The mammalian cochlea possesses two classes of afferent neurons and two classes of efferent neurons.

Transcription

1 1

2 The mammalian cochlea possesses two classes of afferent neurons and two classes of efferent neurons. Type I afferents contact single inner hair cells to provide acoustic analysis as we know it. Type II afferents branch extensively to contact numerous outer hair cells. Medial efferent neurons synapse on outer hair cells. Lateral efferent neurons synapse on Type I afferents beneath inner hair cells. 2

3 Extracellular recording from a single cochlear afferent axon in the VIIIth nerve. Upper panel, average firing rate to sound about 30 spikes/sec. Efferent shocks (black bar) reduces that activity. Lower panel, louder sound, efferent inhibition more potent. 3

4 Idealized vibration pattern of the basilar membrane (cochlear partition on which hair cells reside). Passive mechanism refers to the vibration pattern produced by a pure tone in a dead cochlea (or one without outer hair cells). Active mechanism refers to the pattern of vibration in a healthy, live cochlea, made 50 db more sensitive and much more sharply-tuned by the active mechanical contribution ( electromotility ) of outer hair cells. 4

5 5

6 6

7 Outer hair cells are mechanically active. In response to a sound-induced change in membrane potential, they move. This movement of the outer hair cells adds to that of the cochlear membranes, and so enhances the stimulus energy delivered to the inner hair cell. So, outer hair cell damage deafens the cochlea and is a common form of hearing loss. 7

8 Efferent neurons reside near the superior olivary complex. Medial olivocochlear neurons (MOCs) give rise to axons that synapse on outer hair cells. The axons of lateral olivocochlear neurons (LOCs) synapse onto Type I afferent dendrites beneath inner hair cells. The preponderance of MOCs innervate the contralateral ear, most LOCs are ipsilateral. MOCs can be activated by sound. One experimental method is to use sound in one ear to activate efferent inhibition of the opposite hear. The most common method to activate efferents experimentally is to shock their axons of passage in the floor of the IVth ventricle this mainly activates the larger diameter MOC axons. In the immature cochlea (before the onset of hearing in rodents up to about postnatal day 12) cholinergic efferents innervate IHCs directly. By hearing onset the adult pattern of innervation is found. This change is thought to be due to MOC efferents temporarily contacting IHCs before crossing the tunnel of Corti to find their permanent OHC partners. 8

9 Intracellular recording of inner hair cell membrane potential (high resistance sharp microelectrode). Tonal stimulus produces a plateau depolarization of several millivolts. During efferent (c.o.c.b. crossed olivocochlear bundle) stimulation the receptor potential is reduced in amplitude. Onset and offset of inhibition has time constants in range of 100 ms. 9

10 Average spike rate of a single auditory afferent neuron (extracellular recording in VIIIth nerve). As sound intensity is increased, spike rate increases along a sigmoidal line. When the same acoustic stimuli are combined with electrical shocks to the efferents, the response curve is shifted to higher levels. Sound at best frequency of the fiber. 10

11 Tuning curves of single auditory afferent neurons (VIIIth nerve recording). Inhibition de-tunes the afferent fiber. 11

12 Efferents to OHCs desensitize IHCs. This effect results from the mechanical contribution OHCs make to cochlear vibration. This cellular amplification is required for the sensitivity and sharp tuning of cochlear afferents. 12

13 Cochlear vibration pattern from Promenade round the Cochlea website. 13

14 Electromotility of OHCs enhances overall vibration of cochlear partition. 14

15 Laser Doppler velocimetry was used to measure basilar membrane tuning in control conditions (dotted line) and during electrical stimulation of the efferent pathway (solid line). (Dolan et al., 1997) Sound pressure and frequency of pure tones was varied to produce a criterion motion (velocity) of the basilar membrane. Louder sound was required to produce equivalent movement during efferent stimulation. So, efferent inhibition of OHCs really does change the vibration pattern of the basilar membrane. 15

16 The red-eared turtle, Pseudemys scripta elegans. Physiology of hair cells studied first in cold-blooded vertebrates. 16

17 Cold-blooded turtle tissues make wonderfully hardy ex vivo physiological preparations. Sound presented to the external ear of the half-head produces responses in cochlear afferents that match the known sensitivity and turning of the intact animal s ear. This preparation can last for many hours without losing sensitivity of tuning. 17

18 Closer view of the turtle s basilar papilla (auditory end-organ). The elliptical basilar membrane is ~ 1 mm in length, on which resides a stripe of ~ 1000 hair cells. The basilar papilla is tonotopically arranged, best frequencies of ~ 1000 Hz are found for afferent fiber responses on the right-hand end, and ~ 50 Hz on the left-hand end. 18

19 Extracellular recording from a single afferent in the turtle half-head prep. A tone at the best frequency produces action potentials at ~ 150 per second. Electrical shocks delivered to the efferent axons in the VIIIth nerve (a trick of the anatomy allows this without simultaneously activating afferent axons) suppresses the afferent response for ~ 100 ms. 19

20 Efferent inhibition de-tunes turtle acoustic afferents much like inhibition of mammalian cochlear afferents, at least for moderate levels of inhibition. Control tuning curve (open circles) shows sharp tuning of an afferent fiber in the turtle. During moderate inhibition (filled circles) the CF sensitivity, but not that at low frequencies, was reduced. With stronger inhibition (filled diamonds), suppression was evident at all frequencies. (Art and Fettiplace, 1984) 20

21 Intracellular recording from turtle auditory hair cells. During efferent shock train the membrane potential (shown here relative to the resting potential of -45 mv) is strongly hyperpolarized. The hyperpolarization outlasts the efferent shocks by more than 100 ms. IPSPs (inhibitory postsynaptic potentials) are cholinergic, and blocked by nicotinic antagonist (upper panel) and muscarinic antagonists. 21

22 Acoustic receptor potentials in turtle hair cells are sinusoidal changes in membrane potential that mimic the sound wave. A. During inhibition the membrane is hyperpolarized, and the receptor potential changes in amplitude depending on frequency. At the center or best frequency the receptor potential is made smaller, at higher frequencies the peak-to-peak sinusoid is unchanged, and at lower frequencies the receptor potential actually gets larger! B. Peak-to-peak receptor potential amplitude as a function of acoustic frequency in a different cell. Solid circles control, open circles during moderate inhibition, open squares during maximum hyperpolarization. Hair cell is converted into a low pass filter by strong inhibition. 22

23 Electrical circuit model of hair cell tuning and the possible effects of inhibition. The hair cell s electrical tuning can be modeled by an LRC circuit in which the inductance and capacitance exchange charge to produce resonance (and ringing in the time domain). The LRC circuit s tuning can reduced by shunt damping with a parallel resistor to produce curve c. The LRC circuit s tuning can reduced by series damping, an increase in resistance in series with the inductor, to produce curve b. Moderate inhibition produces shunt damping, simply adding a synaptic conductance in parallel. But, during strong inhibition, dominant voltagedependent potassium conductances of the hair cell membrane are turned off, producing a net increase in membrane resistance, so series damping. 23

24 Tuned systems, i.e., the turtle auditory hair cell (or a piano string to use a more familiar example), responds to transient stimuli by ringing. The frequency of that ring corresponds to the best frequency of the filter, and the duration of ringing is proportional to the sharpness of tuning. So, as in the example, a hair cell sharply-tuned to 371 Hz produces a slowly-decaying oscillation ( ringing ) at 371 Hz after a transient stimulus (an acoustic click). When that acoustic click is paired with efferent inhibition (second record), the sharpness of tuning drops (fewer cycles of oscillation). This is equivalent to the model curve c in the previous electrical model. Stronger inhibition (note the average hyperpolarization is greater in the third record) and even fewer oscillations are seen. If a still larger hyperpolarization were produced, no oscillations would occur curve b in the electrical model. 24

25 Efferent IPSP in turtle hair cell. Average change in membrane potential produced by a single efferent shock. Major component is a pronounced, 100 ms long hyperpolarization. However, a mysterious depolarizing blip precedes and appeared to be part of the synaptic response (was blocked by cholinergic antagonists as before). 25

26 Cellular physiology of cholinergic inhibition from chicken cochlear hair cells. 26

27 The basilar papilla in chickens is ~ 4mm in length and covered by ~10,000 hair cells. These are tonotopically organized, from ~ 50 Hz to ~ 5,000 Hz. Bird hair cells can be divided into tall (analogous to inner hair cells, possessing predominantly afferent innervation) and short (analogous to outer hair cells, possessing predominantly efferent innervation). 27

28 Efferent innervation in cross section of the chicken basilar papilla. Large, calyciform cholinergic endings found on short hair cells. 28

29 A. Brief application of ACh to a short hair cell produces a biphasic change in membrane potential as seen in turtle (boxed inset). B. Same cell in voltage clamp showing biphasic membrane current response to ACh. 29

30 The biphasic current results from sequential activation of two classes of ionic channels in the plasma membrane. The hair cell AChR is a ligand-gated cation channel that allows the entry of sodium and calcium. The cytoplasmic calcium rise activates nearby calcium-dependent potassium channels that produce the predominant outward current to hyperpolarize the hair cell. 30

31 Cytoplasmic BAPTA rapidly buffers free calcium and prevents the ACh-evoked outward current. This is a significant part of the proof that cholinergic inhibition of hair cells relies on calcium-activated potassium channels. 31

32 Adding 10 mm BAPTA to the hair cell cytoplasm prevents calcium concentration from rising enough to activate calcium-dependent potassium channels. Under these conditions the initial cation current through the hair cells nachr can be recorded in isolation. As shown here, this current reverses near 0 mv, confirming that hair cell AChRs are non-selective cation channels, similar in that respect to nicotinic AChRs of muscle and nerve. 32

33 Intracellular recording from mammalian outer hair cells show biphasic membrane currents in response to ACh released from efferent terminals. So, the two channel hypothesis applies to vertebrate hair cells from reptiles to mammals. 33

34 More examples of biphasic synaptic inhibition of mammalian cochlear hair cells (rats). 34

35 35

36 36

37 37

38 α9 alone, but not α10, can form functional channels (RNA injection in oocytes). But, α9 and α10 together form channels 100-fold more efficiently, and these have native calcium-dependence. (Elgoyhen) Best guess is that the native hair cell AChR is a pentamer with two alpha 9, 3 alpha

39 Identification of the constituent gene products enables transgenic experiments. The alpha subunits have been knocked out, overexpressed, and modified. 39

40 Belen Elgoyhen designed a mutation that should increase the mean open time of the channel. Threonine (hydrophilic) substitution for leucine that normally contributes to a hydrophobic girdle at the 9 position change favors open-state. 40

41 41

42 Spontaneous efferent synaptic currents (entirely inward because recorded negative to the potassium equilibrium potential). As predicted, cholinergic synaptic currents in the resulting transgenic mouse were longer-lasting than those of their wildtype littermates. Although amplitude was less on average, the synaptic area (equivalent to charge transfer) was greater overall at the transgenic synapses. 42

43 Electrically-evoked efferent synaptic currents. 43

44 44

45 Oto-acoustic emissions are ear sounds due to active motility of outer hair cells. They are an indication of normal cochlear function. Typically recorded as the distortion product (2f1-f2) produced by two tones (f1, f2) presented simultaneously to the ear. A sensitive microphone in the ear canal is the detector. 45

46 Efferent activity suppresses DPOAEs. In this example the normal efferentevoked suppression (electrical stimulation of efferents in the floor of the IVth ventricle) reduces the amplitude of the DPOAE by db. In the alpha10 knockout mouse in which hair cells no longer respond to ACh, this suppressive effect is lost entirely. 46

47 In the alpha 9 change of function mouse the efferent suppression of DPOAEs is enhanced and greatly prolonged compared to wildtype litter mates. 47

48 48

49 49

50 Efferent synaptic ultrastructure includes a synaptic cistern in the hair cell, coextensive with the efferent contact, and lying very close (~18 nm) to the postsynaptic membrane. Thought to be a calcium store that participates in efferent function. The story is incomplete. 50

51 51