Chapter 5

Notes (16 August 2008) 
 
P. 109: The antibodies used by Siemens et al. (2004) were directed against the cytoplasmic component of cadherin-23. Why therefore they should label the strand extracellularly is a mystery (see Gillespie et al., 2005, for further discussion). Secondly, the dimensions of cadherin-23 and the tip link do not match, unless it is supposed that the tip link is a complex of multiple cadherin-23 molecules. Thirdly, the tip link has to be able to extend to accommodate very large deflections of the hair bundle, while transmitting the effects of very small deflections to the mechanotransducer channels. At first sight, cadherin-23 does not appear to have the structure to allow this to occur; however, some molecular modelling by Sotomayor et al. (2005) provides a model of how this could happen (the molecule could unravel in stages, with Ca²⁺ ions providing temporary adhesion between the different folds of the molecule, hence transmitting tension through the molecule, even though it is much shorter than its full extension). (16 Aug 2008)

P. 117, section 5.3.2.4: Giga-ohm is a more generally used alternative to gigohm. (23 Aug 2008)

P. 127 and Fig. 5.15: The original figures of de Boer and Nuttall (1999) included the imaginary part of the impedance, not shown here for clarity. Their figures show that the imaginary part of the impedance was always negative within the part of the travelling wave shown in the Figure. Negative imaginary impedance shows that the impedance of the membrane was dominated by the stiffness rather than by the mass. This shows that the travelling wave comes to its peak and dies away before reaching the resonant point in the basilar membrane, in agreement with the description in Chapter 3 (Fig. 3.15 and accompanying text). (16 Aug 2008). 

P. 134: He et al. (2003) found that prestin knockout also leads to loss of stiffness of the outer hair cells. This may mean that motor effects arising elsewhere become uncoupled in some way, and so become unable to affect the dimensions of the whole cell. This would mean that prestin may not in fact be the motor protein. However, Zheng et al. (2000a) had also expressed prestin in a model cell system. They found that the cells showed nonlinear membrane capacitances, meaning that the membrane was changing under the influence of the electric field, and giving strong proof that the transfected protein, prestin, was indeed the motor protein. (16 Aug 2008).
 

P. 137: Because thermal noise makes the channel open and close stochastically (i.e. with a random component),there is no lower absolute threshold. Even very weak stimuli will cause some redistribution of the channel between the open and closed states. The actual threshold then depends on the degree to which the signal can be discriminated against the random background noise. 

Channel opening and closing with stimuli at intensities well under the apparent detection threshold explains a conundrum that is sometimes posed in relation to the active amplification and detection of near-threshold stimuli. If a stimulus is only just above its detection threshold at the peak of the travelling wave, its travelling wave after amplification will also only be just above the threshold for activating hair cells. How then does the amplification come about in the first place? - since before amplification, i.e. in the region of the cochlea when the wave is just entering the active region, the wave will be about 40 dB below threshold for activating hair cells The answer is that even a stimulus 40 dB below its apparent detection threshold will have some effect on the hair cells, and will still be amplified. The signal-to-noise ratio will be improved by the cochlear mechanics acting as a narrow band filter, so helping to reduce the amplification of thermal noise.(16 Aug 2008).
 

P. 149: Measurement of the degree of even order distortion faces technical difficulties, since for most realistic situations the difference between the primaries f₂ and f₁, and the difference tone f₂ – f₁ is large. Does one set the primaries or the difference tone to the CF of the region being measured? In either case, the cochlear mechanical filtering will have a substantial effect on the relative levels of the measured responses. A similar problem arises if one tries to measure the degree of even order nonlinearity by comparing the relative levels of the fundamental f₀ and its first harmonic 2f₀. For these reasons, little information is given in the text on the relative degree of even-order distortion. Cooper (1998) also points out a conundrum: from the levels of even order distortion that he measured from the harmonic 2f₀, he would have expected to see much greater d.c. shifts in the position of the basilar membrane, since for most realistic input-output functions with even order nonlinearity we would have expected both to be produced together. Is there therefore a feedback loop that continuously maintains the zero point of the basilar membrane? (see discussion on p. 144: see also Zou et al. (2006). (16 Aug 2008)

Zou, Y, et al., (2006) Cochlear transducer operating point adaptation. J Acoust Soc Am. 119:2232-2241.

P. 134: the question of whether a stimulus-evoked emission (in contrast to a spontaneous continuous oscillation as in Fig. 5.20) must always involve amplification, is a vexed one. For further discussions, see Allen and Fahey (1992), de Boer et al. (2005), and Ren and Nuttall (2006). (16 Aug 2008).

de Boer, E et al. (2005) The Allen-Fahey experiment extended. J Acoust Soc Am. 117:1260-1266.

Ren, T and Nuttall, AL, (2006) Cochlear compression wave: an implication of the Allen-Fahey experiment. J Acoust Soc Am. 119:1940-1942.

Return to An Introduction to the Introduction to the Physiology of Hearing webpage