Neuroscience Letters, 112 (1990) 25-30
25
Elsevier ScientificPublishers Ireland Ltd. NSL 06797
Neuronal substrates involved in processing of communicative acoustic signals in tree shrews: a 2-deoxyglucose study H. Binz, Ch. Zurhorst, E. Zimmermann and H. Rahmann Institute of Zoology, University of Stuttgart-Hohenheim, Stuttgart (F.R.G.)
(Received 9 November 1989; Accepted 28 December 1989) Key words: Central auditory system; Acoustic stimulation; 2-Deoxyglucose; Autoradiography; Tree
shrew Autoradiography with [14C]2-deoxyglucose(2-DG) was used to map functional differencesin activation of the central auditory pathway in adult tree shrews during presentation of particular acoustic stimuli (low frequency, LFS, and high frequency, HFS, pure sinus tones; social calls, SC). Individuals stimulated with broadband-noise (BBN) were used as controls. Stimulus-specificlabelling was found in autoradiographs of cochlear nucleus, superior olivarycomplex, inferior colliculus and auditory cortex. These findings imply a tonotopic organization at least in these auditory brain areas and indicate differencesin the processing of sounds with different functional significance. Acoustic signals are of great importance in the social communication system of tree shrews [2] and are also involved in the detection of prey and predators. However, there is only scarce information about the neuronal processing of sounds in these archaic mammals. Horseradish peroxidase (HRP) studies provided some insight into the interconnections of the central auditory system in tree shrews and suggested a general archaic mammalian pattern. Together with single-unit and multi-unit electrophysiological studies they indicate a tonotopic organization of cochlear nucleus, superior olivary complex, inferior colliculus and auditory cortex for many species other than tupaia. As shown for other mammals especially in inferior colliculus and auditory cortex isofrequency regions are oriented not only in the dorsal-ventral plane with high frequencies represented in more ventral parts of each specific auditory structure and lower frequencies in the dorsal ones. A second plane of tonotopic organization does exist from caudal parts of an acoustic structure towards the rostral parts [4, 11]. The present study was designed to explore possible differences in functional activity in the central auditory system during exposure to stimuli of different frequencies Correspondence: H. Binz, Institute of Zoology, University of Stuttgart-Hohenheim, 7000 Stuttgart 70,
F.R.G. 0304-3940/90/$ 03.50 © 1990 ElsevierScientific Publishers Ireland Ltd.
26 and of communicative or non-communicative meaning. A 3-dimensional (3-D) reconstruction of isofrequency planes within the nuclei was employed to show possible caudo-rostral oriented tonotopy. Adult female tree shrews (Tupaia belangeri) were injected, i.p. with 10/~Ci/100 g body weight of [14C]2-deoxyglucose (2-DG; 59 mCi/mM) in 1.25 ml sterile aqueous solution containing 3% ethanol (Amersham-Buchler). Subjects were placed into a double-walled sound-attenuated chamber with the loudspeaker positioned directly above the animal, for 45 min in the dark during stimulation. Animals were exposed to (1) continuous broadband noise (BBN), (2) pulsed high-frequency sinus tone (7.8 kHz; HFS), (3) pulsed low-frequency sinus tone (1.4 kHz; LFS) both with pulse duration of 55.5 ms and interpulse interval of 165.3 ms, (4) intraspecific low-frequency social call (SC) 'gackern' [2] which corresponds in its time and frequency pattern mainly to stimulus 3. Intensities for all stimuli were adjusted to 70 dB. To avoid artifacts caused by stress and handling, subjects were anesthetized by Halothane in the chamber 10 min before decapitation. Following the 45 min stimulus exposure the anesthetized animals were decapitated, the brain dissected and frozen in isopentane cooled to - 50°C. Brains then were serially sectioned on a cryostat (section thickness: 14/~m). Sections were freeze-dried and exposed to fl-max hyper film (Amersham Buchler) for 21 days and afterwards stained with Cresyl violet. Densitometry was done with a computer-assisted analysis system (Gesotec/Viper 2 planimetry systems; camera: Ikegami, CCD). Image processing of autoradiograms was performed by routines [10], i.e., zeroing at the background. To exclude data variations produced by individual differences Nucleus reticularis gigantocellularis was used as reference. As it was shown for gerbils [9] and in preliminary examinations in tree shrews in our laboratory this structure was not influenced by different acoustic stimulation. To establish a correlation between anatomical structures and marked areas on the autoradiograms user drawn outlines on the histological image were transferred to the autoradiographic image [10]. Areas of interest were measured to provide quantitative data for grey-values and volume. For the 3-D reconstruction boundaries of segments with defined iso-grey-values were marked. All stimuli produced stimulus-specific foci of very high 2-DG uptake when compared to the adjacent structures in all auditory nuclei. Presentation of BBN (controls) resulted in overall labelling of cochlear nucleus, superior olivary complex, inferior coUiculus and auditory cortex, whereas autoradiograms of HFS, LFS and SC animals showed discrete regions of enhanced metabolic activity in all structures with the exception of medial geniculate body. Ventral and dorsal cochlear nucleus show very high activity during acoustic stimulation compared to the level observed in most other auditory structures. Exposure to pure tone stimulation produces two narrow bands, one in each region. HFS bands are located ventrally, while HFS and SC bands are positioned dorsally. Labelling at the superior olivary complex was present but less distinct. HFS produced an enhanced metabolic activity ventrally whereas LFS and SC shifted the marking to a dorsal position in the medial and lateral part. In inferior colliculus the presence of selective labelling was even more striking than
27 STIMULUS:
Broad Band Noise
High Frequency
Low Frequency
Social Call
COLLICULUS INFERIOR
A STIMULUS:
r
Broad Band Noise
High Frequency
Low Frequency
Social Call
L.,i ACOUSTIC CORTEX
B IIH
Fig. 1. Representative autoradiograms showing the patterns of 2-DG uptake in inferior colliculus (A) and
auditory cortex (B) following stimulation with different sounds. Position of the presented slice in the caudoventral plane is shown in Fig. 2. Presentation of detail (bottom of figure) corresponds to autoradiograms of brain sections (top of figure). Length of bar is equivalent to 4 mm (upper set) and 2 mm (lower set). in other regions. Independent o f stimulation type it occurred only in the central nucleus where m o s t o f the neurons o f lateral lemniscus terminate [3]. While B B N animals showed a high level o f D G u p t a k e which was h o m o g e n e o u s t h r o u g h o u t the nucleus in pure tone stimulated individuals especially in the caudal half a characteristic
28
CoIIiculue inferior
A
.tmml
•
o
rne(~"
a
' lat.
[mrnl
Fig. 2.3-D-reconstruction of 2-DG uptake in inferior coliiculus (A) and auditory cortex (B) during stimulation with different sounds. Gray values represent relative optical densities. Arrows mark slices presented in Fig. 1.
banding of typical shape has been observed (Figs. 1A and 2A: representative autoradiograms are presented for each stimulus group; in the 3-D reconstruction slices which correspond to presented autoradiograms are marked with arrows). H F S stimulation produced isofrequency bands which were short at first and tilted slightly downwards. Towards the middle of the nucleus they rapidly increase in length, with the downward tilt drifting towards the lateral border ('Paisley'-pattern). In the rostral half of inferior colliculus the laminar structures seemed to merge. LFS and SC produced a narrow band in the second and in the third quarter. Paralleling the dorsal border of the nucleus on the whole length from the midsagittal plane to the lateral
29
plane, it was curved upward. Labelling produced by SC stimulus corresponded mostly to the LFS but at the caudal end of the nucleus was pronounced. It was a bit more generally spread within the nucleus which is possibly due to the fact that the SC does not consist of a pure sinus tone but is composed of a fundamental frequency and two harmonics. According to data from literature about cytoarchitectonics of the temporal c o r t e x in tree shrews [5] ventral part of Corpus geniculatum mediale receives major input from the inferior colliculus and projects in turn mostly to primary auditory cortex. This was found to be consistent with our data. Corpus geniculatum mediale was exclusively labelled in its ventral part. In temporal cortex only primary auditory cortex was labelled in all different stimulation groups. Terminations of projections from the ventral medial geniculate body are found primarily in layer IV and the adjacent part of layer III. Mapping of primary auditory cortex with microelectrodes in tree shrews [6] showed a tonotopic organization with high frequencies represented rostrodorsally and lower frequencies activating caudoventral areas. These findings do not correspond to our results with autoradiography. In all animals the very rostral parts of auditory cortex showed an unspecific labelling. Autoradiograms of primary auditory cortex (Fig. 1B) showed for BBN stimulation labelling of almost the whole primary auditory cortex being restricted, however, to layers IV and III. HFS elicited a narrow band of pronounced metabolic activity over all layers which was located dorsally throughout the whole primary auditory cortex (Fig. 2B; slices which are presented as autoradiograms are indicated by arrow). Labelling produced by LFS was found to be arranged in a narrow stripe along the ventral part of primary auditory cortex. In SC-stimulated animals labeling was much more distinct than in the other stimulation groups. Whether the lack of metabolic activity in dorsal layers IV and III in the caudal region of primary cortex was due to inhibitory effects related to the communicative signal or not has to be subject of further investigation. In general, the present study revealed by means of autoradiography that auditory pathway of tree shrews shows the features of tonotopy. Similar to other mammals (marmoset [1]; ferret [7]; rat [8]) frequency representation in auditory cortex appears to be tonotopically organized in the dorsoventral plane. Our results, however, do not provide evidence for pronounced differences in tonotopy from caudal to rostral regions as reported for tupaia and other mammals, e.g., gerbils [11]. Communicative signals elicit a pattern of metabolic activity which is quite different to that produced by artificial stimuli. 1 Aitkin, L.M., Merzenich, M.M., Irvine, D.R.F., Clarey, J.C. and Nelson, J.E., Frequency Representation in Auditory Cortex of the common marmoset (Callithrixjacchusjacchus), J. Comp. Neurol., 252 (1986) 175-185. 2 Binz, H. and Zimmermann, E., The vocal repertoire of adult tree shrews, Behaviour, 109 (1989) 142-162. 3 Casseday, J.H., Diamond, I.T. and Harting, J.K., Auditory pathways to the cortex in Tupaia glis, J. Comp. Neurol., 166 (1976) 303-340. 4 Huang, C. and Fex, J., Tonotopic organization in the inferior colliculus of the rat demonstrated with the 2-deoxyglucose method, Exp. Brain Res., 61 (1986) 506-512.
30 5 Oliver, D.L. and Hall, W.C., The medial geniculate body of the tree shrew, Tupaia glis, J. Comp. Neurol., 182 (1978) 459--493. 6 Oliver, D.L. and Merzenich, M.M., Tonotopic organizaton and connections of primary auditory cortex in the tree shrew, Tupaia glis, Anat. Rec., 184 (1976) 491. 7 Phillips, D.P., Judge, P.W. and Kelly, J.B., Primary auditory cortex in the ferret (Mustela putorius): neural response properties and topographic organization, Brain Res., 443 (I 988) 281-294. 8 Sally, S.L. and Kelly, J.B., Organization of auditory cortex in the albino rat: sound frequency, J. Neurophysiol., 59 (1988) 1627-1638. 9 Sharp, F.R., Ryan, A.F., Goodwin, P. and Woolf, N.K., Increasing intensities of wide band noise increase ['4C]2-deoxyglucose uptake in gerbil central auditory structures, Brain Res., 230 (1981) 87-96. 10 Shivaramakrishnan, K. and Tretiak, O.J., Database management in autoradiography, Comput. Med. Imaging Graph., 13 (1989) 115-135. 11 Steffen, H., Simonis, C., Thomas, H., Tillein, J. and Scheich, H., Auditory cortex: multiple fields, their architectonics and connections in the mongolian gerbil. In J. Syka and R.B. Masterton (Eds.), Auditory Pathway: Structure and Function, Plenum, New York, 1987, pp. 123-132.
25
Elsevier ScientificPublishers Ireland Ltd. NSL 06797
Neuronal substrates involved in processing of communicative acoustic signals in tree shrews: a 2-deoxyglucose study H. Binz, Ch. Zurhorst, E. Zimmermann and H. Rahmann Institute of Zoology, University of Stuttgart-Hohenheim, Stuttgart (F.R.G.)
(Received 9 November 1989; Accepted 28 December 1989) Key words: Central auditory system; Acoustic stimulation; 2-Deoxyglucose; Autoradiography; Tree
shrew Autoradiography with [14C]2-deoxyglucose(2-DG) was used to map functional differencesin activation of the central auditory pathway in adult tree shrews during presentation of particular acoustic stimuli (low frequency, LFS, and high frequency, HFS, pure sinus tones; social calls, SC). Individuals stimulated with broadband-noise (BBN) were used as controls. Stimulus-specificlabelling was found in autoradiographs of cochlear nucleus, superior olivarycomplex, inferior colliculus and auditory cortex. These findings imply a tonotopic organization at least in these auditory brain areas and indicate differencesin the processing of sounds with different functional significance. Acoustic signals are of great importance in the social communication system of tree shrews [2] and are also involved in the detection of prey and predators. However, there is only scarce information about the neuronal processing of sounds in these archaic mammals. Horseradish peroxidase (HRP) studies provided some insight into the interconnections of the central auditory system in tree shrews and suggested a general archaic mammalian pattern. Together with single-unit and multi-unit electrophysiological studies they indicate a tonotopic organization of cochlear nucleus, superior olivary complex, inferior colliculus and auditory cortex for many species other than tupaia. As shown for other mammals especially in inferior colliculus and auditory cortex isofrequency regions are oriented not only in the dorsal-ventral plane with high frequencies represented in more ventral parts of each specific auditory structure and lower frequencies in the dorsal ones. A second plane of tonotopic organization does exist from caudal parts of an acoustic structure towards the rostral parts [4, 11]. The present study was designed to explore possible differences in functional activity in the central auditory system during exposure to stimuli of different frequencies Correspondence: H. Binz, Institute of Zoology, University of Stuttgart-Hohenheim, 7000 Stuttgart 70,
F.R.G. 0304-3940/90/$ 03.50 © 1990 ElsevierScientific Publishers Ireland Ltd.
26 and of communicative or non-communicative meaning. A 3-dimensional (3-D) reconstruction of isofrequency planes within the nuclei was employed to show possible caudo-rostral oriented tonotopy. Adult female tree shrews (Tupaia belangeri) were injected, i.p. with 10/~Ci/100 g body weight of [14C]2-deoxyglucose (2-DG; 59 mCi/mM) in 1.25 ml sterile aqueous solution containing 3% ethanol (Amersham-Buchler). Subjects were placed into a double-walled sound-attenuated chamber with the loudspeaker positioned directly above the animal, for 45 min in the dark during stimulation. Animals were exposed to (1) continuous broadband noise (BBN), (2) pulsed high-frequency sinus tone (7.8 kHz; HFS), (3) pulsed low-frequency sinus tone (1.4 kHz; LFS) both with pulse duration of 55.5 ms and interpulse interval of 165.3 ms, (4) intraspecific low-frequency social call (SC) 'gackern' [2] which corresponds in its time and frequency pattern mainly to stimulus 3. Intensities for all stimuli were adjusted to 70 dB. To avoid artifacts caused by stress and handling, subjects were anesthetized by Halothane in the chamber 10 min before decapitation. Following the 45 min stimulus exposure the anesthetized animals were decapitated, the brain dissected and frozen in isopentane cooled to - 50°C. Brains then were serially sectioned on a cryostat (section thickness: 14/~m). Sections were freeze-dried and exposed to fl-max hyper film (Amersham Buchler) for 21 days and afterwards stained with Cresyl violet. Densitometry was done with a computer-assisted analysis system (Gesotec/Viper 2 planimetry systems; camera: Ikegami, CCD). Image processing of autoradiograms was performed by routines [10], i.e., zeroing at the background. To exclude data variations produced by individual differences Nucleus reticularis gigantocellularis was used as reference. As it was shown for gerbils [9] and in preliminary examinations in tree shrews in our laboratory this structure was not influenced by different acoustic stimulation. To establish a correlation between anatomical structures and marked areas on the autoradiograms user drawn outlines on the histological image were transferred to the autoradiographic image [10]. Areas of interest were measured to provide quantitative data for grey-values and volume. For the 3-D reconstruction boundaries of segments with defined iso-grey-values were marked. All stimuli produced stimulus-specific foci of very high 2-DG uptake when compared to the adjacent structures in all auditory nuclei. Presentation of BBN (controls) resulted in overall labelling of cochlear nucleus, superior olivary complex, inferior coUiculus and auditory cortex, whereas autoradiograms of HFS, LFS and SC animals showed discrete regions of enhanced metabolic activity in all structures with the exception of medial geniculate body. Ventral and dorsal cochlear nucleus show very high activity during acoustic stimulation compared to the level observed in most other auditory structures. Exposure to pure tone stimulation produces two narrow bands, one in each region. HFS bands are located ventrally, while HFS and SC bands are positioned dorsally. Labelling at the superior olivary complex was present but less distinct. HFS produced an enhanced metabolic activity ventrally whereas LFS and SC shifted the marking to a dorsal position in the medial and lateral part. In inferior colliculus the presence of selective labelling was even more striking than
27 STIMULUS:
Broad Band Noise
High Frequency
Low Frequency
Social Call
COLLICULUS INFERIOR
A STIMULUS:
r
Broad Band Noise
High Frequency
Low Frequency
Social Call
L.,i ACOUSTIC CORTEX
B IIH
Fig. 1. Representative autoradiograms showing the patterns of 2-DG uptake in inferior colliculus (A) and
auditory cortex (B) following stimulation with different sounds. Position of the presented slice in the caudoventral plane is shown in Fig. 2. Presentation of detail (bottom of figure) corresponds to autoradiograms of brain sections (top of figure). Length of bar is equivalent to 4 mm (upper set) and 2 mm (lower set). in other regions. Independent o f stimulation type it occurred only in the central nucleus where m o s t o f the neurons o f lateral lemniscus terminate [3]. While B B N animals showed a high level o f D G u p t a k e which was h o m o g e n e o u s t h r o u g h o u t the nucleus in pure tone stimulated individuals especially in the caudal half a characteristic
28
CoIIiculue inferior
A
.tmml
•
o
rne(~"
a
' lat.
[mrnl
Fig. 2.3-D-reconstruction of 2-DG uptake in inferior coliiculus (A) and auditory cortex (B) during stimulation with different sounds. Gray values represent relative optical densities. Arrows mark slices presented in Fig. 1.
banding of typical shape has been observed (Figs. 1A and 2A: representative autoradiograms are presented for each stimulus group; in the 3-D reconstruction slices which correspond to presented autoradiograms are marked with arrows). H F S stimulation produced isofrequency bands which were short at first and tilted slightly downwards. Towards the middle of the nucleus they rapidly increase in length, with the downward tilt drifting towards the lateral border ('Paisley'-pattern). In the rostral half of inferior colliculus the laminar structures seemed to merge. LFS and SC produced a narrow band in the second and in the third quarter. Paralleling the dorsal border of the nucleus on the whole length from the midsagittal plane to the lateral
29
plane, it was curved upward. Labelling produced by SC stimulus corresponded mostly to the LFS but at the caudal end of the nucleus was pronounced. It was a bit more generally spread within the nucleus which is possibly due to the fact that the SC does not consist of a pure sinus tone but is composed of a fundamental frequency and two harmonics. According to data from literature about cytoarchitectonics of the temporal c o r t e x in tree shrews [5] ventral part of Corpus geniculatum mediale receives major input from the inferior colliculus and projects in turn mostly to primary auditory cortex. This was found to be consistent with our data. Corpus geniculatum mediale was exclusively labelled in its ventral part. In temporal cortex only primary auditory cortex was labelled in all different stimulation groups. Terminations of projections from the ventral medial geniculate body are found primarily in layer IV and the adjacent part of layer III. Mapping of primary auditory cortex with microelectrodes in tree shrews [6] showed a tonotopic organization with high frequencies represented rostrodorsally and lower frequencies activating caudoventral areas. These findings do not correspond to our results with autoradiography. In all animals the very rostral parts of auditory cortex showed an unspecific labelling. Autoradiograms of primary auditory cortex (Fig. 1B) showed for BBN stimulation labelling of almost the whole primary auditory cortex being restricted, however, to layers IV and III. HFS elicited a narrow band of pronounced metabolic activity over all layers which was located dorsally throughout the whole primary auditory cortex (Fig. 2B; slices which are presented as autoradiograms are indicated by arrow). Labelling produced by LFS was found to be arranged in a narrow stripe along the ventral part of primary auditory cortex. In SC-stimulated animals labeling was much more distinct than in the other stimulation groups. Whether the lack of metabolic activity in dorsal layers IV and III in the caudal region of primary cortex was due to inhibitory effects related to the communicative signal or not has to be subject of further investigation. In general, the present study revealed by means of autoradiography that auditory pathway of tree shrews shows the features of tonotopy. Similar to other mammals (marmoset [1]; ferret [7]; rat [8]) frequency representation in auditory cortex appears to be tonotopically organized in the dorsoventral plane. Our results, however, do not provide evidence for pronounced differences in tonotopy from caudal to rostral regions as reported for tupaia and other mammals, e.g., gerbils [11]. Communicative signals elicit a pattern of metabolic activity which is quite different to that produced by artificial stimuli. 1 Aitkin, L.M., Merzenich, M.M., Irvine, D.R.F., Clarey, J.C. and Nelson, J.E., Frequency Representation in Auditory Cortex of the common marmoset (Callithrixjacchusjacchus), J. Comp. Neurol., 252 (1986) 175-185. 2 Binz, H. and Zimmermann, E., The vocal repertoire of adult tree shrews, Behaviour, 109 (1989) 142-162. 3 Casseday, J.H., Diamond, I.T. and Harting, J.K., Auditory pathways to the cortex in Tupaia glis, J. Comp. Neurol., 166 (1976) 303-340. 4 Huang, C. and Fex, J., Tonotopic organization in the inferior colliculus of the rat demonstrated with the 2-deoxyglucose method, Exp. Brain Res., 61 (1986) 506-512.
30 5 Oliver, D.L. and Hall, W.C., The medial geniculate body of the tree shrew, Tupaia glis, J. Comp. Neurol., 182 (1978) 459--493. 6 Oliver, D.L. and Merzenich, M.M., Tonotopic organizaton and connections of primary auditory cortex in the tree shrew, Tupaia glis, Anat. Rec., 184 (1976) 491. 7 Phillips, D.P., Judge, P.W. and Kelly, J.B., Primary auditory cortex in the ferret (Mustela putorius): neural response properties and topographic organization, Brain Res., 443 (I 988) 281-294. 8 Sally, S.L. and Kelly, J.B., Organization of auditory cortex in the albino rat: sound frequency, J. Neurophysiol., 59 (1988) 1627-1638. 9 Sharp, F.R., Ryan, A.F., Goodwin, P. and Woolf, N.K., Increasing intensities of wide band noise increase ['4C]2-deoxyglucose uptake in gerbil central auditory structures, Brain Res., 230 (1981) 87-96. 10 Shivaramakrishnan, K. and Tretiak, O.J., Database management in autoradiography, Comput. Med. Imaging Graph., 13 (1989) 115-135. 11 Steffen, H., Simonis, C., Thomas, H., Tillein, J. and Scheich, H., Auditory cortex: multiple fields, their architectonics and connections in the mongolian gerbil. In J. Syka and R.B. Masterton (Eds.), Auditory Pathway: Structure and Function, Plenum, New York, 1987, pp. 123-132.
