Marginal shell of the anteroventral cochlear nucleus: intensity coding in single units of the unanesthetized, decerebrate cat

ItI.WiH[ ELSEVIER

Neuroscience Letters 205(1996)71-74

1 ~ [ ~

Marginal shell of the anteroventral cochlear nucleus: intensity coding in single units of the unanesthetized, decerebrate cat S. Ghoshal, D.O. Kim* Division ¢~fOtolaryngology, Surgical Reearch Center, Department ~[ Surgery, Neuroscience Program, Bioengineering Program, University of Connecticut Health Center, Farmington, CT 06030-1110, USA

Received 29 November 1995; revised version received 10 January 1996; accepted 10 January 1996

Abstract

Single units were recorded in the marginal shell (38 units in 10 cats) and central core (62 units in 15 cats) of the anteroventral cochlear nucleus (AVCN) in unanesthetized decerebrate cats. The recording sites of the shell units were verified in reconstructed electrode tracks, and those of the core units were verified for 18 units and based on the recording depth for 44 units. There was a substantial presence of strongly driven units in the AVCN shell exhibiting non-saturating rate-level functions to pure tone, noise or both with dynamic ranges as wide as 89 dB. This finding supports a hypothesis that the AVCN shell may play a role in encoding acoustic stimulus intensity. The AVCN shell and core populations were different as follows. The shell population had more units which had wide dynamic ranges, low spontaneous rates (SRs) or were acoustically weakly or not driven than the core population. These differences were statistically significant (P < 0.001, Fisher's exact test). Keywords: Auditory; Olivocochlear; Brain stem; Reflex; Vestibular; Somatosensory; Sensory processing; Physiology

The marginal shell of the A V C N comprises the granule-cell layer (GCL), the', small cell cap (SCC) [3,7,17] and the 'external cell-poor rind', encapsulating its central core. The SCC is prominent in humans [16]. The marginal shell of the A V C N is anatomically distinct from its central core in several regards. The auditory nerve (AN) inputs to the SCC part of the shell are nearly exclusively from fibers having low spontaneous rates (SRs) [141 whereas those to the core are from both low- and high-SR fibers; the low- and high-SR AN fibers originate from the inner hair cells. Inputs from the following sources, i.e., collaterals of the efferent medial olivocochlear (MOC) neurons [5], A N fibers from the outer hair cells [4], somatosensory nuclei [11], primary vestibular afferents [6], and putative second-order vestibular neurons [23], terminate predominantly in the A V C N shell with little or weaker termination in the A V C N core. A limited description of A V C N shell units by Bourk [2] reported that they had discharge patterns different

* Corresponding author. Fax: +1 860 6792451; e-mail: kim @neuron.uchc.edu.

from those of the A V C N core. There have been no further physiological studies of the A V C N shell to our knowledge. One hypothesis regarding the function of the A V C N shell is that some of the neurons of the A V C N shell take part in the reflex circuit that includes the outer hair cells and MOC neurons by conveying accurate stimulus-intensity information to the MOC neurons [13]. The nearly exclusive low-SR afferent innervation and termination of the efferent MOC collaterals in the A V C N shell form a basis for this hypothesis. The goals of the present study are to characterize intensity-coding properties of single units in the A V C N shell and to compare them with those of the A V C N core units. Cats were initially sedated by ketamine and anesthetized by halothane gas. After decerebration, no further anesthetic was applied. The method for presentation of acoustic stimuli was described previously [12]. Micropipettes having resistances of 10-50 Mr2, filled with a marker, 10% biotinylated dextran amine (BDA) or 10% horseradish peroxidase (HRP), were used to mark iontophoretically the single-unit recording sites. The tissue sections were processed for HRP [15] and BDA [20]. The electrode tracks were reconstructed incorporating the

0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved PII: S0304-3940(96)12386-5

72

S. Ghoshal, D.O. Kim / Neuroscience Letters 205 (1996) 71-74

b

cat sec.#4Q6 6 ~

& Q

krfl Fig. 1. (a,c) Photomicrographsof marked sites of single-unit recording in AVCN marginal shell. Both sections were Nissl-stained. (b,d) Camera lucida drawings of the corresponding tissue sections. Each asterisk (*) corresponds to a marked site along the electrode track and the short line segments represent other single units recorded. Orientations: D, dorsal; L, lateral. Subdivisions: CG, cerebellar granule-cell layer; CM, cerebellar molecular layer; E, external cell-poor find; G, granule-cell layer; S, small cell cap (SCC). All other subdivisions are defined in Brawer et al. [3]. marked sites, recording depths of single units and tissue shrinkage. This study represents 38 shell units all of which were histologically verified to be in the AVCN marginal shell, either directly (16 units) or indirectly (22 units) in reconstructed electrode tracks using marked sites elsewhere along the track, and 62 AVCN core units consisting of 4 and 14 units directly and indirectly localized, respectively, and 44 units which were recorded at depths between 600 ktm below the dorsolateral surface and 300 ktm above the ventromedial border of the AVCN. Fig. la,c shows photomicrographs of marked sites (arrow heads). Fig. lb,d shows camera lucida drawings. The asterisk (*), corresponding to the marked site, and short line segments on each electrode track represent single units recorded. Both marked sites were clearly within the AVCN marginal shell. Fig. 2 shows rate-level functions for several AVCN units. The stimulus was pure tone at the characteristic

frequency (CF) or wideband noise, both applied at one burst (0.2 s) per second. The CF represents the stimulus frequency to which a unit is most sensitive. Driven rate, i.e., discharge rate to a stimulus minus the background rate, was normalized with respect to the maximum driven rate. A smoothing spline curve which fits the data points is shown in each panel. The dynamic range was defined to be the range of stimulus level where the rate increased monotonically. For saturated rate-level functions (e.g. Fig. 2u-x), the two ends of the dynamic range corresponded to the ordinate values of 0.05 and 0.95. When the function had a positive slope >_0.006/dB at high levels, e.g. Fig. 2f-j, the high end corresponded to the ordinate value of 1.0. When the function was non-monotonic, e.g. Fig. 2k-n, the dynamic range was measured to correspond to the widest monotonically increasing portion of the function. When the rate-level function had a flat portion in the middle with a slope of 0-0.006/dB over a range <25 dB, e.g. Fig. 2c,h,j, such a portion was included in the dynamic range. Fig. 2a-m represents AVCN shell units exhibiting the widest combined dynamic ranges to pure tone and noise. The site of the unit represented in Fig. 2i,j is shown in Fig. lc,d. The dynamic ranges of the shell units, as wide as 89 dB, were considerably wider than those of AVCN core units as shown in Fig. 2u-x for two typical core units. Fig. 2n-t shows examples of other types of ratelevel behaviors among AVCN shell units. Fig. 2n shows an example of a pronounced non-monotonic function of a shell unit to pure tone. The response of this unit to noise was monotonic and showed a wide dynamic range. The site of this unit is shown in Fig. la,b. Fig. 2q,r shows a strong response (670 spikes/s) to pure tone but a weak response (28 spikes/s) to noise, somewhat like a certain class of neurons in the dorsal cochlear nucleus called type II [22]. Fig. 2s,t shows a weak response (27 spikes/s) and, yet, a wide dynamic range (64 dB) to pure tone. To noise, this unit exhibited a high threshold and a steep slope reaching a moderate response (53 spikes/s). The shell and core units were divided into four subgroups based on the magnitude of maximum driven rate to pure tone and noise: strongly driven, >65 spikes/s; moderately driven, 30-65 spikes/s; weakly driven, 5 30 spikes/s; and not driven, <5 spikes/s. Properties of the acoustically weakly and not driven units of the AVCN shell were presented elsewhere [9]. The left column of Fig. 3 shows the proportions of the four subgroups among the shell and core populations. The two populations were different in that the core units were essentially all strongly driven whereas a substantial proportion of the shell units was weakly or not driven. The middle column shows proportions of units with wide dynamic range (>_50 dB) to pure tone or noise among the strongly-driven shell and core units. The distribution patterns among the two populations were approximately mirror images of each other. The right column shows distributions of SR among the

S. GhoshaL D.O. Kim / Neuroscience Letters 205 (1996) 71-74

73

AVCN shell units pure tone @ CF

S03-07 max

J

max. 96 s ~ / ' ~

,

101 spk/s

DR = 89 dB

DR = 74 dB

max. 332 S ~ L / ~ ~

/

_ y

DR = 85 dB

S04-17

b /

Q65-21 * max 253 s p k / ~ . , ~ n S

i

.

d

. IJ

DR =63dB

,

.... 125 _./" .---¢'T spk/s~¢"- -

fmax

DR = 85 dB

297 s p k / s ~ ~...~-'/DR = 62 dB

max 670 spks/s DR = 40 dB q

s

©

ax 93 s p k / s ~

g I f " "

DR = 87 dB

S16-03"

. ~

max

324 spk/s

j max

i J

DR = 75dB

._.._._/~'DR = 67 dB

/'

f

~ I

I

1 ]

I

0 20 40 60 80 100

AVCN core

units

max 253 ~ spks//

max 333 spk/s DR = 68 dB

/

t

DR = 64 dB

I

/'~ /

max 53 spk/s DR = 18 dB

z

-20 0 20 40 60 80 100

~

i ~

ax 266 s p k ~

max 28 spk/s DR = 32 dB

DR = 67 dB

2

/

/

Q65-27 max 27 spk/s

I

S04-12 max 79 spk/s

/

wideband noise

DR~--45 dB

Q62-09

S04-06 max 250 spk/s c

pure tone @ CF

wideband noise

DR = 41 dB

T-

j

--

DR = 3/ dB 4

S03-09

max 98 spk/s o c

0

/

k ~a,"'DR

# max

= 66 dB

~T

~ , . ~ . ~I D R

m I

2O 0 20 40 level

80 100 (dB SPL) 60

0

/ mo, __spk/s 218/f

141 s p k / s . f ~ = 50 dB p I I I--J 20 40 60 80 100

w

V

Y---~

i

DR = 34 dB i

I --

i

x I~.,/

I

I

20 0 20 40 60 80 100

0

1

DR = 45 dB

I

]-

i

T---

20 40 60 80 100

Fig. 2. Rate-level functions of nine AVCN shell units and two AVCN core units. Marked recording sites for two units with asterisks (*), S 16-03 and Q65-21, are shown in Fig. 1. The unit no., CF and SR are listed below for each unit: S03-07, 22.7 kHz, 11 spikes/s (s/s); S04-06, I,¢.7 kHz, 1.0 s/s; S04-17, 10.4 kHz, 0 s/s; S04-12, 15.6 kHz, 16 s/s; S16-03, 3.47 kHz, 4.0 s/s; S03-09, 22.7 kHz, 9.2 s/s; Q65-21, 8.33 kHz, 0 s/s; Q62-09, 0.69 kHz, 12 s/s; Q65-27, 5.95 kHz, 0 s/s; S11-03, 10.4 kHz, 12 s/s; Q52-19, 2.40 kHz, 18 s/s. The short vertical segments along each curve indicate two ends of the dynamic range (DR).

shell and core populations. The differences between the shell and core populations were statistically significant in all of the three comparisons of Fig. 3 as determined with Fisher's exact test [19] (P < 0.001). This study, to our knowledge, represents the first systematic attempt to obtain single-unit recording from the AVCN marginal shell. The main finding was that there was a substantial presence of strongly driven units in the AVCN shell exhibiting non-saturating rate-level functions with dynamic ranges as wide as 89 dB. This finding supports a hypothesis that the AVCN shell may play a role in encoding acoustic stimulus intensity. The observations that the SCC projects to the inferior colliculus [1] and superior olivary complex including medial olivocochlear (MOC) neurons [13] support a suggestion that intensity information in the SCC can be conveyed to these auditory centers. In particular, the SCC may participate in the reflex circuit involving the outer hair cells and MOC neurons [13]. Low-SR AN fibers are known to exhibit wider dynamic ranges than high-SR AN fibers, with a mean of 50 dB and 41 dB, respectively [8]. The present finding of the presence of mostly low-SR units with wide dynamic ranges in the AVCN shell is consistent with the above

report [8] and that of Liberman [14]. It has been hypothesized that the convergence of inputs from low-SR AN fibers and MOC collaterals should allow some neurons of the AVCN shell to exhibit wide dynamic ranges 7..

m a x . d r i v e n rate

dynamic range

A V C N shell

SD, A V C N shell

60 n

AV(?N shell

N =38

N=17

4(,1

N=38

A V C N core N = 62

Sn, AVCN core N = 60

(j;,,.,,...)

O.L ~

J ~.i.

/

~' (111 t

I

AVCIN c o r e

40 L

"~ = 1,2

,,,L£

,, ......... q)

50

.... I O0

1511

2Ill)

SR (spikes~see)

Fig. 3. Left: distributions of four subgroups of the maximum driven rate among shell and core units. Middle: proportions of units with wide dynamic range to pure tone, DR T, or to noise, DR N, >50 dB among strongly-driven shell and core units. Right: distributions of spontaneous; rate (SR) among shell and core units. The no. of units, N, in the upper middle is 17 rather than 20 (53% × 38) because data were insufficient to determine dynamic range for three units. SD, strongly dnven; MD, moderately driven; WD, weakly driven; ND, not driven.

74

S. Ghoshal, D.O. Kim/ Neuroscience Letters 205 (1996) 71-74

[5,13]. In this hypothesis, the MOC collateral input to certain shell units is excitatory, and this excitatory input to the shell units counteracts a suppression of the output of the cochlear amplifier exerted by the MOC neurons via the outer hair cells ([13], Fig. 1). Whether such a mechanism can account for the observed physiological characteristics of the AVCN shell units remains to be investigated. Some of the neurons in the VCN which show onset discharge patterns to short pure-tone bursts were reported to exhibit wide dynamic ranges, 70-80 dB ([10], Fig. 6; [18], Fig. 8; [21], Fig. 12). While two of these studies [18, 21] did not document the units' locations and, hence, their locations regarding shell versus core are not known, one of them [10] documented that the units were in the PVCN core. Regarding the discharge patterns to pure-tone bursts, the present sample of AVCN shell units with wide dynamic ranges included unusual, pause-build and onset discharge patterns (unpublished observations). Supported in part by grants from the NIDCD, NIH (No. R01-DC00360), and from Health Center Research Advisory Committee, University of Connecticut Health Center. We thank K. Parham and H.B. Zhao for help in recording data and discussions, S. Kuwada for discussions, D.K. Morest and D.L. Oliver for advice regarding anatomical aspects, and D.G. Machado for help in processing data. [1] Adams, J.C., Ascending projections to the inferior colliculus, J. Comp. Neurol., 183 (1979) 519-538. [2] Bourk, T.R., Electrical Responses of Neural Units in the Anteroventral Cochlear Nucleus, PhD Dissertation, Mass. Inst. Tech., Cambridge, MA, 1976. [3] Brawer, J.R., Morest, D.K. and Kane, E.C., The neuronal architecture of the cochlear nucleus of the cat, J. Comp. Neurol., 155 (1974) 251-300. [4] Brown, M.C., Berglund. A.M., Kiang, N.Y.S. and Ryugo, D.K., Central trajectories of type 11 spiral ganglion neurons, J. Comp. Neurol., 278 (1988) 581-590. [5] Brown, M.C., Liberman, M.C., Benson, T.E. and Ryugo, D.K., Brainstem branches from olivocochlear axons in cats and rodents, J. Comp. Neurol., 278 (1988) 591-603. [6] Burian, M. and Gstoettner, W., Projection of primary vestibular afferent fibers to the cochlear nucleus of the guinea pig, Neurosci. Lett., 84 (1988) 13-17. [7] Cant, N.B., The synaptic organization of the ventral cochlear nucleus of the cat: the peripheral cap of small cells. In M.A. Mer-

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15] [16] [17] [18]

[19] [20]

[21]

[22]

[23]

chain, J.M. Juiz, D.A. Godfrey and E. Mugnaini (Eds.), The Mammalian Cochlear Nuclei: Organization and Function, Plenum Press, New York, 1993, pp. 91-105. Evans, E.F. and Palmer, A.R., Relationship between the dynamic range of cochlear nerve fibers and their spontaneous activity, Exp. Brain Res., 40 (1980) 115-118. Ghoshal, S. and Kim, D.O., Marginal shell of the anteroventral cochlear nucleus: acoustically weakly-driven and not-driven units in the unanesthetized decerebrate cat, Acta Otolaryngol. (Stockholm), (1996) in press. Godfrey, D.A., Kiang, N.Y.S. and Norris, B.E., Single unit activity in the posteroventral cochlear nucleus of the cat. J. Comp. Neurol. 162 (1975) 247-268. Itoh, K., Kamiya, H., Mitani, A., Yasui, Y., Takada, M. and Mizuno, N., Direct projections from the dorsal column nuclei and the spinal trigeminal nuclei to the cochlear nucleus in the cat, Brain Res., 400 (1987) 145-150. Kim, D.O., Parham, K., Sirianni, J. and Chang, S.O., Spatial response profiles of posteroventral cochlear nucleus neurons and auditory-nerve fibers in unanesthetized decerebrate cats: response to pure tones, J. Acoust. Soc. Am., 89 (1991) 2804-2817. Kim, D.O., Parham, K., Zhao, H. and Ghoshal, S., The olivocochlear feedback gain control subsystem: ascending input from the small cell cap of the cochlear nucleus? In /~. Flock, D. Ottoson and M. Ulfendahl (Eds.), Active Hearing, Elsevier, New York, 1995, pp. 31-51. Liberman, M.C., Central projections of auditory nerve fibers of differing spontaneous rate, 1: anteroventral cochlear nucleus, J. Comp. Neurol., 313 (1991) 240-258. Mesulam, M., Tracing neural connections with horseradish peroxidase, Wiley, New York, 1982. Moore, J.K. and Osen, KK., The cochlear nuclei in man, Am. J. Anat., 154 (1979) 393-418. Osen, K.K., Cytoarchitecture of the cochlear nuclei of the cat, J. Comp. Neurol., 136 (1969) 453-484. Rhode, W.S. and Smith, P.H., Encoding timing and intensity in the ventral cochlear nucleus of the cat, J. Neurophysiol., 56 (1986) 261-286. Rosner, B. Fundamentals of Biostatistics, Duxbury Press, Boston, 1986, 332 pp. Veenman, C.L., Reiner, A. and Honig, M.G., Biotinylated dextran amine as an anterograde tracer for single- and double-labeling studies, J. Neurosci. Methods, 41 (1992) 239-254. Winter, I.M. and Palmer, A.R., Level dependence of cochlear nucleus onset units responses and facilitation by second tones or broadband noise, J. Neurophysiol., 73 (1995) 141-159. Young, E.D., Response characteristics of neurons of the cochlear nuclei, in C.1. Berlin (Ed.), Hearing Science, College-Hill, San Diego, 1984, pp. 423-460. Zhao, H.B., Parham, K., Ghoshal, S. and Kim, D.O., Small neurons in the vestibular nerve root project to the marginal shell of the anteroventral cochlear nucleus in the cat, Brain Res., 700 (1995) 295-298.