Reticular lamina and vibrations in living mouse cochleae

Tianying Rena,1, Wenxuan Hea, and David Kempb

aOregon Hearing Research Center, Department of Otolaryngology, Oregon Health & Science University, Portland, OR 97239; and bUniversity College London Institute, University College London, London WC1E 6BT, United Kingdom

Edited by Mario A. Ruggero, Northwestern University, Evanston, IL, and accepted by Editorial Board Member Peter L. Strick July 1, 2016 (received for review May 9, 2016) It is commonly believed that the exceptional sensitivity of mamma- (Fig. 1 A and B and Materials and Methods). The results from this lian hearing depends on outer hair cells which generate forces for study do not demonstrate the commonly expected cochlear local amplifying sound-induced basilar membrane vibrations, yet how feedback but instead indicate a global hydromechanical mecha- cellular forces amplify vibrations is poorly understood. In this study, nism for outer hair cells to enhance hearing sensitivity. by measuring subnanometer vibrations directly from the reticular lamina at the apical ends of outer hair cells and from the basilar Results membrane using a custom-built heterodyne low-coherence interfer- Vibrations of the Reticular Lamina and Basilar Membrane in Sensitive ometer, we demonstrate in living mouse cochleae that the sound- Cochleae. Vibrations of the reticular lamina and basilar membrane induced reticular lamina vibration is substantially larger than the measured from a sensitive are presented in Fig. 1 C–J. basilar membrane vibration not only at the best frequency but Below 50 dB sound pressure level (SPL) (0 dB SPL = 20 μPa), the surprisingly also at low frequencies. The phase relation of reticular basilar membrane vibration increased with frequency and reached lamina to basilar membrane vibration changes with frequency by up the maximum at ∼48 kHz (best frequency) and then decreased, to 180 degrees from ∼135 degrees at low frequencies to ∼-45 de- forming a sharp peak (Fig. 1C). The peak became broader and grees at the best frequency. The magnitude and phase differences shifted toward low frequencies as the sound level increased. The between reticular lamina and basilar membrane vibrations are ab- ratios of the basilar membrane displacement to the dis- sent in postmortem cochleae. These results indicate that outer hair placement at low sound levels confirm the best frequency of NEUROSCIENCE cells do not amplify the basilar membrane vibration directly through ∼48 kHz (Fig. 1E). The peak ratio decreased from >100 to ∼3as a local feedback as commonly expected; instead, they actively vi- the sound level increased from 20 to 80 dB SPL, indicating ∼30-dB brate the reticular lamina over a broad frequency range. The outer nonlinear compression. The overlapping curves at frequencies hair cell-driven reticular lamina vibration collaboratively interacts below 35 kHz (Fig. 1E) indicate the linear growth at low fre- with the basilar membrane traveling wave primarily through the quencies. The basilar membrane vibration phase (referred to as the cochlear fluid, which boosts peak responses at the best-frequency malleus phase) decreased with frequency (Fig. 1G). Data in Fig. 1 location and consequently enhances hearing sensitivity and frequency C E selectivity. and demonstrate the high sensitivity, sharp tuning, and non- linearity of basilar membrane responses to ultrasonic sounds in the cochlea | outer hair cells | hearing | cochlear amplifier | interferometry living mouse cochlea, which are similar to previous measurements in squirrel monkeys (19), gerbils (20–23), chinchillas (17, 24, 25), guinea pigs (26–28), and mice (7, 29, 30). s the auditory sensory organ, the mammalian cochlea is able By contrast, the displacement of the reticular lamina vibration to detect soft sounds at levels as low as ∼20 μPa, equivalent to A (Fig. 1D) was substantially larger than that of the basilar mem- vibrations of <1-pm displacement (1), which is >100-fold brane vibration. At 20 dB SPL, reticular lamina vibrations were smaller than the diameter of a hydrogen atom. The cochlea’sre- the greatest approximately at the same frequency as the best markable sensitivity is commonly attributed to a micromechanical frequency of the basilar membrane (∼48 kHz). Notably, the peak feedback system, which amplifies soft sounds using forces gen- erated by outer hair cells (2–10). In addition to active bundle movements (2), mammalian outer hair cells are capable of Significance changing their lengths in response to electrical stimulation in vitro (11–14). This electromechanical force production is termed outer The remarkable sensitivity of mammalian hearing depends on hair cell electromotility or somatic motility, which is produced by auditory sensory outer hair cells, yet how these cells enhance the the membrane protein, prestin (15). Transgenic mice without hearing sensitivity remains unclear. By measuring subnanometer prestin or with altered prestin suffer from severe hearing loss vibrations directly from motile outer hair cells in living , (4, 16). Targeted inactivation of prestin over well-defined cochlear we demonstrate that outer hair cells do not directly amplify the segments dramatically reduces the traveling wave’speakresponse basilar membrane vibration by local feedback as commonly (17). Recent micromechanical measurements in sensitive living expected; instead, they actively vibrate the reticular lamina over a broad frequency range. The outer hair cell-driven reticular lamina mouse cochleae showed (18) that electrical stimulation of outer vibration interacts with the basilar membrane traveling wave hair cells induces vigorous vibrations of the reticular lamina at the through the cochlear fluid, resulting in maximal vibrations at the apical surfaces of the outer hair cells. Whereas this experiment best-frequency location, consequently enhancing hearing sensi- demonstrates that outer hair cells are capable of generating sub- tivity. This finding advances our knowledge of mammalian hear- stantial forces to vibrate cochlear microstructures in vivo, the lack ingandmayleadtostrategiesforrestoringhearinginpatients. of frequency selectivity of the electrically induced reticular lamina vibration, however, is inconsistent with the sharp tuning of the Author contributions: T.R. designed research; T.R. and W.H. performed research; T.R., W.H., basilar membrane vibration. Because electrical stimulation is not a and D.K. analyzed data; and T.R., W.H., and D.K. wrote the paper. natural stimulus of the auditory system, we here investigate how The authors declare no conflict of interest. acoustically induced reticular lamina vibrations interact with bas- This article is a PNAS Direct Submission. M.A.R. is a Guest Editor invited by the Editorial ilar membrane vibrations in sensitive living mouse cochleae, using Board. a custom-built scanning heterodyne low-coherence interferometer 1To whom correspondence should be addressed. Email: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1607428113 PNAS Early Edition | 1of6 Downloaded by guest on September 29, 2021 nonlinear compression. Surprisingly, displacements of the reticular lamina at frequencies far below the best frequency were also ∼10- fold greater than those of the basilar membrane (Fig. 1D). Dis- placement ratios of the reticular lamina to the malleus in Fig. 1F show a peak magnitude of ∼1,000 at low sound levels, which is larger than that of the basilar membrane (Fig. 1E). With an in- crease in sound level from 20 to 80 dB SPL, the peak magnitude decreased by ∼45 dB for the reticular lamina vibration (Fig. 1F) and by ∼30 dB for the basilar membrane vibration (Fig. 1E). The phase responses of the reticular lamina (Fig. 1H) were similar to those of the basilar membrane (Fig. 1G). Ratios of the reticular lamina to basilar membrane displacement greater than one (Fig. 1I) confirm that the reticular lamina vi- brated more than the basilar membrane at all frequencies and sound levels. Near 48 kHz, 30-dB SPL tone-induced reticular lamina vibration was approximately sevenfold greater than that of the basilar membrane. This difference decreased with sound level and approached one at 80 dB SPL. At a given sound level, the magnitude ratio decreased with frequency up to ∼48 kHz. The magnitude ratio curves also shifted toward low frequencies with the sound level increase. Surprisingly, the largest-magnitude ratios occurred at frequencies far below the best frequency and at high sound levels, which likely resulted from the saturation near the best frequency. The phase difference of >90 degrees in Fig. 1J indicates that the reticular lamina and basilar membrane vibrated approximately in opposite directions at frequencies below 15 kHz. The lack of substantial phase difference near 48 kHz, however, demonstrates that the reticular lamina and the basilar membrane moved in the same direction at the best frequency.

Vibrations of the Reticular Lamina and Basilar Membrane in Insensitive Cochleae. Under postmortem conditions, the basilar membrane and reticular lamina vibrations (blue lines in Fig. 2 A and B) showed decreased magnitude near the best frequency, peak shift toward low frequencies, and linear growth with the sound level. In contrast with the lack of substantial postmortem change in basilar membrane vibration at frequencies below 30 kHz in Fig. 2A,the reticular lamina displacements decreased by ∼10-fold (or 20 dB) over the same frequency range (Fig. 2B). These postmortem magnitude changes were confirmed by the displacement ratios of the basilar membrane and reticular lamina to the malleus in Fig. 2 C and D. In contrast to the basilar membrane phase responses (Fig. 2E), the reticular lamina vibration showed a substantial postmortem phase decrease at low frequencies (Fig. 2F). Compared with sensitive responses, the ratio of the reticular lamina to basilar membrane displacement decreased dramatically and approached one (blue lines in Fig. 2G)inthepostmortem cochlea, and the phase difference approached zero (blue lines in Fig. 2H). Thus, under postmortem conditions, the reticular lamina and basilar membrane vibrated synchronously in the same di- rection and with similar magnitudes at all observed frequencies and sound levels.

The Relationship Between the Reticular Lamina and Basilar Membrane Fig. 1. Diagram of the and vibrations of the reticular lamina Vibrations. Average displacements at the best frequency as a and basilar membrane in a sensitive mouse cochlea. (A) Intracochlear image function of the sound level from eight cochleae (Fig. 3A) showed through the intact round window membrane in a living mouse. BM, basilar that reticular lamina displacements were significantly greater than membrane; OSL, osseous spiral lamina. (B) Diagram of the organ of Corti. RL, those of the basilar membrane at 20, 30, 40, and 50 dB SPL (t > reticular lamina; IHC, inner hair cell; OHCs, outer hair cells; TM, tectorial 1.89, P < 0.05, n = 8). The displacement difference decreased with membrane; DCs, Deiters’ cells. (C and D) Displacements of the BM (red) and RL the sound level and became insignificant at 60, 70, and 80 dB SPL (blue) as a function of frequency. The noise floor (black dotted line in C)is t < P > n = <0.01 nm. BF, best frequency. (E and F) Displacement ratios of the BM and RL ( 0.73, 0.35, 8). The phase of the reticular lamina in- to the malleus. (G and H) BM and RL phase referred to the malleus. (I) The ratio creased slightly with the sound level up to 70 dB SPL (Fig. 3B). In of the RL to BM displacement. (J) Phase difference between the RL and BM. postmortem cochleae, the reticular lamina and basilar membrane vibrations increased proportionally with the sound level (Fig. 3C), whereas the corresponding phase changed insignificantly (Fig. 3D). displacement of the reticular lamina was ∼10-fold greater than The average displacement ratio of the reticular lamina to the that of the basilar membrane. At sound levels above 40 dB SPL, basilar membrane and the corresponding phase difference from overlapping curves near the best frequency indicate strong five sensitive and six insensitive cochleae are presented as a

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1607428113 Ren et al. Downloaded by guest on September 29, 2021 Discussion In this study, the reticular lamina and basilar membrane vibrations were measured at a longitudinal location ∼0.85 mm from the cochlear base (red arrows in Fig. 4 A and B) as a function of frequency. Because the peak location of the cochlear partition vibration is determined by frequency (31), the change in stimulus frequency from low to high caused the peak of the traveling wave to move from apex to base. Whereas the vibration at ∼48 kHz was NEUROSCIENCE

Fig. 2. Postmortem changes in reticular lamina and basilar membrane vi- brations. Postmortem data were collected immediately after death (blue) and are compared with those measured under sensitive conditions (red) in the same cochlea. (A) Displacements of basilar membrane (BM) as a function of frequency at different sound levels. (B) Reticular lamina (RL) displacements. (C) Displacement ratios of the BM to malleus. (D) Displacement ratios of the RL to malleus. (E and F) BM and RL phase referred to the malleus. (G) Mag- nitude ratios of the RL to BM. (H) Phase difference between the BM and RL. In addition to poor sensitivity, broad tuning, and linear growth, the magnitude and phase differences between the RL and BM were absent under post- mortem conditions.

E–H function of frequency in Fig. 3 . Because the basilar mem- Fig. 3. Differences between the reticular lamina and basilar membrane brane vibrations at low frequencies can be recorded only at rela- vibrations. (A) Reticular lamina (RL) and basilar membrane (BM) displacements tively high sound levels due to sharp tuning in sensitive cochleae, at different sound levels in sensitive cochleae. (B) RL and BM phase. Phase was the data in Fig. 3 E–H were collected at 80 dB SPL. Fig. 3 E and F normalized to that at 80 dB SPL. (C) RL and BM displacements in postmortem shows that reticular lamina vibration at ∼10 kHz was >fivefold cochleae. (D) Insensitive RL and BM phase. (E) Displacement ratios of the RL to larger than the basilar membrane vibration, with a phase lead of BM show that the RL vibrated >fivefold more than the BM at low frequencies. >90 degrees in sensitive cochleae. Near the best frequency, re- This difference decreased with frequency and approached one near the best frequency. (F) RL phase led the BM phase by ∼135 degree at low frequencies ticular lamina and basilar membrane vibrations had a similar and lagged the BM phase by ∼-45 degrees at the best frequency (48.08 ± magnitude and phase. The magnitude and phase differences were 0.25 kHz, n = 5) in sensitive cochleae. (G and H) The magnitude and phase absent in postmortem cochleae (Fig. 3 G and H). differences were absent in postmortem cochleae.

Ren et al. PNAS Early Edition | 3of6 Downloaded by guest on September 29, 2021 measured at the response peak location (Fig. 4A), the vibration at a lower frequency was measured at the basal part of the traveling wave because the response peak moved toward an apical location (Fig. 4B). Therefore, the frequency responses measured in this study can reflect the relationship between the reticular lamina and basilar membrane vibrations at different parts of traveling waves. The longitudinal patterns of the reticular lamina and basilar membrane vibrations at ∼48 kHz were obtained from measured frequency responses (Fig. 1 C, D, G,andH)(Materials and Methods) and are presented by blue and red lines in Fig. 4 C–F. Whereas 40-dB SPL responses occurred within a ∼200-μmregion(Fig.4C and D), a 70-dB SPL tone-induced vibration extended over the entire region from the response peak to the base (Fig. 4 E and F). The reticular lamina vibration was substantially larger than the basilar membrane vibration not only at the peak (Fig. 4C) but also at the basal part of the traveling wave (Fig. 4E). Phase difference between the reticular lamina and basilar membrane vibrations was >120 degrees at the base (Fig. 4F) but insignificant at the response peak (Fig. 4 D and F). Because outer hair cells sit on the basilar membrane via Deiters’ cells, they can move passively, following the basilar membrane traveling wave. Outer hair cells can also change their length (11) in response to the membrane potential change caused by sound stimulation (32). Because the reticular lamina vibration is a vector sum of the basilar membrane vibration and the outer hair cell- driven movement, the active component of the reticular lamina vibration was obtained by vector subtraction of the basilar membrane vibration from the measured reticular lamina vibration and is shown by green lines in Fig. 4 C–F. The overlapping blue and green lines in Fig. 4 C and D indicate that reticular lamina vibration was dominated by the outer hair cell-driven component at 40 dB SPL. In Fig. 4E,the green line diverges from the blue line and almost overlaps the red line near the best frequency, reflecting saturation of the outer hair cell- driven component at high sound levels near the response peak. These data indicate that acoustically induced reticular lamina vibration is dominated by outer hair cell-driven vibration over the entire traveling wave region. This is confirmed by postmortem changes, which show no substantial difference between the reticular lamina and basilar membrane vibrations (blue lines in Figs. 2 G and H and 3 G and H). The data also show that neither destructive interference between the basilar membrane and reticular lamina vibrations at the base nor constructive interference near the peak is significant (indicated by the negligible gap between blue and green lines in Fig. 4 C–F)duetothe large-magnitude difference between the reticular lamina and basilar membrane vibrations. Because force production of outer hair cells depends on metabolic energy in vivo and the generated forces act equally on both ends of the cells but in opposite directions, it is expected that the outer hair cell-driven reticular lamina and basilar membrane vibrations are physiologically vulnerable and in opposite directions (13, 33, 34). The lack of postmortem changes in the basilar membrane vibration at low frequencies (Fig. 2 A and C) and approximately in-phase vi- brations of the reticular lamina and basilar membrane near the best Fig. 4. Longitudinal patterns and interaction of the cochlear partition vi- frequency (Figs. 1J and 3F) indicate that, under physiological con- brations. The active component of the reticular lamina vibration (green lines ditions, locally generated outer hair cell forces do not drive the in panels C–F) was obtained by vector subtraction of the basilar membrane vibration from the measured reticular lamina vibration. (A) Diagram of the basilar membrane effectively, likely because outer hair cell-generated conventional traveling wave at ∼48 kHz shows that the vibration was forces could not be coupled back to the basilar membrane (35). measured at the peak of the traveling wave (red arrow). (B) At a frequency When sound causes the to move out toward the middle lower than the BF, the vibration was measured at a basal part of the trav- ear, displacing fluid out of the scala vestibuli, the basilar membrane eling wave. (C) At 40 dB SPL, the reticular lamina (RL) and basilar membrane at the cochlear base reacts by moving up (Fig. 4H). This upward (BM) vibrations were at a small region centered at the BF location. (E)At movement of the basilar membrane depolarizes outer hair cells by 70 dB SPL, both RL and BM displacements extended from the BF location to deflecting hair bundles toward the excitatory direction (35). Due to the base. (D and F) Near the base, RL phase led BM phase by >120 degrees; they were approximately in-phase at the BF location. (G) Diagram of a single

tone-induced cochlear traveling wave. DOHC, the outer hair cell force-induced displacement; DTW, the traveling wave-induced displacement. (H) At the base of the traveling wave, RL and BM move in approximately opposite directions. traveling wave, the RL and BM move in the same direction, which results in The active RL movement creates a negative fluid pressure in the scala vestibuli the maximal vibration at the apical ends of outer hair cells through con- and a positive fluid pressure inside the organ of Corti. (I) At the peak of the structive interference. Dotted lines show resting positions.

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1607428113 Ren et al. Downloaded by guest on September 29, 2021 configuration change in the motor protein, prestin, outer hair cells Materials and Methods contract, pulling the reticular lamina toward the basilar membrane Eighteen male and female CBA/CaJ mice at 3–5 wk of age were used in this (Fig. 4H). Unexpectedly, the present results indicate a 10-fold study. The animal use protocol was approved by the Oregon Health & Science greater displacement of the reticular lamina than the basilar University Institutional Animal Care and Use Committee. membrane. Furthermore, the observed phase difference of up to 135 degrees between the two vibrations (Fig. 4F) indicates that The Scanning Heterodyne Low-Coherence Interferometer. Based on a scanning they move in approximately opposite directions at the base. The laser interferometer (50, 51), a scanning heterodyne low-coherence interferometer fluid displaced by this opposing reticular lamina motion may have a was built by replacing the light source and related optical and electronic com- significant influence on the developing traveling wave. Indeed, ponents (18). Taking advantage of the small coherence length, low-coherence electrically evoked reticular lamina vibration has already been interferometers can measure vibration with a high axial resolution (27, 29, 52). shown to initiate a conventional traveling wave (18). Due to the use of the 40-MHz carrier frequency for heterodyne detection, the The unexpectedly large opposing motion of the reticular lamina current low-coherence interferometer has unprecedented sensitivity, wide dy- namic range, no 180-degree phase ambiguity, and high temporal resolution. must also result in substantial fluid displacements inside the organ Although the measurement noise floor is below 0.01 nm (black dotted line in of Corti. When the reticular lamina moves toward the basilar Fig. 1C) and heterodyne interferometry does not require a stable optical path membrane (Fig. 4H) at a basal location, the reticular lamina at a ∼ length, occasional animal movement disturbed low-level measurements. Thus, more apical location 0.5 wavelength (or 180 degrees) from the data points showing irregular patterns, particularly with <10-dB signal-to-noise basal location moves away from the basilar membrane at the same ratio, were not presented. The use of low-coherence light and an objective lens time. Active differential movements of the reticular lamina could with relatively large numerical aperture (NA = 0.42) provides adequate spatial therefore result in longitudinal fluid movements. As the traveling selectivity for measuring the reticular lamina and basilar membrane vibrations waves propagate forward, these longitudinal fluid movements help through the intact round window membrane in mice. transfer energy toward best-frequency locations (6, 36) (Fig. 4G), boosting the peak response. Measurement of the Reticular Lamina and Basilar Membrane Vibrations. Ani- Our interpretation of the present results is also consistent with mal anesthesia and surgical procedures were the same as described previously the following previous studies. Out-of-phase vibrations of the re- (18). Under direct visualization, low-coherence light from the interferometer ticular lamina and basilar membrane at low frequencies (Figs. 1J was focused on the center of the outer hair cell region of the cochlear F partition through the intact round window membrane (Fig. 1A). The dis- and 3 ) are compatible with electrically evoked opposite move- ∼ ments at both ends of outer hair cells (11, 12) and at the reticular tance from the cochlear base to the measurement location was 0.85 mm. The transverse locations of the basilar membrane and reticular lamina were lamina (or tectorial membrane) and the basilar membrane in ex- determined by measuring the backscattered light level and the vibration NEUROSCIENCE cised cochleae (13, 33, 34). The acoustically induced in-phase magnitude and phase as a function of the transverse location. When the vibrations near the best frequency are similar to electrically evoked object beam of the interferometer was focused on the basilar membrane, or reticularlaminaandbasilarmembranevibrations(18)andlow- reticular lamina, acoustical tones at different frequencies and levels were frequency tone-induced vibrations of Hensen’s cells and basilar delivered to the . The tone frequency was changed from 0.75 to membrane (37). The proposed outer hair cell-driven fluid move- 67.5 kHz by <0.75 kHz per step. ment inside the organ of Corti is in agreement with the electrically induced fluid movement inside the tunnel of Corti in vitro (38). Data Analysis and Statistical Methods. Igor Pro (Version 6.35A5, WaveMetrics) Because the outer hair cell-generated pressure inside the organ of was used for processing and analyzing data. Frequency responses of the re- Corti acts in all directions, the proposed mechanism can interpret ticular lamina and basilar membrane were presented by plotting displacement complex vibrations of the organ of Corti (18, 27, 29, 30). Phase and phase as a function of frequency. The transfer functions were estimated by inconsistency between our results and others’ likely resulted the displacement ratio of the reticular lamina or basilar membrane to the from differences in phase detection techniques, animal species, malleus at different frequencies. The relationship between the reticular lamina and measurement locations. Whereas the proposed cochlear and basilar membrane vibrations was presented by the ratio of the reticular lamina displacement to the basilar membrane displacement and the phase micromechanical mechanism is well supported by experimental – difference obtained by subtracting the basilar membrane phase from the reticular data, it does not exclude other mechanisms (33, 39 49). lamina phase. The longitudinal patterns of the reticular lamina and basilar In summary, heterodyne low-coherence interferometry in living membrane vibrations at the best frequency of the measured location were mouse cochleae demonstrates that the sound-induced reticular presented by plotting displacement and phase as a function of the longitudinal lamina vibration is substantially larger than basilar membrane location. The longitudinal location was derived from the stimulus frequency vibration and is physiologically vulnerable not only at the best according to the frequency–location function in mouse cochleae (31) frequency but also at low frequencies and that the reticular lamina [d = ð156.5 − 82.5 × logðfÞÞ=100 × 5.41, where d is distance from the base in and basilar membrane vibrate approximately in the same direction millimeters and f is frequency in kHz]. The grouped results were presented by near the best frequency. These results indicate that, contrary to mean and SE calculated across animals at given frequencies and stimulus levels. expectations, outer hair cells do not amplify the basilar membrane Displacement difference between the reticular lamina and basilar membrane vibration directly by local feedback. Rather, they actively vibrate vibration at the best frequency was determined using a two-tailed t test, and P the reticular lamina over a broad frequency range. The interaction value <0.05 was considered statistically significant. of the outer hair cell-driven reticular lamina vibration with the ACKNOWLEDGMENTS. We thank Peter Barr-Gillespie and John V. Brigande for basilar membrane traveling wave through the cochlear fluid results valuable comments on the manuscript, Alfred L. Nuttall and other colleagues at in maximum responses at the best-frequency location, which con- Oregon Hearing Research Center for helpful discussion of the data, and Edward sequently enhances hearing sensitivity and frequency selectivity. Porsov for technical help. This study was funded by NIH Grant R01 DC004554 (to T.R.).

1. Dalhoff E, Turcanu D, Zenner HP, Gummer AW (2007) Distortion product otoacoustic 6. Reichenbach T, Hudspeth AJ (2014) The physics of hearing: Fluid mechanics and the emissions measured as vibration on the eardrum of human subjects. Proc Natl Acad active process of the inner ear. Rep Prog Phys 77(7):076601. Sci USA 104(5):1546–1551. 7. Russell IJ, et al. (2007) Sharpened cochlear tuning in a mouse with a genetically 2. Hudspeth AJ (2014) Integrating the active process of hair cells with cochlear function. modified tectorial membrane. Nat Neurosci 10(2):215–223. Nat Rev Neurosci 15(9):600–614. 8. Peng AW, Ricci AJ (2011) Somatic motility and hair bundle mechanics, are both nec- 3. Kemp DT (1986) Otoacoustic emissions, travelling waves and cochlear mechanisms. essary for cochlear amplification? Hear Res 273(1-2):109–122. Hear Res 22:95–104. 9. Dong W, Olson ES (2013) Detection of cochlear amplification and its activation. 4. Dallos P (2008) Cochlear amplification, outer hair cells and prestin. Curr Opin Biophys J 105(4):1067–1078. Neurobiol 18(4):370–376. 10. Robles L, Ruggero MA (2001) Mechanics of the mammalian cochlea. Physiol Rev 81(3):1305–1352. 5. Fettiplace R, Hackney CM (2006) The sensory and motor roles of auditory hair cells. 11. Brownell WE, Bader CR, Bertrand D, de Ribaupierre Y (1985) Evoked mechanical re- Nat Rev Neurosci 7(1):19–29. sponses of isolated cochlear outer hair cells. Science 227(4683):194–196.

Ren et al. PNAS Early Edition | 5of6 Downloaded by guest on September 29, 2021 12. Santos-Sacchi J (1989) Asymmetry in voltage-dependent movements of isolated outer 31. Müller M, von Hünerbein K, Hoidis S, Smolders JW (2005) A physiological place- hair cells from the organ of Corti. J Neurosci 9(8):2954–2962. frequency map of the cochlea in the CBA/J mouse. Hear Res 202(1-2):63–73. 13. Mammano F, Ashmore JF (1993) Reverse transduction measured in the isolated co- 32. Corey DP, Hudspeth AJ (1979) Ionic basis of the receptor potential in a vertebrate hair chlea by laser Michelson interferometry. Nature 365(6449):838–841. cell. Nature 281(5733):675–677. 14. Frank G, Hemmert W, Gummer AW (1999) Limiting dynamics of high-frequency 33. Gummer AW, Hemmert W, Zenner HP (1996) Resonant tectorial membrane motion in electromechanical transduction of outer hair cells. Proc Natl Acad Sci USA 96(8): the inner ear: Its crucial role in frequency tuning. Proc Natl Acad Sci USA 93(16): 4420–4425. 8727–8732. 15. Zheng J, et al. (2000) Prestin is the motor protein of cochlear outer hair cells. Nature 34. Nowotny M, Gummer AW (2011) Vibration responses of the organ of Corti and the 405(6783):149–155. tectorial membrane to electrical stimulation. J Acoust Soc Am 130(6):3852–3872. 16. Liberman MC, et al. (2002) Prestin is required for electromotility of the outer hair cell 35. Reichenbach T, Hudspeth AJ (2010) A ratchet mechanism for amplification in low- and for the cochlear amplifier. Nature 419(6904):300–304. frequency mammalian hearing. Proc Natl Acad Sci USA 107(11):4973–4978. 17. Fisher JA, Nin F, Reichenbach T, Uthaiah RC, Hudspeth AJ (2012) The spatial pattern of 36. Lighthill J (1981) Energy flow in the cochlea. J Fluid Mech 106:149–213. cochlear amplification. Neuron 76(5):989–997. 37. Khanna SM, Hao LF (2000) Amplification in the apical turn of the cochlea with neg- 18. Ren T, He W, Barr-Gillespie PG (2016) Reverse transduction measured in the living ative feedback. Hear Res 149(1-2):55–76. cochlea by low-coherence heterodyne interferometry. Nat Commun 7:10282. 38. Karavitaki KD, Mountain DC (2007) Evidence for outer hair cell driven oscillatory fluid 19. Rhode WS (1971) Observations of the vibration of the basilar membrane in squirrel flow in the tunnel of Corti. Biophys J 92(9):3284–3293. monkeys using the Mössbauer technique. J Acoust Soc Am 49(4):2, 1218. 39. Markin VS, Hudspeth AJ (1995) Modeling the active process of the cochlea: Phase 20. Overstreet EH, 3rd, Temchin AN, Ruggero MA (2002) Basilar membrane vibrations relations, amplification, and spontaneous oscillation. Biophys J 69(1):138–147. near the round window of the gerbil cochlea. J Assoc Res Otolaryngol 3(3):351–361. 40. Zwislocki JJ, Kletsky EJ (1979) Tectorial membrane: A possible effect on frequency 21. Versteegh CP, van der Heijden M (2012) Basilar membrane responses to tones and analysis in the cochlea. Science 204(4393):639–641. tone complexes: Nonlinear effects of stimulus intensity. J Assoc Res Otolaryngol 13(6): 41. Neely ST, Kim DO (1983) An active cochlear model showing sharp tuning and high 785–798. sensitivity. Hear Res 9(2):123–130. 22. van der Heijden M, Versteegh CP (2015) Energy flux in the cochlea: Evidence against 42. Allen JB (1980) Cochlear micromechanics–a physical model of transduction. J Acoust power amplification of the traveling wave. J Assoc Res Otolaryngol 16(5):581–597. Soc Am 68(6):1660–1670. 23. Ren T, Nuttall AL (2001) Basilar membrane vibration in the basal turn of the sensitive 43. Hubbard A (1993) A traveling-wave amplifier model of the cochlea. Science gerbil cochlea. Hear Res 151(1-2):48–60. 259(5091):68–71. 24. Ruggero MA, Narayan SS, Temchin AN, Recio A (2000) Mechanical bases of frequency 44. Ghaffari R, Aranyosi AJ, Richardson GP, Freeman DM (2010) Tectorial membrane tuning and neural excitation at the base of the cochlea: Comparison of basilar- travelling waves underlie abnormal hearing in Tectb mutant mice. Nat Commun 1:96. membrane vibrations and auditory-nerve-fiber responses in chinchilla. Proc Natl Acad 45. van der Heijden M (2014) Frequency selectivity without resonance in a fluid wave- Sci USA 97(22):11744–11750. guide. Proc Natl Acad Sci USA 111(40):14548–14552. 25. Recio A, Rich NC, Narayan SS, Ruggero MA (1998) Basilar-membrane responses to 46. Wang Y, Steele CR, Puria S (2016) Cochlear outer-hair-cell power generation and clicks at the base of the chinchilla cochlea. J Acoust Soc Am 103(4):1972–1989. viscous fluid loss. Sci Rep 6:19475. 26. Cooper NP, Rhode WS (1992) Basilar membrane mechanics in the hook region of cat 47. Ramamoorthy S, Deo NV, Grosh K (2007) A mechano-electro-acoustical model for the and guinea-pig cochleae: Sharp tuning and nonlinearity in the absence of baseline cochlea: Response to acoustic stimuli. J Acoust Soc Am 121(5 Pt1):2758–2773. position shifts. Hear Res 63(1-2):163–190. 48. Liu Y, Gracewski SM, Nam JH (2015) Consequences of location-dependent organ of 27. Chen F, et al. (2011) A differentially amplified motion in the ear for near-threshold Corti micro-mechanics. PLoS One 10(8):e0133284. sound detection. Nat Neurosci 14(6):770–774. 49. Ó Maoiléidigh D, Hudspeth AJ (2013) Effects of cochlear loading on the motility of 28. Nilsen KE, Russell IJ (2000) The spatial and temporal representation of a tone on the active outer hair cells. Proc Natl Acad Sci USA 110(14):5474–5479. guinea pig basilar membrane. Proc Natl Acad Sci USA 97(22):11751–11758. 50. Ren T (2002) Longitudinal pattern of basilar membrane vibration in the sensitive 29. Lee HY, et al. (2015) Noninvasive in vivo imaging reveals differences between cochlea. Proc Natl Acad Sci USA 99(26):17101–17106. tectorial membrane and basilar membrane traveling waves in the mouse cochlea. 51. Ren T (2004) Reverse propagation of sound in the gerbil cochlea. Nat Neurosci 7(4): Proc Natl Acad Sci USA 112(10):3128–3133. 333–334. 30. Gao SS, et al. (2014) Vibration of the organ of Corti within the cochlear apex in mice. 52. Hong SS, Freeman DM (2006) Doppler optical coherence microscopy for studies of J Neurophysiol 112(5):1192–1204. cochlear mechanics. J Biomed Opt 11(5):054014.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1607428113 Ren et al. Downloaded by guest on September 29, 2021