TRPA1 antibody labelling of hair cells in mouse semicircular canal. Red, TRPA1; blue, tubulin in kinocilia; green, actin in stereocilia. Image courtesy of D. Corey, Harvard Medical School, Boston, Massachusetts, USA.

The long search for the mechanosensitive transduction channel that allows hair cells in the mammalian inner ear to function might be over. Corey et al., writing in Nature, provide evidence in support of the idea that the channel, or at least a component of it, is the transient receptor potential (TRP) channel TRPA1.

Although it has been clear for some time that auditory transduction in vertebrates depends on the mechanical deflection of bundles of stereocilia on hair cells, and that this deflection opens ion channels in the tips of the stereocilia that are mechanically gated, the identification of these channels has proved tricky. However, the known permeability and conductance characteristics of the mechanosensitive channel are consistent with those of TRP channels, which are responsible for sensory transduction in other modalities including taste, thermal sensation and insect hearing.

Therefore, Corey and colleagues used in situ hybridization for all of the TRP channels in the mouse genome to search for one that was expressed in the mouse inner ear. TRPA1 was expressed in hair cells of both the cochlea and the vestibular system in mice, and its expression peaked at the developmental time point when these hair cells first become mechanosensitive. Antibody labelling showed that the cellular distribution of the channels was consistent with a role for TRPA1 in mechanotransduction in the stereocilia.

To test the idea that TRPA1 was involved in transduction, the authors used various methods to inhibit its expression. In zebrafish hair cells they used morphelino oligonucleotides, which inhibit the translation or splicing of mRNA, and in mouse hair cells they used adenoviruses to introduce small inhibitory RNAs (siRNAs) that targeted the TRPA1 message. In both cases, channel function was reduced, as measured either by dye accumulation or electrical responses (decreased microphonic potentials in zebrafish and decreased transduction currents in mouse cells).

In zebrafish, another TRP channel, Trpn1, is also required for hair-cell transduction, but this channel is not found in mammalian genomes. Corey et al. suggest that the two channels might form heteromeric channels in zebrafish. Interestingly, although they are not closely related phyogenetically, both TRPA1 and Trpn1 contain many ankyrin domains close to their amino termini. The authors propose that these repeats form a spring that could correspond to the 'gating spring' that has been biophysically identified as functioning in the hair-cell transduction channel complex. Yet another function for this versatile protein could come from its ability to undergo fast adaptation after opening. The rapid conformational change that would result from channel closing might amplify the vibration of the basilar membrane in the cochlea. This 'cochlear amplifier' is believed to mediate frequency tuning. If TRPA1 turns out to represent the transduction channel, gating spring and cochlear amplifier in one, it would be a remarkable example of evolutionary ingenuity.