Notes on Yost, Chapter 8: Peripheral auditory nervous system and hair cells pp. 105-126.

11 February 2004

 

The goals of this chapter are (a) to describe the several voltage differences present in the inner ear that may be presumed to “fuel” neural energy and neurotransmission; (b) to describe the function of the inner and outer haircells as the transducer for mechanical to bioelectrical energy; and (c) to describe neural processing in the auditory nerve. Chapters 7 and 8 are then a unit, leading in to Chapter 9.

Cochlear potentials are present in the ear in the quiet (“resting potentials” - the “endolymphatic potential” and the internal “hair cell potential”) or are produced in the ear by sound waves (the “summating potential: SP”; the “cochlear microphonic: CM”; and the “action potential”, usually called the “compound action potential: the "CAP”). Yost describes the methods for measuring the potentials, most of which were reasonably understood in the 1950s. It is important to remember that like the dB scale, voltages are always measured with respect to some reference level. In the world outside the ear we usually measure voltages against “ground” or “earth”. In the ear the potential in one part is measured against that of another part of the ear or against some neutral place. Figure 8.1 comes from a famous paper by Halloway Davis, one of the greats of auditory science, and shows the several two-point places for measuring these potentials. Notice the use of 5 electrodes: a reference electrode on the neck; a wire electrode at A resting in the scala tympani; another wire electrode at B in the scala vestibuli; a micropipette filled with an ionized solution in the scala media; and another micropipette in the auditory nerve.

The electrical potential (DC) between the endolymph and the neck reference electrode of +80 millivolts (mV) is called for good reason the “endolymphatic potential”. Yost points out that this is the highest positive potential found anywhere in the organism, and it is believed to derive from metabolic activity in the stria vascularis. Hair cells, like most other neurons, have an internal potential of about -70 mV, but the stereocilia sit in endolymph and thus span an electrical potential of 150 mV, the largest electrical potential found in the body. Note that this is 0.15 volts, which is getting very close to the potential that we expect to find in working flashlight batteries (at least “close” if one is used to working in “orders of magnitude”). Damage to the stria vascularis reduces the endocochlear potential and almost completely eliminates the cochlear microphonic, and at least in the guinea pig is one of the consequences of age. Yost goes on to point to a third resting potential, which seems not to have a name, or at least, I do not know its name and Yost does not give it: it is a small potential of +2 to +5 mV between the scala vestibuli and the scala tympani: what it does, if anything, and where it originates is not known. Figure 8.2 gives these several resting potentials. Davis provided another picture, which showed that the Tunnel of Corti is filled with cortilymph which is like perilymph: endolymph is only present in the scale media beyond the reticular lamina, which is a boundary at the top of the hair cells, just below the tectorial membrane.

Point 2 describes the summating potential, a response to sound, which is recorded in the scala media compared to the reference electrode. It is a DC shift, but may be positive or negative. Yost says that it is not well understood, but Pickles attributes it to DC potentials in the hair cells. Note in Figure 8.3 there is a downward DC shift to the presence of a sound, which is the summating potential. The amplitude of the wave that follows the steady state of the tonal signal is the cochlear microphonic, and the several bumps at the beginning of the response wave define the compound action potential as it appears at the round window.

Point 3 is the more familiar cochlear microphonic, discovered by Wever and Bray in 1930. The CM matches the input in frequency and in amplitude for low frequencies, but as Yost says, it has a non-monotonic relationship to the input, at least at high levels (Figure 8.4). Unfortunately, Yost does not reference the source of 8.4, and I have not seen an original article talking about this effect. I presume it has something to do with distortion rather than “damage” as Yost suggests. The source of the CM is the hair cell itself, and this is shown by watching the voltage change in direction as the electrode passes from the scala tympani into the scala media. Yost puts it at the tip of the hair cell, when the stereocilia pass through the reticular lamina, and notes that in amplitude the CM matches the movement of the basilar membrane. The CM was used by Davis to follow the traveling wave along the baseline, with the informative results shown in Figure 8.5.

Point 4 is the compound action potential, which is derived from the eighth nerve. For a complex stimulus it has several components because it reflects the several volleys of different frequencies down the nerve, spread out over several milliseconds: but as these components are positive and negative, and occur at different times, they may sometimes cancel each other. As a result the CAP is less informative than it would seem to be. The CAP is more or less linear over a 100 dB range. Yost makes the point that as it can be measured in humans it can be used to provide clinical information about the state of the peripheral auditory system. However, it is now seldom used for this purpose, but the information is picked up either in the ABR or in OAE (Otoacoustic Emissions).

      The next topic is hair cells and stereocilia. Figure 8.6 is the familiar picture showing the shearing force across the hair cell provided by the bending of the basilar membrane. Yost points out that the geometry of the stereocilia (their curved arrangement in an arc) provides strength and resistance to breakage, and the linkages provided by cross bridges (seen in Figure 7.11) add to their strength. However, intense prolonged noise insults do disarrange the stereocilia and are responsible for temporary and permanent threshold shifts. Yost points out that measurements from hair cells taken in a dish show that the transmission of ions occurs on the cuticular plate (at the top of the hair cell, besides the stereocilia). [While I have heard that there is a tubule through the middle of the stereocilia.] Note that the outer hair cells vibrate “in sympathy” with imposed vibration at their resonant frequency (or with neurochemical application), which the inner hair cells do not. It is generally accepted that in their vibration they amplify the movement of the basilar membrane, but there are controversial differences in opinion in detail about how this works (see Figure 8.7).

The next section describes the “echoes” and “emissions” of the cochlea, some of which do not result from electrical-muscular events but are real reflected echoes, which others are actually produced in the inner ear by the mechanical vibrations of the hair cells. There are several emissions, sound induced and spontaneous (see Figure 8.8), that are believed to come from the outer hair cells, though Yost states that their source is not known for certain (he is probably being too cautious). The cubic difference tone (2f1 - f2) is seen in the acoustic emission just as it is along the basilar membrane. It is useful clinically because it means that a single input can test 3 sites along the basilar membrane (f1, f2 and the cubic difference tone), and also it shows that the echo is given by the outer hair cells, rather than being a passive echo. The spontaneous OAE is the cause of one kind of tinnitus which is relatively rare, about 5% of all cases of tinnitus (which is the hearing of sounds as if they are in the ear when there is no objective acoustic stimulus present). From a scientific standpoint the echoes are signs that the ear is a non-linear active system that adds energy at certain frequencies (and remember the non-linearity in the strength of displacement of the basilar membrane with increased level for best frequency input. Linearity holds for off-frequency tones, and for impaired ears).

The next section describes the auditory nerve, as we see in schematic view in Figures 8.9 and 8.11, more realistically in the electron micrograph of Figure 8.12. Obviously the anatomy of the afferent nerve is going to be critical in sensory function — if particular fibers were connected to all of the hair cells without any selectivity, then, Yost suggests, we would not be able to make fine tonal discriminations. This is probably true, but timing could be the key to specificity.

There are two kinds of afferent fibers: the radial fibers (R) are Type I and the outer spiral fibers (OS) are Type II. The thick (to be myelinated) radial fibers are about 90-95% of the fibers and go only to the inner hair cells, each to only one or perhaps two hair cells. One hair cell may receive about 10 to 20 fibers in different mammals (on average 8 in humans Yost says). The OS Type II fibers wander in a systematic way through the outer hair cells, innervating large groups of them. Figures 8.12 and 13 are other fantastic pictures of the cochlear anatomy showing the outer hair cells and the Type II ending.

The efferent fibers are called the “olivocochlear bundle”, less often nowadays, the Bundle of Oort, or of Rasmussen. The OCB derives from different parts of the olivary complex (the large set of nuclei where binaural information is first present, though I don’t think binaural information has any specialized connection to the efferent system). One part of the efferent system comes from the lateral nuclei and mostly does not cross the midline (the uncrossed OCB); the second part comes from the medial nuclei, and most (but not all) does cross the midline to the ear on the other side (the crossed OCB). The lateral (mostly uncrossed) goes to the inner hair cells, the medial (mostly crossed) goes to the outer hair cells (you can see in Figure 8.14 that some medial does not cross over and some lateral does cross over, about 10% of each). Following the implications of the anatomy and the presumed functions of inner vs. outer hair cells, one can imagine that the crossed medial fibers probably control outer hair cell motility, while the uncrossed lateral fibers probably control afferent input from the inner hair cells. I do not know of any writing concerning the possible benefit of this functional/structural relationship.

When these afferent and afferent fibers leave the basilar membrane they are initially unmyelinated until they reach the modiolus. There they are twisted together with the low frequency fibers in the center of the twist, the high frequency fibers on the outer layers. The point is always that there is an orderly anatomical arrangement by frequency along the basilar membrane and in the auditory nerve: this is preserved throughout the auditory system, and is called “tonotopic organization.”