- Email: [email protected]
Efferent tracts and cochlear frequency selectivity
277
Heating Research, 24 (1986) 277-283 Eisevier
HRR 00821
Efferent tracts and cochlear frequency selectivity Pierre Bonfils I,*, Marie-Claude Remcmd
1
and RCmy Pujol ’
’ CNRS-UA. 635, Laboraioire de Neurophysiologie, CoN1ge de France, II, Place Marcelin-Berthelot, ’ INSERM
75231 Paris Cedex OS, and - U. 254, Laboratoire de Neurobiologie de I%ludition, CHR Hbpital St. Charles, 34059 Montpeilier Cedex, France (Received 6 January 1986; accepted 25 May 1986)
The co&ear innervation of guinea pigs was sectioned medially in a rostrocaudal direction at the level of the floor of the fourth ventricle, to study the effects of efferent pathways on co&ear ~crophonic (CM) suppression, the compound action potential (CAP) masking phenomenon, the input-output CAP function, and co&ear frequency selectivity estimated with tuning curves of single auditory nerve fibers. Sectioning reduced CM suppression without having any effect on absolute CM amplitude; it also reduced CAP masking. 7he input-output CAP function was not changed at intensities below 75 dB, and tire single-unit tuning curves recorded before and after nerve sectioning were unaffected. Thus, the crossed efferent tracts (i.e., mainly the medial system) seems to be involved in the masking function itself, rather than one of the mechanisms responsible for high frequency co&ear selectivity. cochlea, cochlear efferent, frequency selectivity,
single-unit
tuning curve, suppression,
The existence of a crossed OlivoeochIear bundle (COCB) was first demo~s~at~ in cats, rats, and opossum (Rasmussen, 1946), then in birds, reptiles (Boord, 196I), and man (Gaeek, 1961). More recently, War, and Guinan (1978) used axonal transport techniques to distinguish between two different efferent systems: namely, a medial system, which predominantly innervates the contralateral outer hair cells (OHCs); and a lateral system, which operates mainly at the ipsilateral inner hair cell (IHC) level. From the results of these others studies (Strutz and Spatz, 1980; Strutz and Schmidt, 1982; Guinan et al., 1983,1984), the COCB can be considered to carry mostly fibers from the medial system, which terminates on the OHCs. The crossed efferent tracts have been considered to be naturally inhibitory, based upon the results of electrical stimulation on the floor of the fourth ventricle (Galambos, 1956; Fex, 1962; Correspondence should be addressed to: Dr. Pierre Bonfils at XNSERM-U. 254, Laboratoire de Neurobiologie de I’Audition, CHR HGpital St. Charles, 34059 Montpelher Cedex, France. 037%595.5/86/$03.50
mashing
Wiederhold, 1970; Wiederhold and Kiang, 1970; Gifford and Guinan, 1983). However, recently Mountain (1980) and Siegel and Rim (1982) proved that such st~ulation of the COCB affects cochlear micromechanics. Then, Brown et al. (1983) demonstrated its effect on the frequency tuning curves of IHCs. As electrical stimulation at the level of the brainstem sometimes produces facial twitches, it is not a very specific stimulus. Thus, it seems unjustified to attribute its effect only to the COCB. Our approach has been different, based upon medial sectioning of the cochlear innervation at the level of the floor of the fourth ventricle. We observed the effects of this sectioning on the eoehlear microphonic (CM) suppression eurves, on tone-on-tone masking of the compound action potential (CAP), on the ~put-output CAP function, and on cochlear frequency selectivity estimated using tuning curves of single auditory nerve fibers.
Pigmented guinea pigs, weighing 250 to 500 g each and free from middle-ear infection or their
0 1986 Elsevier Science Publishers B.V. (Biomedical
Division)
278
sequelae, were used. Two separate series of experiments were performed before and after sectioning at the level of the floor of the fourth ventricle. The sectioning was done in a rostrocaudal direction, between the midbrain and the spinal cord, strictly along the midline, ventrally limited by the skull base, without involving the basilar artery. In the first series of experiment, CM suppression and CAP masking were recorded. The animals (n = 23) were anesthetized by a neuroleptanesthesia technique (Evans, 1979). The bulla was opened and differential electrodes were placed in the basal turn of the cochlea (Tasaki et al., 1952). Acoustic stimuli were delivered in a closed system and sound pressure levels were measured at the tympanic membrane with a Briiel & Kjaer sound probe. CM suppression was measured using a tone-on-tone stimulus paradigm, while CM response to a continuous probe tone, set 10-15 dB above detection threshold, was monitored with a frequency analyzer (Tektronix 3L5). A second continuous tone (the masker) of variable frequency and intensity was added. The CM response to the probe tone was measured as a function of each frequency of the suppressing tone in lo-dB steps. From these functions iso-suppression and iso-intensity curves were derived. CAP recordings were made using the same pair of electrodes connected together. The probe stimulus was a tone pip set lo-20 dB above threshold. Its frequency (8-10 kHz) was chosen to correspond to the estimated frequency position of the differential electrodes. For simultaneous masking experiments, the second tone (the masker) was presented continuously, while under forward masking conditions the masker, a tone pip of 50-ms duration, was presented 5 ms before the sound probe (Dallos and Cheatham, 1976; Harris and Dallos, 1979). For each frequency of the masker, the Nl-Pl amplitude of the tone probe (averaged 128 times) was measured as a function of the masker level, in IO-dB steps. From these AP masking functions both iso-intensity and iso-suppression curves could be derived. In the second series of experiments, CAP input-output functions and single-unit frequency tuning curves were recorded. The animals (la = 17) were anesthetized by a single intraperitoneal injection of urethane (1600 mg/kg), supplemented by
additional doses when necessary. This kind of anesthetic is said to keep the spontaneous firing rates of efferents at a higher level than neuroleptanesthesia (Robertson, 1985). Curatization with gallamine (Flaxedyl”) prevented movement of the animal and permitted artificial ventilation. The left bulla was opened and an electrode was sealed near the round window. The opening of the bulla was quickly closed. For CAP input-output functions, the Nl-PI amplitude (averaged 128 times, repetition period 200 ms) was measured as a function of the intensity level. Access to the cochlear nerve was gained through an occipital cr~iotomy and aspiration of the cerebellum. The brainstem was gently pushed medially with small cotton balls introduced between the posterior cochlear nucleus and the temporal bone. The harmlessness of these actions was ensured by monitoring the CAP Nl magnitude at the round window. If Nl threshold varied more than 10 dB, the results have not been taken into account. Glass micropipettes, filled with 2.7 M KCl, were used to record from the auditory nerve fibers. Their resistance in situ varied between 5 and 20 MJ;2. Acoustic stimuli were delivered in a closed system (repetition period 200 ms), and sound pressure levels were measured at the tympanic membrane with a Brtiel & Kjaer sound probe. In all experiments, the temperature of the animal was monitored with a rectal probe and maintained close to 37.5”C with a heating pad. The anatomical site of sectioning was controlled in two ways. The site and depth of the lesion were verified by light microscopic examination after cryostat sectioning and Nissl staining. Second, the synaptic pole of OHCs was examined by electron microscopy: three weeks after sectioning a number of efferent endings had disappeared (see also Iurato et al., 1978). Results Tone-cm-tone masking experiments
CM iso-suppression curves recorded from the first turn of the cochlea showed a maximum of suppression around the best frequency of the recording site (Fig. 1). Sectioning the medial efferent system had no effect on CM absolute amphtude, but a clear alteration could be seen on CM
279
7 $j
80-
B
70-
Y J
60-
y
50-
u z
40 -
2 I
3o
o---o
before
r----o
after
sectw medial
section
CM
I
*
3
4
**~t~~
J
5678910
MASiER
15
FREQUENCY
&Hz)
Fig. 1. CM iso-suppression curves (constant 25 percent reduction of the probe tone), recorded from the first turn of the guinea pig co&lea. The two curves have been plotted before ) and after (------) midline sectioning of the ( efferent fibers in the floor of the fourth ventricle. Maximum suppression was obtained at frequencies corresponding to the recording sites of the electrodes. After sectioning, the suppression was reduced only in this region. Probe tone: 8.5 kHz, 40 dB SPL.
seen when the CM magnitude in response to a constant probe tone was plotted versus the masker level (Fig. 2). Again, if the frequency of the masker was lower than the frequency of the recording site, sectioning the efferent tracts had no effect at any intensity of the masker. However, when the masker was fixed at frequencies corresponding to the electrode position, there was reduction in the suppression: this reduction was the same at any masker level, and was equivalent to a lo-d3 decrease in masker intensity. AP threshold, for frequencies of 8-15 kHz, was not altered by sectioning of the medial efferent fibers, but sectioning did have an effect on the CAP masking phenomenon. CAP masking curves, recorded with simultaneously and forward masking, were plotted before and after medial sectioning. Under both masking conditions, the masking effect was reduced after sectioning by 5 to 30 dB (Fig. 3). Depending upon the animal, this effect was more pronounced with either simultaneous or forward masking (Fig. 4). In most cases, this reduction of the masking effect was more obvious at low levels of suppression.
suppression curves in the region of rn~rn~ suppression: the peak of suppression was reduced by about 10 dB as shown by dashed line in Fig. 1. On the low-frequency side of the curve, the degree of suppression was unchanged before and after sectioning. The reduction of suppression was more clearly
MASKER 40
50
60
10
80
MASKER
40
90 LEVEL
50 IdI3
60 SPL
70
80
FREOUENCY
(kliz)
90
1
Fig. 2. CM response to a constant probe tone (8.5 kHz, 40 dB SPL), plotted versus the intensity of a masking tone presented at 7 kHz (left) and 12 kHz (right). The data were obtained ) and after (- - - - - -) medial sectioning of before (----the efferent fibers. While there was no effect at 7 kHz, a marked reduction was notable at 12 kHz; this reduction was similar (around 10 dB) for all IeveIs of the masker.
Fig. 3. CAP masking curves recorded during simultaneous (left) and forward (right) masking. The masker was adjusted to produce a constant 25 percent reduction of the CAP response to a probe tone of 8 kHz, 30 dB SPL (100 percent amplitude: 1.5-2 mV). The curves shown were obtained before ) and after (------) sectioning of the crossed ( efferent fibers. Sectioning ted to a reduction of the masking effect that, in this case, was greater witb forward masking than simultaneous masking.
A
B
10
20
30
40
50
60
70
80
90
dB
INTENSITY
Fig. 4. Effect of increasing the intensity of a masking tone on the CAP response to a constant probe tone of 8 kHz, 30 dB SPL (A), or 10 kHz, 40 dB SPL (B). The responses were recorded during simultaneous (left) and forward (right) maskand after (- - - - - -) sectioning of the ing, before ( p) crossed efferent fibers. The degree of reduction in the effect of masking varied from 5 to 20 dB; depending upon the animal, this effect was more pronounced under one or the other masking condition.
) and Fig. 5. Input-output CAP function before (--after (- - - - - -) medial sectioning at the level of the brainstem. There was no change, either at threshold or for CAP values below 75 dB. Above 75 dB the CAP magnitude increased after sectioning.
CAP level function The CAP level function presented two slopes. The first one, for low intensities, was slow and decreased progressively to reach a plateau at 50 dB (8-10 kHz), 60 dB (12-14 kHz), and 65 dB
) and after (- - - - - -) medial sectioning of the medial efferent fibers in Fig. 6. Single auditory nerve tuning curves before (three guinea pigs (stimulus repetition period: 200 ms). There was no change at threshold, no effect on Q,,, and no effect on the tips of the curves.
281
(above 14 kHz). The second slope was steeper: the CAP magnitude increased very quickly and saturated at 75-85 dB SPL (averaged 128 times, repetition period 200 ms). Medial sectioning at the level of the floor of the fourth ventricle did not change the input-output function for intensities lower than 75 dB (10 kHz), whatever the frequency used. For levels above 75 dB, there was augmentation of the CAP amplitude after nerve sectioning (Fig. 5). Single-unit tuning curves It was decided to take into account only recordings in which the same fiber had been studied both before and after sectioning. Thus, a technical difficulty arose in the proximity of the recording and sectioning sites. In addition, sectioning was carried out only when a fiber with a characteristic frequency (CF) around 10 kHz was found, i.e., a fiber that originated from a co&ear region with a high degree of efferent innervation (Guinan et al., 1984). Three different units that met these criteria will now be discussed. Their spontaneous discharge rates varied between 20 and 40 spikes/s. Q,, values were between 5 and 7.3, in accordance with other data (Evans, 1975). As shown in Fig. 6, sectioning at the level of the floor of the fourth ventricle had no effect on the threshold at CF, or on the Q,,, or on the tips of the tuning curves. Discussion
Two different efferent systems have been identified coming from the brainstem to the cochlea in the cat (Warr and Guinan, 1978; Guinan et al., 1983, 1984), guinea pig (Strutz and Spatz, 1980) and chicken (Strutz and Schmidt, 1982). The medial system originates, in the cat, from the medial nucleus of the trapezoid body (NTB) and ends on OHCs. The lateral system comes from the lateral superior olivary nucleus (LSO). Medial sectioning along the midline, then, should deprive the cochlea of innervation from the COCB, including the crossed portions of the medial and lateral efferent systems. However, especially in the guinea pig (Robertson, 1985), the lateral system seems to have very few contralateral neurons; this makes medial sectioning almost selective for the medial
system. On the other hand, the real effect of electrical stimulation seems more difficult to appreciate considering the complexity of the auditory afferent and efferent brainstem pathways (Harrison and b-wing, 1964; Irwing and Harrison, 1967; Goldberg and Brown, 1968; Warr and Guinan, 1978). Yet, in 1970, Wiederhold and Kiang had doubts about the selectivity of such stimulation on the COCB. Thus, we feel that sectioning is an appropriate method of investigating the physiology of the medial efferent system in the guinea pig. The effects, reported in the literature, of medial sectioning at the level of the floor of the fourth ventricle have varied with the technique used. The majority of authors. found no effect with COCB sectioning: no change in absolute thresholds in the cat (Trahiotis, 1970; Igarashi et al., 1972); no alteration in the measurable perceptual signal-tonoise ratio 30, 50, and 70 dB above subjects’ pure tone thresholds (Igarashi et al., 1972); no change in psychoacoustic frequency discrimination (Igarashi et al., 1979a) or in psychoacoustic pure intensity discrimination (Igarashi et al., 1979b); no change in the CM or CAP magnitude in response to clicks in chinchilla chronically exposed (Iurato et al., 1978) and no alteration in threshold, amplitude, peak or interpeak latencies of waves I to IV of the auditory brainstem responses in the cat (O-U&i et al., 1982). Such negative results led some authors (Pfaltz, 1969) to say that the crossed efferent bundle does not function under physiological conditions. Nevertheless, in our experiments some physiological changes were detected after medial sectioning. First, CM suppression decreased in the region of normal maximum suppression, around the best frequency of the recording site (8-10 kHz): however, CM absolute magnitude did not change. In the same way, while the CAP thresholds were unchanged, the CAP masking effect was reduced after sectioning. The crossed olivocochlear efferent tracts seem to play a role in masking (Remond and Pujol, 1984). A similar reduction of the masking effect could also explain some previous results (Carlier and Pujol, 1982). One has to consider, in this respect, that the maximum effect in the forward masking paradigm happens when the masking tone precedes the stimulus tone by about 10
282
ms, a delay in agreement with the efferent latenties (Fex, 1962; Cody and Johnstone, 1982; Robertson, 1984). The cochleotopic specificity of this site of sectioning is, also, in agreement with the well-known frequency selectivity of the efferent fibers (Robertson, 1984; Robertson and Gummer, 1985). This fact is, perhaps, also relevant to Dewson’s data (1968): he found that monkeys, trained to discriminate between speech sounds presented at 70 dB SPL in 2400 Hz low-pass noise of varying intensity, were significantly impaired in their performances following surgical sectioning of the COCB. The magnitude of the deficit seemed to be related to the extent of destruction of the efferent cochlear fibers. A second, albeit small, effect of medial sectioning was found in the input-output CAP function for levels higher than 75 dB. At these high levels of stimulation the CAP magnitude increased after sectioning. This may be related to some results reported by Handrock and Zeisberg (1982): in guinea pigs both ITS and PTS are significantly increased after severing of the vestibular nerve. From this, it seems that the crossed efferent tract (perhaps its small lateral component) could have a protective function above 75 dB. The finding of no change in the CAP input-output function below 75 dB, i.e., at a level where the slope reflects the gradual recruitment of afferents near the CF, leads us to speculate that the COCB plays no role in cochlear frequency selectivity. Indeed, medial sectioning has no effect on cochlear frequency selectivity as determined from evaluation of three single-unit tuning curves recorded before and after sectioning. In conclusion, our results indicate that sectioning along the midline decreases the masking phenomenon, while not affecting the single-unit tuning curves. Thus, the COCB seems to be involved in the masking function itself, rather than in mechanisms leading to high-frequency selectivity. Therefore, only techniques that involve recording from single units should be used in studies trying to determine the role of efferent pathways in co&ear frequency selectivity. Acknowledgements This research was funded in part by a grant to Pierre Bonfils from the Fondation pour la Re-
cherche Medicale Franqaise. The authors are greatly indebted to Dr. J.-P. Legouix, Director of the CNRS Laboratory at College de France, Paris, in which the work was carried out. Thanks are due to C. Saulnier and M. Lasek for technical assistance, and to A. Bara for editing the manuscript. References Boord, R.L. (1961): The efferent co&ear bundle in the caiman and the pigeon. Exp. Neural. 3, 225-239. Brown, M.C., Nuttal, A. and Masta, RI. (1983): Intracellular recordings from cochlear inner hair cells: effects of stimulation of the crossed olivocochlear efferents. Science 222, 69-71. Carlier, E. and Pujol, R. (1982): Sectioning the efferent bundle decreases co&ear frequency selectivity. Neurosci. Lett. 28, 101-106. Cody, A.R. and Johnstone, B.M. (1982): Acoustic-evoked activity of single efferent neurons in the guinea pig cochlea. J. Acoust. Sot. Am. 12, 280-282. Dallos, P. and Cheatham, M.A. (1976): Compound action potential tuning curves. J. Acoust. Sot. Am. 59, 591-597. Dewson, J.H. (1968): Efferent olivocochlear bundle: some relationship to stimulus discrimination in noise. J. Neurophysiol. 31, 122-130. Evans, E.F. (1975): Cochlear nerve and cochlear nucleus, In: Handbook of Sensory Physiology, vol. V, part 2, pp. l-108. Editors: W.D. Keidel and W.D. Neff. Springer-Verlag, Berlin, Heidelberg, New York. Evans, E.F. (1979): Neuroleptanesthesia for the guinea pig. Arch. Otolaryngol. 105, 185-186. Fex, J. (1962): Auditory activity in centrifugal and centripetal cochlear fibres in the cat. Acta Oto-Laryngol. (Stockh.) 55, Suppl. 189. Gacek, R.R. (1961): The efferent cochlear bundle in man. Arch. Otolaryngol. 74, 680-694. Galambos, R. (1956): Suppression of auditory nerve activity by stimulation of efferent fibers to cochlea. J. Neurophysiol. 19, 424-437. Gifford, M.G. and Guinan J.J., Jr. (1983): Effects of crossed olivocochlear bundle stimulation on cat auditory nerve fiber response to tones. J. Acoust. Sot. Am. 74, 115-123. Goldberg, J.M. and Brown, P.B. (1968): Functional organisation of the dog superior olivary complex: an anatomical and electrophysiological study. J. Neurophysiol. 31, 639-656. Guinan J.J., Jr., Warr, W.B. and Norris, B.E. (1983): Differential olivocochlear projections from lateral versus medial zones of the superior olivary complex. J. Comp. Neurol. 221, 358-370. Guinan J.J., Jr., Warr, W.B. and Norris, B.E. (1984): Topographic organisation of the olivocochlear projections from the lateral and medial zones of the superior olivary complex. J. Comp. Neurol. 226, 21-27. Handrock, M. and Zeisberg, J. (1982): The influence of the efferent system on adaptation, temporary and permanent threshold shift. Arch. Otol. Rhino]. Laryngol. 234,191-195.
283 Harris, D.M. and Dallos, P. (1979): Forward masking of auditory nerve fiber responses. J. Neurophysiol. 42, 1083-1107. Harrison, J.M. and Irwing, R. (1964): Nucleus of the trapezoid body: dual afferent innervation. Science 143, 473-474. Igarashi, M., Cranford, J.L., Nakai, Y. and Alford, B.R. (1972): Behavioral auditory function after transection of crossed olivocochlear bundle in the cat. I. Pure tone threshold and perceptual signal-to-noise ratio. Acta Oto-Laryngol. @to&h.) 73,455-466. Igarashi, M., Cranford, J.L., Nakai, Y. and Alford, B.R. (1979a): Behavioral auditory function after transection of crossed olivocochlear bundle in the cat. IV. Study on pure tone frequency discrimination. Acta Oto-Laryngol. (Stockh.) 87,79-83. Igarashi, M., Cranford, J.L., Allen, E. and Alford, B.R. (1979b): Behavioral auditory function after transection of crossed olivocochlear bundle in the cat. V. Pure tone intensity discrimination. Acta Oto-Laryngol. (Stockh.) 87, 429-433. Irwing, R. and Harrison, J.M. (1967): The superior olivary complex and audition: a comparative study. J. Comp. Neurol. 130,77-86. Iurato, S., Smith, C.A., Eldredge, D.H., et al. (1978): Distribution of the crossed olivocochlear bundle in the chinchilla’s cochlea. J. Comp. Neurol. 182, 57-76. Mountain, D.C. (1980): Changes in endol~phatic potential and crossed olivocochlear bundle stimulation alter cochlear mechanics. Science 210,71-72. O-U&i, T., Igarashi, M. and Kulecz, W.B. (1982): Transection of crossed olivocochlear bundle and auditory brainstem responses in the cat. Acta Oto-Laryngol. (Stockh.) 94, 1-6. Pfaltz, R.K.J. (1969): Absence of function for the crossed olivocochlear bundle under physiological conditions. Arch. Klin. Exp. Ohren-Nasen Kehlkopfheilkd. 193, 89-100. Rasmussen, G. (1946): The olivary peduncle and other fibrous projections of the superior olivary complex. J. Comp. Neurol. 84, 141-219. Remond, M.-C. and Pujol, R. (1984): CM suppression and CAP masking after sectioning the medial efferent system.
Abstract XXIO, Workshop on Inner Ear Biology, Taormina, Italy, p. 46. Robertson, D. (1984): Horseradish peroxidase injection of physiologica.lly characterized afferent and efferent neurons in the guinea pig spiral ganglion. Hear. Res. 15, 113-121. Robertson, D. (1985): Brainstem location of efferent neurons projecting to the guinea pig cochlea. Hear. Res. 20, 79-84. Robertson, D. and Gummer, M. (1985): Physiological and morphological characterization of efferent neurons in the guinea pig cochlea. Hear. Res. 20,63-78. Siegel, J.H. and Kim, D.O. (1982): Efferent neural control of cochlear mechanics. Olivocochlear bundle stimulation affects co&ear biomech~c~ no~ne~ty. Hear. Res. 6, 171-182. Strutz, J. and Schmidt, CL. (1982): Acoustic and vestibular efferent neurons in the chicken. A horseradish peroxidase study. Acta Oto-Laryngol. (Stockh.) 94, 45-51. Strutz, J. and Spatz, W. (1980): Superior olivary and extraolivary origin of centrifugal innervation of the cochlea in guinea pig. A horseradish peroxidase study. Neurosci. Lett. 17, 227. Tasaki, I., Davis, H. and Legouix, J.-P. (1952): The space-time pattern of the cochlear microphonics (guinea pig) as recorded by differential electrodes. J. Acoust. Sot. Am. 24, 502-519. Trahiotis, C. (1970): Behavioral investigation of some possible effects of sectioning the crossed olivocochlear bundle. J. Acoust. Sot. Am. 47, 592-596. Warr, W.B. and Guinan, J.J., Jr. (1978): Efferent innervation of the organ of Corti: two separate systems. Brain Res. 173, 152-15.5. Wiederhold, M.L. (1970): Variations of the effects of electric stimulation of the crossed olivocochlear bundle on cat single auditory nerve fiber responses to tone burst. J. Acoust. Sot. Am. 48, 966-977. Wiederhold, M.L. and Kiang, N.Y.S. (1970): Effects of electric stimulation of the crossed olivocochlear bundle on single auditory nerve fibers in the cat. J. Acoust. Sot. Am. 48, 950-965.
Heating Research, 24 (1986) 277-283 Eisevier
HRR 00821
Efferent tracts and cochlear frequency selectivity Pierre Bonfils I,*, Marie-Claude Remcmd
1
and RCmy Pujol ’
’ CNRS-UA. 635, Laboraioire de Neurophysiologie, CoN1ge de France, II, Place Marcelin-Berthelot, ’ INSERM
75231 Paris Cedex OS, and - U. 254, Laboratoire de Neurobiologie de I%ludition, CHR Hbpital St. Charles, 34059 Montpeilier Cedex, France (Received 6 January 1986; accepted 25 May 1986)
The co&ear innervation of guinea pigs was sectioned medially in a rostrocaudal direction at the level of the floor of the fourth ventricle, to study the effects of efferent pathways on co&ear ~crophonic (CM) suppression, the compound action potential (CAP) masking phenomenon, the input-output CAP function, and co&ear frequency selectivity estimated with tuning curves of single auditory nerve fibers. Sectioning reduced CM suppression without having any effect on absolute CM amplitude; it also reduced CAP masking. 7he input-output CAP function was not changed at intensities below 75 dB, and tire single-unit tuning curves recorded before and after nerve sectioning were unaffected. Thus, the crossed efferent tracts (i.e., mainly the medial system) seems to be involved in the masking function itself, rather than one of the mechanisms responsible for high frequency co&ear selectivity. cochlea, cochlear efferent, frequency selectivity,
single-unit
tuning curve, suppression,
The existence of a crossed OlivoeochIear bundle (COCB) was first demo~s~at~ in cats, rats, and opossum (Rasmussen, 1946), then in birds, reptiles (Boord, 196I), and man (Gaeek, 1961). More recently, War, and Guinan (1978) used axonal transport techniques to distinguish between two different efferent systems: namely, a medial system, which predominantly innervates the contralateral outer hair cells (OHCs); and a lateral system, which operates mainly at the ipsilateral inner hair cell (IHC) level. From the results of these others studies (Strutz and Spatz, 1980; Strutz and Schmidt, 1982; Guinan et al., 1983,1984), the COCB can be considered to carry mostly fibers from the medial system, which terminates on the OHCs. The crossed efferent tracts have been considered to be naturally inhibitory, based upon the results of electrical stimulation on the floor of the fourth ventricle (Galambos, 1956; Fex, 1962; Correspondence should be addressed to: Dr. Pierre Bonfils at XNSERM-U. 254, Laboratoire de Neurobiologie de I’Audition, CHR HGpital St. Charles, 34059 Montpelher Cedex, France. 037%595.5/86/$03.50
mashing
Wiederhold, 1970; Wiederhold and Kiang, 1970; Gifford and Guinan, 1983). However, recently Mountain (1980) and Siegel and Rim (1982) proved that such st~ulation of the COCB affects cochlear micromechanics. Then, Brown et al. (1983) demonstrated its effect on the frequency tuning curves of IHCs. As electrical stimulation at the level of the brainstem sometimes produces facial twitches, it is not a very specific stimulus. Thus, it seems unjustified to attribute its effect only to the COCB. Our approach has been different, based upon medial sectioning of the cochlear innervation at the level of the floor of the fourth ventricle. We observed the effects of this sectioning on the eoehlear microphonic (CM) suppression eurves, on tone-on-tone masking of the compound action potential (CAP), on the ~put-output CAP function, and on cochlear frequency selectivity estimated using tuning curves of single auditory nerve fibers.
Pigmented guinea pigs, weighing 250 to 500 g each and free from middle-ear infection or their
0 1986 Elsevier Science Publishers B.V. (Biomedical
Division)
278
sequelae, were used. Two separate series of experiments were performed before and after sectioning at the level of the floor of the fourth ventricle. The sectioning was done in a rostrocaudal direction, between the midbrain and the spinal cord, strictly along the midline, ventrally limited by the skull base, without involving the basilar artery. In the first series of experiment, CM suppression and CAP masking were recorded. The animals (n = 23) were anesthetized by a neuroleptanesthesia technique (Evans, 1979). The bulla was opened and differential electrodes were placed in the basal turn of the cochlea (Tasaki et al., 1952). Acoustic stimuli were delivered in a closed system and sound pressure levels were measured at the tympanic membrane with a Briiel & Kjaer sound probe. CM suppression was measured using a tone-on-tone stimulus paradigm, while CM response to a continuous probe tone, set 10-15 dB above detection threshold, was monitored with a frequency analyzer (Tektronix 3L5). A second continuous tone (the masker) of variable frequency and intensity was added. The CM response to the probe tone was measured as a function of each frequency of the suppressing tone in lo-dB steps. From these functions iso-suppression and iso-intensity curves were derived. CAP recordings were made using the same pair of electrodes connected together. The probe stimulus was a tone pip set lo-20 dB above threshold. Its frequency (8-10 kHz) was chosen to correspond to the estimated frequency position of the differential electrodes. For simultaneous masking experiments, the second tone (the masker) was presented continuously, while under forward masking conditions the masker, a tone pip of 50-ms duration, was presented 5 ms before the sound probe (Dallos and Cheatham, 1976; Harris and Dallos, 1979). For each frequency of the masker, the Nl-Pl amplitude of the tone probe (averaged 128 times) was measured as a function of the masker level, in IO-dB steps. From these AP masking functions both iso-intensity and iso-suppression curves could be derived. In the second series of experiments, CAP input-output functions and single-unit frequency tuning curves were recorded. The animals (la = 17) were anesthetized by a single intraperitoneal injection of urethane (1600 mg/kg), supplemented by
additional doses when necessary. This kind of anesthetic is said to keep the spontaneous firing rates of efferents at a higher level than neuroleptanesthesia (Robertson, 1985). Curatization with gallamine (Flaxedyl”) prevented movement of the animal and permitted artificial ventilation. The left bulla was opened and an electrode was sealed near the round window. The opening of the bulla was quickly closed. For CAP input-output functions, the Nl-PI amplitude (averaged 128 times, repetition period 200 ms) was measured as a function of the intensity level. Access to the cochlear nerve was gained through an occipital cr~iotomy and aspiration of the cerebellum. The brainstem was gently pushed medially with small cotton balls introduced between the posterior cochlear nucleus and the temporal bone. The harmlessness of these actions was ensured by monitoring the CAP Nl magnitude at the round window. If Nl threshold varied more than 10 dB, the results have not been taken into account. Glass micropipettes, filled with 2.7 M KCl, were used to record from the auditory nerve fibers. Their resistance in situ varied between 5 and 20 MJ;2. Acoustic stimuli were delivered in a closed system (repetition period 200 ms), and sound pressure levels were measured at the tympanic membrane with a Brtiel & Kjaer sound probe. In all experiments, the temperature of the animal was monitored with a rectal probe and maintained close to 37.5”C with a heating pad. The anatomical site of sectioning was controlled in two ways. The site and depth of the lesion were verified by light microscopic examination after cryostat sectioning and Nissl staining. Second, the synaptic pole of OHCs was examined by electron microscopy: three weeks after sectioning a number of efferent endings had disappeared (see also Iurato et al., 1978). Results Tone-cm-tone masking experiments
CM iso-suppression curves recorded from the first turn of the cochlea showed a maximum of suppression around the best frequency of the recording site (Fig. 1). Sectioning the medial efferent system had no effect on CM absolute amphtude, but a clear alteration could be seen on CM
279
7 $j
80-
B
70-
Y J
60-
y
50-
u z
40 -
2 I
3o
o---o
before
r----o
after
sectw medial
section
CM
I
*
3
4
**~t~~
J
5678910
MASiER
15
FREQUENCY
&Hz)
Fig. 1. CM iso-suppression curves (constant 25 percent reduction of the probe tone), recorded from the first turn of the guinea pig co&lea. The two curves have been plotted before ) and after (------) midline sectioning of the ( efferent fibers in the floor of the fourth ventricle. Maximum suppression was obtained at frequencies corresponding to the recording sites of the electrodes. After sectioning, the suppression was reduced only in this region. Probe tone: 8.5 kHz, 40 dB SPL.
seen when the CM magnitude in response to a constant probe tone was plotted versus the masker level (Fig. 2). Again, if the frequency of the masker was lower than the frequency of the recording site, sectioning the efferent tracts had no effect at any intensity of the masker. However, when the masker was fixed at frequencies corresponding to the electrode position, there was reduction in the suppression: this reduction was the same at any masker level, and was equivalent to a lo-d3 decrease in masker intensity. AP threshold, for frequencies of 8-15 kHz, was not altered by sectioning of the medial efferent fibers, but sectioning did have an effect on the CAP masking phenomenon. CAP masking curves, recorded with simultaneously and forward masking, were plotted before and after medial sectioning. Under both masking conditions, the masking effect was reduced after sectioning by 5 to 30 dB (Fig. 3). Depending upon the animal, this effect was more pronounced with either simultaneous or forward masking (Fig. 4). In most cases, this reduction of the masking effect was more obvious at low levels of suppression.
suppression curves in the region of rn~rn~ suppression: the peak of suppression was reduced by about 10 dB as shown by dashed line in Fig. 1. On the low-frequency side of the curve, the degree of suppression was unchanged before and after sectioning. The reduction of suppression was more clearly
MASKER 40
50
60
10
80
MASKER
40
90 LEVEL
50 IdI3
60 SPL
70
80
FREOUENCY
(kliz)
90
1
Fig. 2. CM response to a constant probe tone (8.5 kHz, 40 dB SPL), plotted versus the intensity of a masking tone presented at 7 kHz (left) and 12 kHz (right). The data were obtained ) and after (- - - - - -) medial sectioning of before (----the efferent fibers. While there was no effect at 7 kHz, a marked reduction was notable at 12 kHz; this reduction was similar (around 10 dB) for all IeveIs of the masker.
Fig. 3. CAP masking curves recorded during simultaneous (left) and forward (right) masking. The masker was adjusted to produce a constant 25 percent reduction of the CAP response to a probe tone of 8 kHz, 30 dB SPL (100 percent amplitude: 1.5-2 mV). The curves shown were obtained before ) and after (------) sectioning of the crossed ( efferent fibers. Sectioning ted to a reduction of the masking effect that, in this case, was greater witb forward masking than simultaneous masking.
A
B
10
20
30
40
50
60
70
80
90
dB
INTENSITY
Fig. 4. Effect of increasing the intensity of a masking tone on the CAP response to a constant probe tone of 8 kHz, 30 dB SPL (A), or 10 kHz, 40 dB SPL (B). The responses were recorded during simultaneous (left) and forward (right) maskand after (- - - - - -) sectioning of the ing, before ( p) crossed efferent fibers. The degree of reduction in the effect of masking varied from 5 to 20 dB; depending upon the animal, this effect was more pronounced under one or the other masking condition.
) and Fig. 5. Input-output CAP function before (--after (- - - - - -) medial sectioning at the level of the brainstem. There was no change, either at threshold or for CAP values below 75 dB. Above 75 dB the CAP magnitude increased after sectioning.
CAP level function The CAP level function presented two slopes. The first one, for low intensities, was slow and decreased progressively to reach a plateau at 50 dB (8-10 kHz), 60 dB (12-14 kHz), and 65 dB
) and after (- - - - - -) medial sectioning of the medial efferent fibers in Fig. 6. Single auditory nerve tuning curves before (three guinea pigs (stimulus repetition period: 200 ms). There was no change at threshold, no effect on Q,,, and no effect on the tips of the curves.
281
(above 14 kHz). The second slope was steeper: the CAP magnitude increased very quickly and saturated at 75-85 dB SPL (averaged 128 times, repetition period 200 ms). Medial sectioning at the level of the floor of the fourth ventricle did not change the input-output function for intensities lower than 75 dB (10 kHz), whatever the frequency used. For levels above 75 dB, there was augmentation of the CAP amplitude after nerve sectioning (Fig. 5). Single-unit tuning curves It was decided to take into account only recordings in which the same fiber had been studied both before and after sectioning. Thus, a technical difficulty arose in the proximity of the recording and sectioning sites. In addition, sectioning was carried out only when a fiber with a characteristic frequency (CF) around 10 kHz was found, i.e., a fiber that originated from a co&ear region with a high degree of efferent innervation (Guinan et al., 1984). Three different units that met these criteria will now be discussed. Their spontaneous discharge rates varied between 20 and 40 spikes/s. Q,, values were between 5 and 7.3, in accordance with other data (Evans, 1975). As shown in Fig. 6, sectioning at the level of the floor of the fourth ventricle had no effect on the threshold at CF, or on the Q,,, or on the tips of the tuning curves. Discussion
Two different efferent systems have been identified coming from the brainstem to the cochlea in the cat (Warr and Guinan, 1978; Guinan et al., 1983, 1984), guinea pig (Strutz and Spatz, 1980) and chicken (Strutz and Schmidt, 1982). The medial system originates, in the cat, from the medial nucleus of the trapezoid body (NTB) and ends on OHCs. The lateral system comes from the lateral superior olivary nucleus (LSO). Medial sectioning along the midline, then, should deprive the cochlea of innervation from the COCB, including the crossed portions of the medial and lateral efferent systems. However, especially in the guinea pig (Robertson, 1985), the lateral system seems to have very few contralateral neurons; this makes medial sectioning almost selective for the medial
system. On the other hand, the real effect of electrical stimulation seems more difficult to appreciate considering the complexity of the auditory afferent and efferent brainstem pathways (Harrison and b-wing, 1964; Irwing and Harrison, 1967; Goldberg and Brown, 1968; Warr and Guinan, 1978). Yet, in 1970, Wiederhold and Kiang had doubts about the selectivity of such stimulation on the COCB. Thus, we feel that sectioning is an appropriate method of investigating the physiology of the medial efferent system in the guinea pig. The effects, reported in the literature, of medial sectioning at the level of the floor of the fourth ventricle have varied with the technique used. The majority of authors. found no effect with COCB sectioning: no change in absolute thresholds in the cat (Trahiotis, 1970; Igarashi et al., 1972); no alteration in the measurable perceptual signal-tonoise ratio 30, 50, and 70 dB above subjects’ pure tone thresholds (Igarashi et al., 1972); no change in psychoacoustic frequency discrimination (Igarashi et al., 1979a) or in psychoacoustic pure intensity discrimination (Igarashi et al., 1979b); no change in the CM or CAP magnitude in response to clicks in chinchilla chronically exposed (Iurato et al., 1978) and no alteration in threshold, amplitude, peak or interpeak latencies of waves I to IV of the auditory brainstem responses in the cat (O-U&i et al., 1982). Such negative results led some authors (Pfaltz, 1969) to say that the crossed efferent bundle does not function under physiological conditions. Nevertheless, in our experiments some physiological changes were detected after medial sectioning. First, CM suppression decreased in the region of normal maximum suppression, around the best frequency of the recording site (8-10 kHz): however, CM absolute magnitude did not change. In the same way, while the CAP thresholds were unchanged, the CAP masking effect was reduced after sectioning. The crossed olivocochlear efferent tracts seem to play a role in masking (Remond and Pujol, 1984). A similar reduction of the masking effect could also explain some previous results (Carlier and Pujol, 1982). One has to consider, in this respect, that the maximum effect in the forward masking paradigm happens when the masking tone precedes the stimulus tone by about 10
282
ms, a delay in agreement with the efferent latenties (Fex, 1962; Cody and Johnstone, 1982; Robertson, 1984). The cochleotopic specificity of this site of sectioning is, also, in agreement with the well-known frequency selectivity of the efferent fibers (Robertson, 1984; Robertson and Gummer, 1985). This fact is, perhaps, also relevant to Dewson’s data (1968): he found that monkeys, trained to discriminate between speech sounds presented at 70 dB SPL in 2400 Hz low-pass noise of varying intensity, were significantly impaired in their performances following surgical sectioning of the COCB. The magnitude of the deficit seemed to be related to the extent of destruction of the efferent cochlear fibers. A second, albeit small, effect of medial sectioning was found in the input-output CAP function for levels higher than 75 dB. At these high levels of stimulation the CAP magnitude increased after sectioning. This may be related to some results reported by Handrock and Zeisberg (1982): in guinea pigs both ITS and PTS are significantly increased after severing of the vestibular nerve. From this, it seems that the crossed efferent tract (perhaps its small lateral component) could have a protective function above 75 dB. The finding of no change in the CAP input-output function below 75 dB, i.e., at a level where the slope reflects the gradual recruitment of afferents near the CF, leads us to speculate that the COCB plays no role in cochlear frequency selectivity. Indeed, medial sectioning has no effect on cochlear frequency selectivity as determined from evaluation of three single-unit tuning curves recorded before and after sectioning. In conclusion, our results indicate that sectioning along the midline decreases the masking phenomenon, while not affecting the single-unit tuning curves. Thus, the COCB seems to be involved in the masking function itself, rather than in mechanisms leading to high-frequency selectivity. Therefore, only techniques that involve recording from single units should be used in studies trying to determine the role of efferent pathways in co&ear frequency selectivity. Acknowledgements This research was funded in part by a grant to Pierre Bonfils from the Fondation pour la Re-
cherche Medicale Franqaise. The authors are greatly indebted to Dr. J.-P. Legouix, Director of the CNRS Laboratory at College de France, Paris, in which the work was carried out. Thanks are due to C. Saulnier and M. Lasek for technical assistance, and to A. Bara for editing the manuscript. References Boord, R.L. (1961): The efferent co&ear bundle in the caiman and the pigeon. Exp. Neural. 3, 225-239. Brown, M.C., Nuttal, A. and Masta, RI. (1983): Intracellular recordings from cochlear inner hair cells: effects of stimulation of the crossed olivocochlear efferents. Science 222, 69-71. Carlier, E. and Pujol, R. (1982): Sectioning the efferent bundle decreases co&ear frequency selectivity. Neurosci. Lett. 28, 101-106. Cody, A.R. and Johnstone, B.M. (1982): Acoustic-evoked activity of single efferent neurons in the guinea pig cochlea. J. Acoust. Sot. Am. 12, 280-282. Dallos, P. and Cheatham, M.A. (1976): Compound action potential tuning curves. J. Acoust. Sot. Am. 59, 591-597. Dewson, J.H. (1968): Efferent olivocochlear bundle: some relationship to stimulus discrimination in noise. J. Neurophysiol. 31, 122-130. Evans, E.F. (1975): Cochlear nerve and cochlear nucleus, In: Handbook of Sensory Physiology, vol. V, part 2, pp. l-108. Editors: W.D. Keidel and W.D. Neff. Springer-Verlag, Berlin, Heidelberg, New York. Evans, E.F. (1979): Neuroleptanesthesia for the guinea pig. Arch. Otolaryngol. 105, 185-186. Fex, J. (1962): Auditory activity in centrifugal and centripetal cochlear fibres in the cat. Acta Oto-Laryngol. (Stockh.) 55, Suppl. 189. Gacek, R.R. (1961): The efferent cochlear bundle in man. Arch. Otolaryngol. 74, 680-694. Galambos, R. (1956): Suppression of auditory nerve activity by stimulation of efferent fibers to cochlea. J. Neurophysiol. 19, 424-437. Gifford, M.G. and Guinan J.J., Jr. (1983): Effects of crossed olivocochlear bundle stimulation on cat auditory nerve fiber response to tones. J. Acoust. Sot. Am. 74, 115-123. Goldberg, J.M. and Brown, P.B. (1968): Functional organisation of the dog superior olivary complex: an anatomical and electrophysiological study. J. Neurophysiol. 31, 639-656. Guinan J.J., Jr., Warr, W.B. and Norris, B.E. (1983): Differential olivocochlear projections from lateral versus medial zones of the superior olivary complex. J. Comp. Neurol. 221, 358-370. Guinan J.J., Jr., Warr, W.B. and Norris, B.E. (1984): Topographic organisation of the olivocochlear projections from the lateral and medial zones of the superior olivary complex. J. Comp. Neurol. 226, 21-27. Handrock, M. and Zeisberg, J. (1982): The influence of the efferent system on adaptation, temporary and permanent threshold shift. Arch. Otol. Rhino]. Laryngol. 234,191-195.
283 Harris, D.M. and Dallos, P. (1979): Forward masking of auditory nerve fiber responses. J. Neurophysiol. 42, 1083-1107. Harrison, J.M. and Irwing, R. (1964): Nucleus of the trapezoid body: dual afferent innervation. Science 143, 473-474. Igarashi, M., Cranford, J.L., Nakai, Y. and Alford, B.R. (1972): Behavioral auditory function after transection of crossed olivocochlear bundle in the cat. I. Pure tone threshold and perceptual signal-to-noise ratio. Acta Oto-Laryngol. @to&h.) 73,455-466. Igarashi, M., Cranford, J.L., Nakai, Y. and Alford, B.R. (1979a): Behavioral auditory function after transection of crossed olivocochlear bundle in the cat. IV. Study on pure tone frequency discrimination. Acta Oto-Laryngol. (Stockh.) 87,79-83. Igarashi, M., Cranford, J.L., Allen, E. and Alford, B.R. (1979b): Behavioral auditory function after transection of crossed olivocochlear bundle in the cat. V. Pure tone intensity discrimination. Acta Oto-Laryngol. (Stockh.) 87, 429-433. Irwing, R. and Harrison, J.M. (1967): The superior olivary complex and audition: a comparative study. J. Comp. Neurol. 130,77-86. Iurato, S., Smith, C.A., Eldredge, D.H., et al. (1978): Distribution of the crossed olivocochlear bundle in the chinchilla’s cochlea. J. Comp. Neurol. 182, 57-76. Mountain, D.C. (1980): Changes in endol~phatic potential and crossed olivocochlear bundle stimulation alter cochlear mechanics. Science 210,71-72. O-U&i, T., Igarashi, M. and Kulecz, W.B. (1982): Transection of crossed olivocochlear bundle and auditory brainstem responses in the cat. Acta Oto-Laryngol. (Stockh.) 94, 1-6. Pfaltz, R.K.J. (1969): Absence of function for the crossed olivocochlear bundle under physiological conditions. Arch. Klin. Exp. Ohren-Nasen Kehlkopfheilkd. 193, 89-100. Rasmussen, G. (1946): The olivary peduncle and other fibrous projections of the superior olivary complex. J. Comp. Neurol. 84, 141-219. Remond, M.-C. and Pujol, R. (1984): CM suppression and CAP masking after sectioning the medial efferent system.
Abstract XXIO, Workshop on Inner Ear Biology, Taormina, Italy, p. 46. Robertson, D. (1984): Horseradish peroxidase injection of physiologica.lly characterized afferent and efferent neurons in the guinea pig spiral ganglion. Hear. Res. 15, 113-121. Robertson, D. (1985): Brainstem location of efferent neurons projecting to the guinea pig cochlea. Hear. Res. 20, 79-84. Robertson, D. and Gummer, M. (1985): Physiological and morphological characterization of efferent neurons in the guinea pig cochlea. Hear. Res. 20,63-78. Siegel, J.H. and Kim, D.O. (1982): Efferent neural control of cochlear mechanics. Olivocochlear bundle stimulation affects co&ear biomech~c~ no~ne~ty. Hear. Res. 6, 171-182. Strutz, J. and Schmidt, CL. (1982): Acoustic and vestibular efferent neurons in the chicken. A horseradish peroxidase study. Acta Oto-Laryngol. (Stockh.) 94, 45-51. Strutz, J. and Spatz, W. (1980): Superior olivary and extraolivary origin of centrifugal innervation of the cochlea in guinea pig. A horseradish peroxidase study. Neurosci. Lett. 17, 227. Tasaki, I., Davis, H. and Legouix, J.-P. (1952): The space-time pattern of the cochlear microphonics (guinea pig) as recorded by differential electrodes. J. Acoust. Sot. Am. 24, 502-519. Trahiotis, C. (1970): Behavioral investigation of some possible effects of sectioning the crossed olivocochlear bundle. J. Acoust. Sot. Am. 47, 592-596. Warr, W.B. and Guinan, J.J., Jr. (1978): Efferent innervation of the organ of Corti: two separate systems. Brain Res. 173, 152-15.5. Wiederhold, M.L. (1970): Variations of the effects of electric stimulation of the crossed olivocochlear bundle on cat single auditory nerve fiber responses to tone burst. J. Acoust. Sot. Am. 48, 966-977. Wiederhold, M.L. and Kiang, N.Y.S. (1970): Effects of electric stimulation of the crossed olivocochlear bundle on single auditory nerve fibers in the cat. J. Acoust. Sot. Am. 48, 950-965.