Models of Ca2+ release channel adaptation.

13. G. D. Housley and J. F. Ashmore, J. Physiol. (London) 448, 73 (1992). 14. J. F. Ashmore and R. W. Meech, Nature 322, 368 (1986); A. H. Gitter and H. P. Zenner, in Basic Issues in Hearing, H. Duifhuis, J. W. Horst, H. P. Wit, Eds. (Academic Press, London, 1988), pp. 32-39; J. F. Ashmore, in Cochlear Mechanisms, J. P. Wilson and D. T. Kemp, Eds. (Plenum, New York, 1989), pp. 107-113; J. Santos-Sacchi, J. Neurosci. 11, 3096 (1991). 15. J. Santos-Sacchi and D. P. Dilger, Hearing Res. 35, 143 (1988). 16. A. Forge, Cell Tissue Res. 265, 473 (1991). 17. M. Kdssl and 1. J. Russell, J. Neurosci. 12, 1575 (1992); I. J. Russell and M. Kdssl, Proc. R. Soc. London B 247, 97 (1992). The values used here for extracellular responses were confirmed by 1. J. Russell as being representative of the most sensitive preparations. 18. P. Dallos, Hearing Res. 12, 89 (1983). 19. P. M. Sellick, R. Patuzzi, B. M. Johnstone, J. Acoust. Soc. Am. 72,131 (1982). 20. Control of cochlear amplification by extracellular voltages has been mentioned in passing by a number of authors, for example, H. Davis [Am. J. Otolaryngol. 2, 153 (1981)] and (15). 21. C. D. Geisler, G. K. Yates, R. B. Patuzzi, B. M. Johnstone, Hearing Res. 44, 241 (1990); M. A. Cheatham and P. Dallos, unpublished data. 22. C. D. Hopkins, Am. Zool. 21, 211 (1981). 23. A. J. Hudspeth and D. P. Corey, Proc. Natl. Acad. Sci. U.S.A. 74, 2407 (1977). 24. Detailed descriptions of the experimental methods have been published (11). OHCs were obtained from the cochleas of young anesthetized albino guinea pigs (care and maintenance of animals was in accord with institutional guidelines). For these specific experiments, cells (n = 12) were harvested only from the third and fourth turns of the cochlea; their lengths ranged between 60 and 75 pum and their diameter was uniform (8 to 9 pm). In other experiments, with cells not fully inserted into the microchamber (n = 206), the full range of cell lengths was used. We note that the all-pass nature of the response applies to cells of any length and for any degree of insertion. After removal of segments of the organ of Corti, cells were transferred after enzymatic incubation with Type IV collagenase (0.5 mg/mI; Sigma) to the experimental bath containing either Leibovitz's L-15 medium (Gibco) or Medium 199 (Gibco), supplemented with 15 mM Hepes and 5 mM bovine serum albumin (Sigma) and adjusted to 300 mosM (pH 7.35). Microchambers that held the cells were fabricated from borosilicate glass and had aperture diameters similar to those of OHCs. Cells were drawn into siliconized microchambers by gentle suction. Inserted cells were inspected at high magnification and discarded if there was any sign of induced trauma. All experiments were conducted at room temperature. Electrical command signals were generated from the low-impedance output of a waveform generator board in an IBM 486 clone and were delivered between the electrolytes surrounding and filling the microchamber. Making the fluid within the microchamber positive hyperpolarized the included membrane segment and depolarized the excluded membrane (11). Although the microchamber method did not permit us to measure it directly, the asymmetry of the electromotile response is indicative of the cell's resting potential. Cells that are likely to have relatively high resting potentials produce larger shortening than extension-directed responses. Conversely, depolarized cells generate either a symmetrical electromotile response or one with extension dominance (11) [J. Santos-Sacchi, J. Neurosci. 9, 2954 (1989)]. All cells in this study had pronounced contraction-directed response asymmetry and, by inference, high membrane potential. Pseudorandom noise is often used to identify the linear filter portion of a nonlinear physiological system [A. R. Maoler, Scand. J. Rehab. Med. Suppl. 3, 37 (1974); P. A. Marmarelis and V. Z. Marmarelis, Analysis of Physiological Sysfems (Plenum, New York, 1978)]. We used this signal in order to reduce data collection time. For our parameters, the 3-dB down point of the input was at 19,924 Hz. There are 161 spectral components within this bandwidth, and the voltage applied across the entire cell per spectral line was approximately ±0.6 mV. The extensive data obtained with ternary noise was confirmed in several cells by use of sinusoidal stimuli. The noise floor for the measurement shown in Fig. 2 was between 0.1 and 0.2 nm; thus, noise was clearly not a determinant of the high-frequency asymptote. The cell was imaged through a slit on a photodiode. Cell contraction and expansion modulated the light flux and the photocurrent. The entire stimulus-delivery and measuring apparatus was calibrated by use of the ternary signal input to illuminate the photodiode with a wide-band light-emitting diode and by measurement of its output. The resulting frequency response (corner frequency 18 kHz) of the entire system was used to correct all experimental data. System gain was calibrated for each experimental run by controlled displacement of the image of the cell in the slit. 25. We thank our colleagues M. A. Cheatham, B. Clark, S. Echteler, G. Emadi, D. Z. Z. He, M. Ruggero, J. Siegel, I. Sziklai, and S. Vranic-Sowers for their contributions and their comments on the manuscript. Supported by NIH and the American Hearing Research Foundation.