cell bundle. EMBO J., 21, 6689–6699.[CrossRef][Web of Science][Medline]
- Lin, J.C., Ho, W.H., Gurney, A. and Rosenthal, A. (2003) The netrin-G1 ligand NGL-1 promotes the outgrowth of thalamocortical axons. Nat. Neurosci., 6, 1270–1276.[CrossRef][Web of Science][Medline]
-
Kuja-Panula, J., Kiiltomaki, M., Yamashiro, T., Rouhiainen, A. and Rauvala, H. (2003) AMIGO, a transmembrane protein implicated in axon tract development, defines a novel protein family with leucine-rich repeats. J. Cell Biol., 160, 963–973.
[Abstract/Free Full Text] - Ferrary, E., Tran Ba Huy, P., Roinel, N., Bernard, C. and Amiel, C. (1988) Calcium and the inner ear fluids. Acta Otolaryngol., 460 (suppl.), 13–17.
-
Siemens, J., Kazmierczak, P., Reynolds, A., Sticker, M., Littlewood-Evans, A. and Muller, U. (2002) The Usher syndrome proteins cadherin 23 and harmonin form a complex by means of PDZ-domain interactions. Proc. Natl Acad. Sci. USA, 99, 14946–14951.
[Abstract/Free Full Text] -
Küssel-Andermann, P., El-Amraoui, A., Safieddine, S., Nouaille, S., Perfettini, I., Lecuit, M., Cossart, P., Wolfrum, U. and Petit, C. (2000) Vezatin, a novel transmembrane protein, bridges myosin VIIA to the cadherin-catenins complex. EMBO J., 19, 6020–6029.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. E. McConnell, J. N. Higginbotham, D. A. Shifrin Jr., D. L. Tabb, R. J. Coffey, and M. J. Tyska The enterocyte microvillus is a vesicle-generating organelle J. Cell Biol., June 29, 2009; 185(7): 1285 - 1298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Li, J. Xue, and E. H. Peterson Architecture of the Mouse Utricle: Macular Organization and Hair Bundle Heights J Neurophysiol, February 1, 2008; 99(2): 718 - 733. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Maerker, E. van Wijk, N. Overlack, F. F.J. Kersten, J. McGee, T. Goldmann, E. Sehn, R. Roepman, E. J. Walsh, H. Kremer, et al. A novel Usher protein network at the periciliary reloading point between molecular transport machineries in vertebrate photoreceptor cells Hum. Mol. Genet., January 1, 2008; 17(1): 71 - 86. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Nishimura-Akiyoshi, K. Niimi, T. Nakashiba, and S. Itohara Axonal netrin-Gs transneuronally determine lamina-specific subdendritic segments PNAS, September 11, 2007; 104(37): 14801 - 14806. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Gosens, E. van Wijk, F. F.J. Kersten, E. Krieger, B. van der Zwaag, T. Marker, S. J.F. Letteboer, S. Dusseljee, T. Peters, H. A. Spierenburg, et al. MPP1 links the Usher protein network and the Crumbs protein complex in the retina Hum. Mol. Genet., August 15, 2007; 16(16): 1993 - 2003. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Michalski, V. Michel, A. Bahloul, G. Lefevre, J. Barral, H. Yagi, S. Chardenoux, D. Weil, P. Martin, J.-P. Hardelin, et al. Molecular Characterization of the Ankle-Link Complex in Cochlear Hair Cells and Its Role in the Hair Bundle Functioning J. Neurosci., June 13, 2007; 27(24): 6478 - 6488. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Kremer, E. van Wijk, T. Marker, U. Wolfrum, and R. Roepman Usher syndrome: molecular links of pathogenesis, proteins and pathways Hum. Mol. Genet., October 15, 2006; 15(suppl_2): R262 - R270. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Mburu, Y. Kikkawa, S. Townsend, R. Romero, H. Yonekawa, and S. D. M. Brown Whirlin complexes with p55 at the stereocilia tip during hair cell development PNAS, July 18, 2006; 103(29): 10973 - 10978. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. van Wijk, B. van der Zwaag, T. Peters, U. Zimmermann, H. te Brinke, F. F.J. Kersten, T. Marker, E. Aller, L. H. Hoefsloot, C. W.R.J. Cremers, et al. The DFNB31 gene product whirlin connects to the Usher protein network in the cochlea and retina by direct association with USH2A and VLGR1 Hum. Mol. Genet., March 1, 2006; 15(5): 751 - 765. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Senften, M. Schwander, P. Kazmierczak, C. Lillo, J.-B. Shin, T. Hasson, G. S. G. Geleoc, P. G. Gillespie, D. Williams, J. R. Holt, et al. Physical and Functional Interaction between Protocadherin 15 and Myosin VIIa in Mechanosensory Hair Cells J. Neurosci., February 15, 2006; 26(7): 2060 - 2071. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Adato, G. Lefevre, B. Delprat, V. Michel, N. Michalski, S. Chardenoux, D. Weil, A. El-Amraoui, and C. Petit Usherin, the defective protein in Usher syndrome type IIA, is likely to be a component of interstereocilia ankle links in the inner ear sensory cells Hum. Mol. Genet., December 15, 2005; 14(24): 3921 - 3932. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Krendel and M. S. Mooseker Myosins: Tails (and Heads) of Functional Diversity Physiology, August 1, 2005; 20(4): 239 - 251. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Adato, V. Michel, Y. Kikkawa, J. Reiners, K. N. Alagramam, D. Weil, H. Yonekawa, U. Wolfrum, A. El-Amraoui, and C. Petit Interactions in the network of Usher syndrome type 1 proteins Hum. Mol. Genet., February 1, 2005; 14(3): 347 - 356. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||