|Home | About | Journals | Submit | Contact Us | Français|
Achieving the exquisite sensitivity and frequency selectivity of the mammalian ear requires active amplification of input sound. In this issue of Neuron, Dallos and colleagues demonstrate that the molecular motor prestin, which drives shape changes in the soma of mechanosensory hair cells, underlies mechanical feedback mechanisms for sound amplification in mammals.
Our ability to perceive sound illustrates the extraordinary signal-processing capability of the nervous system. The mammalian auditory system responds to sound-induced vibrations of well less than a nanometer, can amplify signals by more than 100-fold, and has a wide dynamic range that enables humans to perceive frequencies from 20 to 20,000 Hz. Oscillations in air pressure outside the ear induce fluid motions that travel down the cochlear duct and vibrate the basilar membrane (Figure 1). As a consequence of gradual changes in the mechanical properties of the basilar membrane from the base to the apex of the cochlea, each basilar-membrane segment responds to a specific frequency. The vibrations are transferred to the organ of Corti, which sits on top of the basilar membrane and contains three rows of outer hair cells (OHCs) and one of inner hair cells (IHCs). The hair cells convert mechanical vibrations into electrical signals, transmitting the information to neurons. Because hair cells at the cochlear base and apex respond respectively to high and low frequencies, sounds are relayed to the nervous system as a tonotopic map (LeMasurier and Gillespie, 2005).
Gold predicted in 1948 that because viscous damping in the cochlea would otherwise dissipate sound energy, active feedback mechanisms must amplify passive basilar-membrane movements induced by sound (Gold, 1948). The underlying active process has been dubbed the “cochlear amplifier” and depends critically on OHCs, which have few afferent contacts and are thought to act locally in the cochlea. When a pure tone stimulates a passive basilar membrane resonance at a unique location along the cochlear duct, OHCs are activated. OHCs add energy back into the system, enhancing basilar membrane vibration; these movements are detected by IHCs, which are innervated by the preponderance of the afferent neurons that transmit sound information to the CNS. The cochlear amplifier shows remarkable compressive nonlinearity, which results from saturation of the cochlear amplifier, ensuring that the softest sounds are amplified substantially more than loud ones (LeMasurier and Gillespie, 2005).
Two mechanisms have been proposed for cochlear amplification in mammals. The first is based on the observation that OHCs change their length in response to changes in membrane potential, a phenomenon that is called electromotility. The second view holds that amplification arises from active hair bundle movement. While the two mechanisms are not necessarily in conflict, the relative contributions of each have been difficult to dissect because of their mechanical interconnectedness.
The manuscript by Dallos and colleagues in this issue of Neuron addresses the contribution of electromotility to cochlear amplification in mammals (Dallos et al., 2008). Originally observed in isolated OHCs from the guinea pig, electromotility refers to behavior where hyperpolarization lengthens an OHC soma, while depolarization shortens it. Changes in membrane potential evoke nonlinear capacitance changes similar to the gating charge of voltage-dependent ion channels. Electromotility has been demonstrated in OHCs of several mammalian species, but it is not observed in inner hair cells (IHC) or in hair cells from non-mammalian species (Ashmore, 2008). In a search for the molecular motor, Dallos and colleagues identified in a subtractive hybridization screen a gene they named prestin (from presto, which means fast in Italian), which is selectively expressed in OHCs (Zheng et al., 2000). Prestin is a member of the sulfate anion-transporter family. When expressed in cell lines, it confers both nonlinear capacitance and voltage-dependent motility, presumably from voltage-dependent changes in protein conformation. Immunolocalization studies have shown that prestin is enriched in the lateral membrane of OHCs, where it likely is a component of the 11 nm particles that are present in OHCs but not in IHCs (Ashmore, 2008).
In 2002, to define the function of prestin, Liberman, Zuo, and colleagues generated prestin knockout mice (Liberman et al., 2002). Electromotility in isolated OHCs from knockout mice was dramatically impaired. Measurement of the auditory brainstem response (ABR) and compound action potential (CAP) demonstrated a more than 100-fold drop in sound sensitivity. The frequency sensitivity of the inner ear was also perturbed, with broadened frequency tuning and a shift in the characteristic frequency to lower frequencies, typical of manipulations that cause the loss of the cochlear amplifier. Subsequent studies analyzed the properties of mechanotransduction currents in hair cells of the knockout mice but did not observe any differences from wild-type hair cells (Dallos et al., 2006). Taken together, these findings demonstrate that in the absence of prestin, the ability of the inner ear to amplify sound and discriminate frequencies is impaired.
While the results obtained with the prestin knockout mice are compelling, it has remained unclear whether the observed defects are a direct consequence of defects in the cochlear amplifier. Significantly, the analysis of the prestin mutant mice revealed that they had shorter-than-average OHCs and substantial hair-cell loss in high-frequency regions (Liberman et al., 2002). Moreover, because prestin is required for normal axial stiffness of the OHC soma (Ashmore, 2008), reduction in of OHC stiffness when prestin is eliminated should affect the passive mechanical properties of the cochlea. This prediction was illustrated strikingly when Mellardo Lagarde and colleagues measured basilar-membrane motion in prestin knockout mice; while they found as expected broadened frequency tuning, sensitivity was normal, i.e., 100-fold greater than when the cochlear amplifier was poisoned (Mellado Lagarde et al., 2008). These results show that simple loss of prestin leads to dramatic mechanical changes in the cochlea that may be due simply to reduced stiffness of hair cells.
Where does this leave prestin-mediated electromotility and its importance for cochlear function? Electromotility could be necessary to feed mechanical energy back into the system to amplify basilar membrane vibrations to produce sensitive, sharply tuned responses of the cochlea. To test this model, it would be useful to engineer a mouse line where OHCs express a mutant prestin that lacks electromotility but is properly localized and present at normal concentrations, allowing OHCs to retain their wild-type axial stiffness. A mouse with just these properties is reported in the current issue of Neuron (Dallos et al., 2008). Dallos and colleagues generated a knockin mouse that carries two point mutations affecting a domain of prestin between the last transmembrane domain and the intracellular C terminus. The mutations were chosen because they perturb nonlinear capacitance of tissue-culture cells expressing mutant prestin without affecting the protein’s targeting (Zheng et al., 2005). In the knockin mice, the mutant protein was properly targeted to the OHCs basolateral membrane. OHC length and axial stiffness were not affected in the mutants, and characteristic features of mechanotransduction currents, such as the kinetics of channel activation and adaptation were not affected. However, nonlinear capacitance and electromotility of isolated OHCs was drastically reduced (>90%), which is predicted to lead to a loss in the sound sensitivity of the cochlea by about 100-fold. Measurement of CAPs confirmed the loss in sound sensitivity and frequency selectivity of the knockin mice. These data provide compelling evidence that prestin-mediated electromotility, not just passive mechanics, is an integral part of the active mechanism necessary for sound amplification and frequency tuning in the cochlea.
Several important issues remain, with the most important being the relationship between stereocilia amplification mechanisms and prestin. The results of Dallos and colleagues (2008) do not demonstrate whether prestin contributes cycle-by-cycle energy input to the basilar membrane, although the molecule’s speed suggests that it is capable of this function. Because prestin responds to membrane voltage, for electromotility to operate at kilohertz rates in the cochlea, hair cells must overcome the intrinsic limit of current-to-voltage conversion that is imparted by the cell’s membrane time constant; however, several mechanisms have been proposed to overcome this constraint (Ashmore, 2008). In another model, stereocilia mechanisms could mechanically amplify in the cochlea, with prestin controlling amplification by biasing hair bundles to a mechanically sensitive operating point (Fettiplace, 2006). If this is the role for prestin, the molecule’s amazing speed must be an epiphenomenon of evolution; Dallos et al. (2008) rightly question this assumption.
Another unsettled issue is how sharp frequency tuning arises, as electromotility is not intrinsically tuned. One solution is that stereocilia mechanotransduction, which is tuned in the cochlea (Beurg et al., 2006), sets a narrow frequency window for a hair cell and prestin only amplifies filtered input signals. It is also possible that prestin does not amplify mechanically but instead is involved in focusing input sound energy to a restricted area on the basilar membrane (Ren and Gillespie, 2007); this role is more in tune with the results of Mellano Lagarde, which illustrate that addition of prestin reduces the passive sensitivity of the cochlea.
Finally, the Mellano Lagarde data show clearly that all of prestin’s roles in the cochlea are not yet understood, and that measuring basilar-membrane motion in the knockin mice reported by Dallos et al. (2008) is of critical importance. Comparison of the knockout and knockin mice will allow dissection of prestin’s passive contribution to basilar-membrane mechanics, including not only vibration amplitude but also efficient excitation of IHCs, the output cells of the cochlea.