Ohc Somatic

© 2008 Nature Publishing Group http://www.nature.com/natureneuroscience

B R I E F C O M M U N I C AT I O N S

Outer hair cell somatic, not hair bundle, motility is the basis of the cochlear amplifier Marcia M Mellado Lagarde1,2, Markus Drexl1,2, Victoria A Lukashkina1, Andrei N Lukashkin1 & Ian J Russell1 Sensitivity, dynamic range and frequency tuning of the cochlea are attributed to amplification involving outer hair cell stereocilia and/or somatic motility. We measured acoustically and electrically elicited basilar membrane displacements from the cochleae of wild-type and Tecta DENT/DENT mice, in which stereocilia are unable to contribute to amplification near threshold. Electrically elicited responses from Tecta DENT/DENT mice were markedly similar to acoustically and electrically elicited responses from wild-type mice. We conclude that somatic, and not stereocilia, motility is the basis of cochlear amplification. Mechanoelectrical sensory transduction in the mammalian cochlea takes place in the organ of Corti, which is sandwiched between the basilar and tectorial membranes (Fig. 1a). Sensory transduction is initiated when sound-induced vibrations of the basilar membrane deflect the hair bundles of stereocilia at the apical poles of the outer hair cells (OHCs), gating the mechanosensitive transducer channels at their tips1. Deflection occurs through radial shear between the reticular lamina and the tectorial membrane to which the hair bundles project. The resulting flow of current into the sensory-motor OHCs initiates active mechanical forces that amplify low-level and compress high-level basilar membrane displacements2. Amplification is frequency dependent, and, for every location in the cochlea, effective amplification occurs around the characteristic frequency to which the location is tuned, thus providing the exquisite sensitivity and enormous dynamic range of the cochlea3,4. Two mechanisms have been suggested to function as the cochlear amplifier. One is voltage-dependent somatic motility resulting from the activity of the motor protein prestin in the OHC lateral membranes1. The other relies on hair-bundle motility driven by calcium currents1. The prestin motor is unique to OHCs of the mammalian cochlea and is essential for cochlear frequency tuning and sensitivity1. Prestin, however, has not been shown to be the unequivocal source of cochlear amplification1. Calcium-dependent hair-bundle motility is a ubiquitous feature of hair cells1, but only in vitro studies have predicted that it is the source of amplification in the mammalian cochlea5,6. Here, we investigate the involvement of hair-bundle and somatic motility in the amplification and tuning of basilar membrane responses

to acoustic and electrical cochlear stimulation in wild-type Tecta+/+ and mutant TectaDENT/DENT mice7. The OHC hair bundles of Tecta+/+ mice are coupled mechanically to the tectorial membrane, but those of TectaDENT/DENT mice are freestanding and are not attached to the tectorial membrane, which is vestigial and is not associated with the organ of Corti in TectaDENT/DENT mice7. Basilar membrane responses to acoustic cochlear stimulation depend on sensory transduction that is mediated via displacements of the OHC hair bundles. The hair bundles of Tecta+/+ mice are displaced through interaction with the tectorial membrane, whereas those of TectaDENT/DENT mice are displaced by fluid flow when basilar membrane velocity is sufficiently large7. Electrical stimulation of the cochlea bypasses sensory transduction8 and directly drives both OHC somatic and hair-bundle motility1,5,6. In Tecta+/+ mice, however, the presence of the tectorial membrane will permit electrically elicited hair-bundle movements to interact with the tectorial membrane, an opportunity that is precluded to the hair bundles of TectaDENT/DENT mice7. By exploiting the differences between Tecta+/+ and TectaDENT/DENT mice in hair-bundle tectorial membrane interaction, we were able to conclude that somatic, and not hair-bundle, motility is the basis of cochlear amplification for near-threshold displacements of the basilar membrane. We measured the frequency tuning, sensitivity and gain (amplification) of acoustically elicited basilar membrane displacements in the basal, high-frequency turn of the cochleae of Tecta+/+ mice (Supplementary Methods online). These measurements were compared with those taken from TectaDENT/DENT mice, which lack a functional tectorial membrane, and with those from the electrically elicited basilar membrane responses from both Tecta+/+ and TectaDENT/DENT mice described below. In response to tones at the characteristic frequency of the basilar membrane location, the dependence of basilar membrane displacement on sound pressure level (SPL) for Tecta+/+ mice was more nonlinear, compressive and sensitive than that measured from TectaDENT/DENT mice7 (Fig. 1b). Threshold frequency-tuning curves, where the sound pressure required to elicit a basilar membrane threshold displacement is plotted as a function of frequency (Fig. 1c,d and Supplementary Fig. 1 online), showed that those of TectaDENT/DENT mice were about 25 dB less sensitive at the characteristic frequency and had much broader tuning with a smaller Q10dB (ratio of the characteristic frequency to the bandwidth measured 10 dB from the tip)7 than the basilar membrane tuning curves from Tecta+/+ mice (Table 1). The tuning curves had additional threshold minima below the characteristic frequency. One of these at R1 (B0.6 octaves below the characteristic frequency; Table 1) has been described previously and is attributed to the resonance of the tectorial membrane7. It was present, therefore, in tuning curves from Tecta+/+, but not TectaDENT/DENT, mice. The other minimum at R2 (B0.3 octaves below the characteristic frequency; Table 1), which is apparent in the tuning curves of both Tecta+/+ and TectaDENT/DENT mice, is recognized here for the first time, to the best of our knowledge,

1School of Life Sciences, University of Sussex, Falmer, Brighton, BN1 9QG, UK. 2These authors contributed equally to this work. Correspondence should be addressed to I.J.R. ([email protected]).

Received 5 February; accepted 28 April; published online 30 May 2008; doi:10.1038/nn.2129

746

VOLUME 11

[

NUMBER 7

[

JULY 2008 NATURE NEUROSCIENCE

B R I E F C O M M U N I C AT I O N S

Inner hair cell

b Hair bundle Reticular lamina

OHC Basilar membrane

R1 R2 CF

50 40

0.0

30

–1.0

20 –2.0

10

–3.0

20

30

–4.0 50 60 70

40

60 kHz 54 kHz 64 kHz

1.0

0.1 20 30 40 50 60 70 80 Sound pressure (dB SPL)

d

70 60

90 80 70 60 50 40 30 20

R1 R2

f

CF

1,000

100

10 30

40 50 60 70 80 Stimulus frequency (kHz)

Basilar membrane displacement/ malleus displacement

Basilar membrane displacement/ malleus displacement

10,000

R2 CFe

0.5 0.0 –0.5 –1.0 –1.5 –2.0 –2.5 –3.0 –3.5 20 30 40 50 60 70 Stimulus frequency (kHz)

Stimulus frequency (kHz)

e

50 kHz

Basilar membrane phase (cycles)

Basilar membrane threshold (dB SPL)

c

Basilar membrane phase (cycles)

© 2008 Nature Publishing Group http://www.nature.com/natureneuroscience

Outer pillar Laser cell beam Scala tympani

10.0

Basilar membrane displacement (nm)

Scala media Tectorial membrane

Basilar membrane threshold (dB SPL)

a

10,000

Figure 1 Basilar membrane responses to acoustic stimulation of the mouse cochlea. (a) Cross-section of the cochlea. The organ of Corti and the location of the laser diode beam for measuring basilar membrane vibrations are shown. (b) Basilar membrane displacement as a function of sound pressure measured at the characteristic frequency (CF, 60 kHz, squares) and at 50 kHz (circles) from the 60-kHz location in the cochlea of a Tecta+/+ mouse (red), and at the characteristic frequency (64 kHz, squares) and at 54 kHz (circles) from the 64-kHz location in the cochlea of a Tecta DENT/DENT mouse (blue). The dotted line represents a slope of 1. (c,d) Threshold frequency tuning curves (open symbols), referred to the malleus, measured from the basal turn basilar membrane in the cochleae of a Tecta+/+ (c) and a Tecta DENT/DENT (d) mouse. The phase of basilar membrane displacement relative to the motion of the malleus (solid symbols) measured 15 dB above threshold is shown (right vertical axis). (e,f) Basilar membrane displacement divided by malleus displacement as a function of stimulus frequency measured from a Tecta +/+ (e) mouse for tones at levels between 25–60-dB SPL in 5-dB steps and a Tecta DENT/DENT (f) mouse for tones at levels between 48–80 dB SPL in 2-dB steps. Black traces in e are for post mortem measurements and for levels 470-dB SPL in f. The vertical dashed lines indicate the characteristic frequency and CFe, R2 and R1 frequencies. All procedures involving animals were carried out in accordance with the UK Home Office regulations with approval from the University of Sussex ethics committee.

R2 CFe

1,000

100

10 30

40 50 60 70 80 Stimulus frequency (kHz)

although it has been apparent in previous measurements. The minima at R1 and R2 are associated with phase jumps of B1801 (Fig. 1c,d), an indication that these minima represent resonances in the basilar membrane vibrations. We suggest that R2 is the result of an additional mode of vibration of the basilar membrane9,10 that, similar to R1, does not appear as a threshold minimum in neural tuning curves11. The ratio of the magnitudes of basilar membrane vibrations to those of the malleus in the mechanical input pathway to the cochlea provides a measure of cochlear amplification. We examined a Tecta+/+ preparation (Fig. 1e) and found the ratio to be B3,000 at the characteristic frequency, compared with B50 post mortem (a difference of B30 dB, which we refer to here as ‘gain’). Gain was greatly reduced at the characteristic frequency of tuning curves measured from TectaDENT/DENT mice (Fig. 1f and Table 1), confirming earlier findings7. Amplification associated with the characteristic frequency was reduced in the majority of TectaDENT/DENT mice; therefore, we estimated the characteristic frequency (CFe ) according to the recording location on the basilar membrane and from our finding that the characteristic frequency is normally B0.3 octaves above the more apparent minimum at R2. The minima at R1 and R2 were apparently not associated with substantial cochlear amplification.

For near-threshold acoustic stimulation about the characteristic frequency, feedback from the OHCs contributes maximally to basilar membrane displacements in a frequency-dependent manner2–4,8. Nearthreshold basilar membrane displacements elicited through extra cochlear stimulation of both Tecta+/+ and TectaDENT/DENT mice in response to stimulation via an electrode on the round window membrane8 are very similar to those measured from Tecta+/+ mice in response to acoustic stimulation. Growth of electrically elicited basilar membrane displacements for near-threshold responses at or near the characteristic frequency was, therefore, nonlinear and compressive (Fig. 2a) and basilar membrane threshold tuning curves for both Tecta+/+ and TectaDENT/DENT mice (Fig. 2b,c and Table 1) were very similar in form and characteristics to those obtained in response to acoustic stimulation in Tecta+/+ mice. The R2 resonance was present in the tuning curves of both Tecta+/+ and TectaDENT/DENT mice, but R1 was absent from tuning curves measured from TectaDENT/DENT mice. The basilar membrane responses due to electrical stimulation are presumably the result of the simultaneous excitation of OHCs in the basal turn of the cochlea. There were, however, phase jumps of B1801 associated with R2 and R1 and an accumulating phase lag of B4501 that was associated with the characteristic frequency, which are similar to those obtained with acoustic stimulation. The gains of electrically elicited vibrations of the basilar membrane were obtained by comparing the sensitivities of basilar membrane displacement (nm per mA) for threshold currents with those at high levels (224 mA root mean square, RMS; Fig. 2d,e). The gains were found to be very similar for both Tecta+/+ and TectaDENT/DENT mice and were not significantly different from those of the acoustically stimulated cochleae of Tecta+/+ mice (P r 0.05; Table 1).

Table 1 Characteristics of acoustically and electrically elicited basilar membrane tuning curves Acoustic

R1 Oct o CF

Tecta+/+

R2 Oct o CF

Characteristic frequency (kHz)

Characteristic frequency sensitivity (dB SPL)

0.61 ± 0.07

0.27 ± 0.04

64.5 ± 2.5

TectaDENT/DENT Electrical

–

0.34 ± 0.08

62.2 ± 4.9

55 ± 8 Sensitivity (mA RMS)

19 ± 3

Tecta+/+ TectaDENT/DENT

0.60 ± 0.08 –

0.33 ± 0.06 0.37 ± 0.04

65.2 ± 3.4 63.2 ± 5.2

6.75 ± 1.75 6.41 ± 1.63

Q10dB

Gain (dB)

10.0 ± 1.4

27.5 ± 4.9

5

–

6.7 ± 2.7

5

29.9 ± 5.4 28.5 ± 2.4

5 9

9.5 ± 1.6 9.6 ± 0.4

n

Measurements from the basal turn of the cochleae of Tecta+/+ and TectaDENT/DENT mice. Acoustic stimulation, threshold gain at the characteristic frequency versus gain post mortem; Electrical stimulation, threshold gain at the characteristic frequency versus gain at maximum stimulus current; Oct o CF, octaves below the characteristic frequency of the measurement location on the basilar membrane.

NATURE NEUROSCIENCE VOLUME 11

[

NUMBER 7

[

JULY 2008

747

B R I E F C O M M U N I C AT I O N S Figure 2 Basilar membrane responses to electrical stimulation of the mouse cochlea. 100 (a) Basilar membrane displacement as a function 100 10.0 R1 R2 CF R2 CF of stimulating current (mA RMS) measured at the 70 kHz 0.0 0.5 64 kHz characteristic frequency (70 kHz, squares) and at 16 kHz 10 –0.5 10 0.0 16 kHz (circles) from the 70-kHz location in the 16 kHz –1.0 1.0 cochlea of a Tecta+/+ mouse (red), and at the –0.5 –1.5 –2.0 characteristic frequency (64 kHz, squares) –1.0 1 1 –2.5 and at 16 kHz (circles) from the 64-kHz –1.5 –3.0 location in the cochlea of a Tecta DENT/DENT –3.5 –2.0 0.1 1 10 100 1,000 30 40 50 60 70 80 30 40 50 60 70 80 mouse (blue). The dotted line represents a slope Stimulus current (µA RMS) Stimulus frequency (kHz) Stimulus frequency (kHz) of 1. (b,c) Threshold frequency tuning curves (open symbols) measured from the basal turn 10,000 10,000 100 basilar membrane in the cochlea of a Tecta+/+ Before R2 CF R1 R2 CF After salicylate 10.0 (b) and a Tecta DENT/DENT (c) mouse. The threshold stimulus current (left ordinate) is expressed in 1,000 1,000 1.0 mA RMS. The phase of basilar membrane 10 displacement (solid symbols) is plotted for a 100 100 0.1 constant stimulus current of 15 dB above threshold for both genotypes. (d,e) Sensitivity of µA RMS 1 10 10 basilar membrane displacement to stimulating 30 40 50 60 70 80 90 100 30 40 50 60 70 80 30 40 50 60 70 80 current in 2-dB steps (6.7–223.6 mA RMS) Sound pressure (dB SPL) Stimulus frequency (kHz) Stimulus frequency (kHz) versus stimulus frequency measured from a Tecta+/+ (d) and a Tecta DENT/DENT (e) mouse. Black traces represent stimulus currents between 35.4–223.6 mA RMS. The vertical dashed lines indicate the characteristic frequency and the R2 and R1 frequencies. The data in each figure come from different Tecta+/+ and Tecta DENT/DENT mice. (f) Means ± s.d. (n ¼ 5) of cochlear microphonic potentials (CM) as functions of SPL recorded before and after salicylate treatment. Inset, basilar membrane displacement as a function of electrical stimulus level recorded before and after placing a crystal of salicylate on the round window membrane (typical example of the data obtained in Tecta DENT/DENT mice; Tecta+/+ data are similar.). The dashed line represents the measurement noise floor. Stimulus current (µA RMS)

Stimulus current (µA RMS)

CM magnitude (µV)

0 50 10 0 15 0 20 0 25 0

f

Basilar membrane sensitivity –1 (nm µA )

© 2008 Nature Publishing Group http://www.nature.com/natureneuroscience

Basilar membrane sensitivity –1 (nm µA )

748

e

Basilar membrane displacement (nm)

Electrically evoked basilar membrane vibrations, but not the soundevoked receptor potentials (cochlear microphonic potentials), in Tecta+/+ and TectaDENT/DENT mice were completely blocked by sodium salicylate in the range of currents used in these experiments12 (Fig. 2f), presumably by blocking the functioning of the motor protein prestin and, therefore, OHC somatic motility13,14. Basilar membrane frequency tuning measured from Tecta+/+ mice in response to acoustic stimulation and from both Tecta+/+ and TectaDENT/DENT mice in response to extracochlear electrical stimulation are very similar near threshold and reflect the electromechanical properties of the cochlea when dominated by contributions from the OHCs2–4,7,11. Measurements made from guinea pigs8 were the first to show that basilar membrane responses to electrical extracochlear and acoustic stimulation were similar. This similarity is notable because acoustic and electrical stimulation excite cochlear amplification in different ways. Electrical stimuli delivered to the round window, which drives cochlear amplification directly8, can excite both OHC somatic and hair-bundle motility5,6. Either of these potential sources could, therefore, underlie cochlear amplification in Tecta+/+ mice. At low stimulus levels, electrically elicited hair-bundle motility cannot, however, amplify basilar membrane responses measured from TectaDENT/DENT mice. In TectaDENT/DENT mice, the sensory bundles of the OHCs are freestanding and therefore cannot react against the tectorial membrane7. For hair-bundle motility to be the source of cochlear amplification, OHC hair bundles must interact with the tectorial membrane if they are to exert forces on the basilar membrane5. Accordingly, the amplification and compression of the basilar membrane responses that we have measured from the cochleae of TectaDENT/DENT mice in response to near-threshold electrical stimulation, which are similar in all respects to near-threshold acoustically and electrically elicited responses from Tecta+/+ mice, cannot be the result of hair-bundle motility. This finding leads us to propose that, at levels close to the threshold of hearing, it is not necessary to invoke a role for hairbundle motility in amplifying the vibrations of the basilar membrane. Indeed, electrically evoked basilar membrane vibrations in both Tecta+/+ and TectaDENT/DENT mice are completely suppressed by salicylate.

c

Basilar membrane phase (cycles)

d

b

Basilar membrane phase (cycles)

Basilar membrane displacement (nm)

a

Salicylate has been shown to block prestin-based OHC somatic motility13,14, but has no direct action on the hair cell mechanoelectrical transducer channels15. We are thus led to attribute amplification in the basal turn of the cochlea to the prestin-based somatic motility of OHCs. Note: Supplementary information is available on the Nature Neuroscience website.

ACKNOWLEDGMENTS We thank G. Richardson for making the Tecta mice available. We thank J. Hartley for technical assistance and C. Kros for helpful comments on the manuscript. This work was supported by the Medical Research Council. M.M.M.L. was supported by a Federation of European Neuroscientists—International Brain Research Organization Fellowship, and M.D. was supported by a Deutsche Forschungsgemeinschaft Fellowship. AUTHOR CONTRIBUTIONS M.M.M.L. and M.D. contributed equally to measuring and analyzing the responses to acoustical and electrical stimulation of the cochlea. V.A.L. measured and analyzed responses to acoustical stimulation. A.N.L. contributed to the experimental design, directed the research and wrote the software. I.J.R. designed the experiments and contributed largely to data analysis and writing the paper. Published online at http://www.nature.com/natureneuroscience/ Reprints and permissions information is available online at http://npg.nature.com/ reprintsandpermissions/

1. Fettiplace, R. & Hackney, C.M. Nat. Rev. Neurosci. 7, 19–29 (2006). 2. Dallos, P. in The Cochlea (eds. Dallos, P., Popper, A.N. & Fay, R.R.) 1–43 (Springer, New York, 1996). 3. Lukashkin, A.N., Walling, M.N. & Russell, I.J. Curr. Biol. 17, 1340–1345 (2007). 4. Robles, L. & Ruggero, M.A. Physiol. Rev. 81, 1305–1352 (2001). 5. Chan, D.K. & Hudspeth, A.J. Biophys. J. 89, 4382–4395 (2005). 6. Kennedy, H.J., Crawford, A.C. & Fettiplace, R. Nature 433, 880–883 (2005). 7. Legan, P.K. et al. Neuron 28, 273–285 (2000). 8. Nuttall, A.L. & Ren, T. Hear. Res. 92, 170–177 (1995). 9. Nowotny, M. & Gummer, A.W. Proc. Natl. Acad. Sci. USA 103, 2120–2125 (2006). 10. Karavitaki, K.D. & Mountain, D.C. Biophys. J. 92, 3294–3316 (2007). 11. Allen, J.B. & Fahey, P.F. J. Acoust. Soc. Am. 94, 809–817 (1993). 12. Murugasu, E. & Russell, I.J. Aud. Neurosci. 1, 139–150 (1995). 13. Oliver, D. et al. Science 292, 2340–2343 (2001). 14. Santos-Sacchi, J., Song, L., Zheng, J. & Nuttall, A.L. J. Neurosci. 26, 3992–3998 (2006). 15. Kennedy, H.J., Evans, M.G., Crawford, A.C. & Fettiplace, R. J. Neurosci. 26, 2757–2766 (2006).

VOLUME 11

[

NUMBER 7

[

JULY 2008 NATURE NEUROSCIENCE