Download Corticofugal Amplification of Subcortical Responses to Single Tone

RAPID COMMUNICATION
Corticofugal Amplification of Subcortical Responses to Single Tone
Stimuli in the Mustached Bat
YUNFENG ZHANG AND NOBUO SUGA
Department of Biology, Washington University, St. Louis, Missouri 63130
Zhang, Yunfeng and Nobuo Suga. Corticofugal amplification of
subcortical responses to single tone stimuli in the mustached bat.
J. Neurophysiol. 78: 3489–3492, 1997. Since 1962, physiological
data of corticofugal effects on subcortical auditory neurons have
been controversial: inhibitory, excitatory, or both. An inhibitory
effect has been much more frequently observed than an excitatory
effect. Recent studies performed with an improved experimental
design indicate that corticofugal system mediates a highly focused
positive feedback to physiologically ‘‘matched’’ subcortical neurons, and widespread lateral inhibition to ‘‘unmatched’’ subcortical
neurons, in order to adjust and improve information processing.
These results lead to a question: what happens to subcortical auditory responses when the corticofugal system, including matched
and unmatched cortical neurons, is functionally eliminated? We
temporarily inactivated both matched and unmatched neurons in
the primary auditory cortex of the mustached bat with muscimol
(an agonist of inhibitory synaptic transmitter) and measured the
effect of cortical inactivation on subcortical auditory responses.
Cortical inactivation reduced auditory responses in the medial geniculate body and the inferior colliculus. This reduction was larger
(60 vs. 34%) and faster (11 vs. 31 min) for thalamic neurons than
for collicular neurons. Our data indicate that the corticofugal system amplifies collicular auditory responses by 1.5 times and thalamic responses by 2.5 times on average. The data are consistant
with a scheme in which positive feedback from the auditory cortex
is modulated by inhibition that may mostly take place in the cortex.
INTRODUCTION
Neurons in the deep layers of the auditory cortex (AC)
project to the medial geniculate body (MGB) or to the inferior colliculus (IC) (Huffman and Henson 1990; Saldana
et al. 1996). The corticofugal projections are tonotopically
organized (Andersen et al. 1980; Herbert et al. 1991). Physiological data of corticofugal effects on the MGB and IC
neurons have been controversial: inhibitory (Amato et al.
1969; Massopust and Ordy 1962; Sun et al. 1996; Watanabe
et al. 1966), excitatory (Andersen et al. 1972; Villa et al.
1991), or both (Ryugo and Weinberger 1976; Sun et al.
1989; Syka and Popelar 1984). These studies did not reveal
the functional role of the corticofugal projections because
of limitations in experimental design.
The recent findings made in the AC of the mustached bat,
Pteronotus parnellii parnellii, indicate that cortical neurons
mediate a highly focused positive feedback, incorporated
with widespread lateral inhibition, via corticofugal projections. Therefore an inhibitory or excitatory corticofugal effect depends on a topographic relationship between cortical
and subcortical neurons (Yan and Suga 1996; Zhang et al.
1997). In cats, He (1997) found corticofugal effects similar
to the above.
However, Sun et al. (1996) reported that in the big brown
bat, Eptesicus fuscus, the corticofugal pathway modulates
auditory responses of collicular neurons only by inhibition. If
they are correct, a large cortical inactivation should eliminate
inhibition in the IC and should increase collicular auditory
responses. However, if the excitation due to positive feedback, found in the mustached bat, is larger than lateral inhibition, a large cortical inactivation should decrease the auditory responses of subcortical neurons. The aim of the present
study is to measure the total amount of change produced by
the corticofugal system by entirely inactivating the corticofugal fibers originating from a particular subdivision of the
primary auditory cortex, called the Doppler-shifted constantfrequency (DSCF) processing area.
The DSCF area is Ç1.4 mm diam and consists of neurons
sharply tuned to sounds between 60.6 and 62.3 kHz (Suga
and Manabe 1982; Suga et al. 1987). For frequencies between 61.0 and 61.5 kHz, best frequency shifts at a rate of
Ç66 Hz per cortical minicolumn, which is Ç20 mm wide.
DSCF neurons tuned to a particular frequency augment the
auditory responses of thalamic and collicular neurons (hereafter, subcortical neurons) tuned to the same frequency (not
different by ú0.2 kHz), and reduce the responses of subcortical neurons tuned to other frequencies (different by ú0.2
kHz). This means that single subcortical neurons receive
positive feedback from one or a few cortical minicolumns,
and receive lateral inhibition from many, perhaps, all other
minicolumns in the DSCF area (Zhang et al. 1997). We
found that inactivation of both matched and unmatched cortical DSCF neurons evokes a prominent decrease in subcortical auditory responses.
METHODS
Materials, surgery, recording of neural activity, acoustic stimulation, and data acquisition were basically the same as those described elsewhere (Suga et al. 1983). The essential portions of
these are summarized below. Four adult mustached bats (Pteronotus parnellii parnellii) were used for the present experiments. Under neuroleptanalgesia (Innovar 4.08 mg/kg body wt), the dorsal
surface of the bat’s skull was exposed, and a 1.8-cm-long metal
post was glued onto the skull. Four days after the surgery, the
unanesthetized bat was placed in a styrofoam restraint suspended
by an elastic band at the center of a soundproof, echo-attenuated
room maintained at 30–327C. The head was immobilized by fixing
the post on the skull to a metal rod with set screws, and adjusted
directly at a condenser loudspeaker located 74 cm away. ‘‘DSCF’’
neurons are tuned to 60.6 to Ç62.3 kHz sounds and are clustered
in the DSCF area of the AC, the ventral division of the MGB, and
the dorsoposterior division of the IC. To record their auditory
0022-3077/97 $5.00 Copyright q 1997 The American Physiological Society
/ 9k20$$no65
11-12-97 00:43:32
neupa
LP-Neurophys
3489
3490
Y. ZHANG AND N. SUGA
responses, a tungsten-wire microelectrode was inserted into one of
these structures through holes of Ç50 mm diam made in the skull.
DSCF neurons were identified by their best frequencies (BFs) and
locations in the AC, MGB, and IC. A window discriminator was
used to isolate action potentials of single neurons.
The acoustic stimuli were 23-ms-long tone bursts with a 0.5ms rise-decay time. These were generated by a voltage-controlled
oscillator and an electronic switch and were delivered at a rate of
5/s. The frequency of a tone burst was varied manually or by a
computer. Collicular, thalamic, and cortical DSCF neurons were
tuned to particular frequencies (BFs), and particular amplitudes
(best amplitudes). The amplitudes of the tone bursts of a computerized frequency (F) scan were set at the best amplitude of a given
neuron, which was usually Ç30 dB above minimum threshold.
The F scan consisted of 21 time blocks, in 20 blocks of which
frequency was changed in 0.10-kHz steps. In the 21st (last) block,
no stimulus was presented in order to count background discharges.
The duration of each block was 200 ms, so that the duration of the
F scan was 4,200 ms. The F scan was used to obtain a frequencyresponse curve (Fig. 3). To measure the time course of a change in
subcortical auditory responses evoked by cortical inactivation (Fig.
2), a modified F scan was used in which the frequency of tone bursts
was fixed at the BF of the subcortical auditory neuron under study.
After mapping the DSCF area, a hole of 0.2–0.3 mm diam was
made in the skull at the center of the DSCF area, and a well was
placed over the hole. A few days later, single subcortical DSCF
neurons were recorded, and 0.2 mg muscimol (1.0 mg/ ml saline)
was applied to the DSCF area with a 1.0-ml Hamilton microsyringe
to inactivate the DSCF area. Gelfoam (gelatin sponge) was placed
in the well to prevent leakage of muscimol and to increase the
contact time with the cortical surface. To study the effect of cortical
inactivation on the auditory responses of subcortical neurons, the
responses of single subcortical neurons to tone bursts in the F or
modified F scan were recorded with a computer before, during,
and after the cortical inactivation. The F or modified F scan was
delivered once every 5 s during data acquisition.
Off-line data processing included plotting peristimulus time
(PST) and PST cumulative (PSTC) histograms displaying the responses to 50 identical acoustic stimuli. Frequency-response curves
were based on the responses to 50 identical F scans. The magnitude
of auditory responses was expressed by the number of impulses
per 50 identical stimulus blocks after subtracting background discharges counted in the last block of the F or modified F scan
repeated 50 times. The t-test was used to test the significance of
the difference between the auditory responses obtained before and
after muscimol application, and between the responses of thalamic
and collicular neurons.
the five thalamic (B) and six collicular neurons (D). After
muscimol application, the reduction in auditory response for
the thalamic neurons began simultaneously with that of the
collicular neurons. However, the reduction in responses developed faster [11 { 4.2 (SD) min vs. 31 { 15 min for 1/2
of the maximum decrease] and was larger (60 { 35% vs.
34 { 7.8% at the maximum decrease) for the thalamic neurons than for the collicular neurons. This result indicates
that, normally, auditory responses are amplified by the corticocollicular projection and are further amplified by corticothalamic projection. The latency and duration of the plateau
reduction (plateau reduction is defined as the reduction to
within 10% above the maximum reduction) and the latency
of 50% recovery were almost the same for the thalamic and
collicular neurons.
Background discharge rates ranged between 1.0 and 4.2/
s (3.3 { 0.87/s, n Å 5) for the thalamic neurons and between
1.4 and 3.9/s (3.2 { 0.75/s, n Å 6) for the collicular neurons. Muscimol application evoked a large reduction in background discharges of all thalamic neurons. The mean and
standard deviation of percent reduction are 75 { 9.3%, n Å
5 (Fig. 2, A and B). Muscimol application evoked no sig-
RESULTS
The effect of inactivation of the DSCF area with muscimol
was studied for five thalamic and six collicular DSCF neurons. The sample size was small because of the difficulty
in obtaining a long-term recording of single-unit activity.
Nevertheless, the results were consistent. Cortical inactivation reduced the auditory responses of every subcortical neuron studied and evoked no change in its BF. Figure 1 shows
such reduction in the auditory responses of a thalamic (A)
and a collicular neuron (B). The amount of the reduction
was larger for the thalamic neurons than for the collicular
neurons. Both the initial and later portions of the response
were reduced by approximately the same amount. Figure 2
shows the time courses of muscimol’s effect on auditory
responses and background discharges of a thalamic ( A) and
a collicular neuron (C), and the average time courses for
/ 9k20$$no65
FIG . 1. Decrease in the auditory responses of a thalamic (A) and a
collicular neuron (B) evoked by a large cortical inactivation, including
‘‘matched’’ cortical neurons, with muscimol. Each peristimulus-time (PST)
histogram displays the responses of a single neuron to 50 identical stimuli.
These responses were recorded before (a: control condition), during (b:
muscimol), and after (c: recovery condition) the inactivation. The acoustic
stimulus (tone burst) was a 60.98 kHz, 45 dB SPL for A and a 61.08 kHz,
47 dB SPL for B. Horizontal bars below the histograms indicate 23-mslong acoustic stimuli. The PST histograms in a–c are also shown by the
PST cumulative (PSTC) histograms in d. MGB, medial geniculate body;
IC, inferior colliculus.
11-12-97 00:43:32
neupa
LP-Neurophys
CORTICOFUGAL AMPLIFICATION OF AUDITORY RESPONSES
3491
FIG . 2. Changes in the auditory responses
( s ) and background discharges ( ● ) of thalamic (A and B) and collicular (C and D) neurons evoked by a large cortical inactivation,
including matched cortical neurons, with muscimol. A and C: data obtained from single
neurons. B and D: average of data obtained
from 5 and 6 neurons, respectively. In B and
D, each symbol and a bar, respectively, represent a mean and a standard deviation. Auditory responses are expressed by the total number of impulses discharged for 50 presentations of an identical tone burst. (Background
discharges were subtracted from the auditory
responses.) Background discharges are expessed by the total number of impulses discharged in 50 200-ms-long time blocks (i.e.,
the number of impulses per 10 s), in which
no acoustic stimuli were delivered. Open and
filled triangles, respectively, represent the auditory responses and background discharges
in the control condition. Arrows indicate the
time when muscimol was applied to the auditory cortex.
nificant reduction in three of the six collicular neurons (Fig.
2C), but a significant reduction in the remaining three (35 {
22.3%, n Å 3, P õ 0.05). The mean reduction for all six
collicular neurons was insignificant (Fig. 2D).
The reduction in auditory response by muscimol applied
to the DSCF area was larger and faster for thalamic neurons
than for collicular neurons. Therefore one may consider a
possibility that muscimol applied to the AC diffused to the
MGB and IC. Muscimol applied to the FM-FM area, which
is located immediately dorsoanterior to the DSCF area, reduced the auditory responses of thalamic FM-FM neurons
(J. Yan and N. Suga, unpublished observations), but did not
at all reduce the responses of three thalamic and one collicular DSCF neuron tested. Muscimol applied to the DSCF
area had no effect on thalamic FM-FM neurons, which were
located immediately dorsomedial to thalamic DSCF neurons.
These control experiments indicate that the effect of muscimol described in the present paper was due to the inactivation of the corticofugal system originating from the DSCF
area. Also, the fact that reduction in responses of thalamic
and collicular neurons began simultaneously argues in favor
of a corticofugal effect, rather than a diffusion of muscimol
to each subcortical nucleus.
The absolute amount of reduction by muscimol was largest
for the responses to the tone burst at the BF of a given subcortical DSCF neuron, but the reduction expressed as a percentage
was similar to all the responses to tone bursts at the BF and at
other frequencies (Fig. 3). In the present study, all frequencyresponse curves were obtained at the best amplitude to excite
a given subcortical neuron that was, on the average, 29 { 6.7
dB above minimum threshold for the five thalamic neurons,
and 30 { 5.4 dB above minimum threshold for the six collicular neurons. The 50% bandwidths of the frequency-response
curves showed no change ú0.1 kHz for all subcortical neurons
studied, except for one collicular neuron.
/ 9k20$$no65
DISCUSSION
The dose of muscimol we applied to the DSCF area is
known to evoke a prominent temporary deficit in frequency
discrimination within a frequency range of 60.6 and 62.3
kHz, but no deficit in echo-delay (time interval) discrimination (Riquimaroux et al. 1991). Therefore this dose of muscimol presumably inactivated the whole DSCF area, and we
infer that without the corticofugal system, responses to single
tone bursts would be Ç60 and 34% less than normal at the
MGB and IC, respectively.
A reduction of auditory responses evoked by a focal inactivation of cortical DSCF neurons with 90 nl of 1.0% lidocaine is 54 { 15 % (n Å 4) for matched thalamic neurons
and 21 { 8.9 % (n Å 4) for matched collicular neurons
(Zhang et al. 1997). We expected that the reduction of
auditory responses evoked by the large cortical inactivation
with muscimol would be larger than these. The reduction
was indeed larger for collicular neurons [34 { 7.8% (n Å
6) vs. 21 { 8.9% (n Å 4); P õ 0.05], but not for thalamic
neurons [60 { 35% (n Å 5) vs. 54 { 15% (n Å 4); P ú
0.1]. It remains to be studied whether this is due to less
convergence in the corticothalamic pathway than in the corticocollicular pathway.
In the big brown bat, collicular auditory responses are
strong, so that these responses are attenuated (modulated) in
the IC by corticofugal inhibition (Sun et al. 1996). Similar
data have also been obtained from cats (Amato et al. 1969;
Massopust and Ordy 1962; Watanabe et al. 1966). Because
of the findings by Yan and Suga (1996) and Zhang et al.
(1997), we can offer two explanations as to why Sun et al.
(1996) and others observed only an inhibitory corticofugal
effect. 1) Positive feedback is highly focused to matched
subcortical neurons, so that it is unlikely to be observed without first identifying matching cortical and subcortical re-
11-12-97 00:43:32
neupa
LP-Neurophys
3492
Y. ZHANG AND N. SUGA
FIG . 3. Changes in the frequency-response
curves of thalamic (A and B) and collicular neurons
(C and D) evoked by a large cortical inactivation,
including matched cortical neurons, with muscimol.
Open circles: control condition. Filled circles: muscimol condition. Dashed lines: recovery condition.
Stimulus amplitudes were set at best amplitudes
(BA) of individual neurons, which were Ç30 dB
above minimum threshold (MT).
cording sites. 2) Lateral inhibition is widespread over, presumably, all unmatched neurons within a given cortical subdivision (e.g., the 1.4-mm2 DSCF and the 0.9-mm2 FM-FM
area), so that electrical stimulation of the AC evokes a decrease in the auditory responses of most subcortical neurons,
and lidocaine applied to the AC evokes an increase in the
responses of most subcortical neurons. Our interpretations are
supported by the data obtained from the cat: inactivation of
almost all of the AC by cooling reduced the auditory responses
of MGB neurons (Villa et al. 1991). The difference between
our data and those of Sun et al. is unlikely to be due to
a species difference, because Yan and Suga (unpublished
observations) have obtained the data indicating that cortical
auditory neurons of the big brown bat share identical corticofugal mechanisms with those of the mustached bat.
We thank Dr. J. F. Olsen and A. Kadir for comments on the manuscript.
This work was supported by National Institute of Deafness and Other
Communicative Disorders Grant DC-00175.
Address for reprint requests: N. Suga, Dept. of Biology, Washington
University, 1 Brookings Dr., St. Louis, MO 63130.
Received 12 May 1997; accepted in final form 5 August 1997.
REFERENCES
AMATO, G., LAGRUTTA, V., AND ENIA, F. The control exerted by the auditory cortex on the activity of the medial geniculate body and inferior
colliculus. Arch. Sci. Biol. 53: 291–313, 1969.
ANDERSEN, P., JUNGE, K., AND SVEEN, O. Corticofugal facilitation of thalamic transmission. Brain Behav. Evol. 6: 170–184, 1972.
ANDERSEN, R. A., SYNDER, R. L., AND MERZENICH, M. M. The topographic
organization of corticocollicular projections from physiologically identified loci in the AI, AII, and anterior auditory cortical field of the cat. J.
Comp. Neurol. 191: 479–494, 1980.
HE, J. Modulatory effects of regional cortical activation on the onset response of the cat medical geniculate neurons. J. Neurophysiol. 77: 896–
908, 1997.
HERBERT, H., ASCHOFF, A., AND OSTWALD, J. Topography of projections
/ 9k20$$no65
from the auditory cortex to the inferior colliculus in the rat. J. Comp.
Neurol. 304: 103–122, 1991.
HUFFMAN, R. F. AND HENSON, O. W., JR . The descending auditory pathway
and acousticomotor systems: connections with the inferior colliculus.
Brain Res. 15: 295–323, 1990.
MASSOPUST, L. C., JR . AND ORDY, J. M. Auditory organization of the inferior colliculus in the cat. Exp. Neurol. 6: 465–477, 1962.
RIQUIMAROUX, H., GAIONI, S. J., AND SUGA, N. Cortical computational maps
control auditory perception. Science 251: 565–568, 1991.
RYUGO, D. K. AND WEINBERGER, N. M. Corticofugal modulation of the
medial geniculate body. Exp. Neurol. 51: 377–391, 1976.
SALDANA, E., FELICIANO, M., AND MUGNAINI, E. Distribution of descending
projections from primary auditory neocortex to inferior colliculus mimics
the topography of intracollicular projections. J. Comp. Neurol. 371: 15–
40, 1996.
SUGA, N. AND MANABE, T. Neural basis of amplitude-spectrum representation in auditory cortex of the mustached bat. J. Neurophysiol. 47: 225–
255, 1982.
SUGA, N., NIWA, H., TANIGUCHI, I., AND MARGOLIASH, D. The personalized
auditory cortex of the mustached bat: adaptation for echolocation. J.
Neurophysiol. 58: 643–654, 1987.
SUGA, N., O’NEILL, W. E., KUJIRAI, K., AND MANABE, T. Specificity of
combination-sensitive neurons for processing of complex biosonar signals
in the auditory cortex of the mustached bat. J. Neurophysiol. 49: 1573–
1626, 1983.
SUN, X., CHEN, Q., AND JEN, P.H.-S. Corticofugal control of central auditory
sensitivity in the big brown bat, Eptesicus fuscus. Neurosci. Lett. 212:
131–134, 1996.
SUN, X., JEN, P.H.-S., SUN, D., AND ZHANG, S. Corticofugal influences on
the responses of bat inferior colliculus to sound stimulation. Brain Res.
495: 1–8, 1989.
SYK A, J. AND POPELAR, J. Inferior colliculus in the rat: neuronal response
to stimulation of the auditory cortex. Neurosci. Lett. 51: 235–240, 1984.
VILLA, A.P.E., ROUILLER, E. M., SIMM, G. M., ZURITA, P., DE RIBAUPIERRE,
Y., AND DE RIBAUPIERRE, F. Corticofugal modulation of the information
processing in the auditory thalamus of the cat. Exp. Brain Res. 86: 506–
517, 1991.
WATANABE, T., YANAGISAWA, K., AND KATSUKI, Y. Cortical efferent flow
influencing unit responses of medial geniculate body to sound stimulation.
Exp. Brain Res. 2: 302–317, 1966.
YAN, J. AND SUGA, N. Corticofugal modulation of time-domain processing
of biosonar information in bats. Science 273: 1100–1103, 1996.
ZHANG, Y., SUGA, N., AND YAN, J. Corticofugal modulation of frequency
processing in bat auditory system. Nature 387: 900–903, 1997.
11-12-97 00:43:32
neupa
LP-Neurophys