Recent Advances in Vestibulo-auditory Neurobiology (Neuroembryology and Aging 2004 2005)

A-PDF MERGER DEMO

Recent Advances in Vestibulo-Auditory Neurobiology

Guest Editors

Ying-Shing Chan, Hong Kong Jufang He, Hong Kong

26 figures, 3 in color, and 1 table, 2005

Basel • Freiburg • Paris • London • New York • Bangalore • Bangkok • Singapore • Tokyo • Sydney

S. Karger Medical and Scientific Publishers Basel • Freiburg • Paris • London New York • Bangalore • Bangkok Singapore • Tokyo • Sydney

Disclaimer The statements, options and data contained in this publication are solely those of the individual authors and contributors and not of the publisher and the editor(s). The appearance of advertisements in the journal is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements. Drug Dosage The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.

Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher or, in the case of photocopying, direct payment of a specified fee to the Copyright Clearance Center (see ‘General Information’). © Copyright 2005 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel ISBN-10: 3–8055–8227–7 ISBN-13: 978–3–8055–8227–8

Vol. 3, No. 4, 2004–05

Contents

161 Editorial Overview Chan, Y.S.; He, J. (Hong Kong) 162 Toward Maturation of the Vestibular System: Neural Circuits and Neuronal

Properties Lai, C.H.; Tse, Y.C.; Shum, D.K.Y.; Chan, Y.S. (Hong Kong) 171 What Peripheral Vestibular Manipulations Reveal about the Function and

Plasticity in the Primate Oculomotor System Newlands, S.D. (Galveston, Tex.); Angelaki, D.E. (St. Louis, Mo.) 183 Cellular Mechanisms of Vestibular Compensation Paterson, J.M.; Menzies, J.R.W.; Bergquist, F.; Dutia, M.B. (Edinburgh) 194 Recovery after Vestibular Lesions: From Animal Models to Patients Vidal, P.P.; Straka, H.; Vibert, N.; Moore, L.E.; de Waele, C. (Paris) 207 Peripheral Vestibular Responses to Sound Curthoys, I.S. (Sydney) 215 The Inferior Colliculus: A Center for Convergence of Ascending and

Descending Auditory Information Malmierca, M.S. (Salamanca) 230 From Receptive Field Dynamics to the Rate of Transmitted Information:

Some Facets of the Thalamocortical Auditory System Huetz, C.; Edeline, J.M. (Orsay) 239 Thalamocortical and Corticothalamic Interaction in the Auditory System Zhang, Z.; Chan, Y.S.; He, J. (Hong Kong)

249 250 251 after 252

Author Index/Subject Index Vol. 3, No. 4, 2004–05 Author Index Vol. 3, 2004–05 Subject Index Vol. 3, 2004–05 Contents Vol. 3, 2004–05

© 2005 S. Karger AG, Basel Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Access to full text and tables of contents, including tentative ones for forthcoming issues: www.karger.com/nba_issues

Neuroembryol Aging 2004–05;3:161 DOI: 10.1159/000096793

Published online: November 6, 2006

Editorial Overview Ying-Shing Chan a Jufang He b a

Department of Physiology and Research Center of the Heart, Brain, Hormone and Healthy Aging, LKS Faculty of Medicine, The University of Hong Kong, and b Department of Rehabilitation Sciences, Hong Kong Polytechnic University, Hong Kong, SAR, China

The vestibular system and auditory system utilize a common form of mechanosensitive hair cells located in the fluid-filled inner ear. Activation of these mechanoreceptors by appropriate stimuli triggers a cascade of neural events along distinctive pathways thereby providing the brain with the senses of balance and hearing. The review articles collected in this special issue summarize the latest progress of research in these two systems spanning from the dynamics of development to plasticity in adult circuitry. They also explore current advances in cellular mechanism, signal processing and clinical application. The first article by Lai et al. [pp 162–170] outlines the prenatal and postnatal developmental profiles of vestibular neural circuits. Emphasis is placed on the postnatal maturation of central vestibular neurons, including the role of glutamatergic transmission and the coding of gravity-related spatial information. The article of Newlands and Angelaki [pp 171–182] provides an overview on the compensatory processes of the primate oculomotor system following manipulations of the peripheral vestibular apparatus. Focus is placed on the plasticity of otolith/canal interaction and eye-head coordination especially during functional recovery associated with vestibular compensation. This article also probes into the spatial orientation function. Paterson et al. [pp 183–193] focus on the cellular mechanisms of vestibular compensation pertaining to the remodeling of circuitry and synaptic plasticity in the vestibular nucleus and cerebellum. Insights on the involvement of the neuroendocrine system in balance disorders are also presented. The article of Vidal et al. [pp 194–206] addresses the recovery of vestibulo-ocular reflexes and postural deficits during ves-

© 2005 S. Karger AG, Basel 1661–3406/05/0034–0161$22.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

tibular compensation. Plastic changes in the membrane properties of medial vestibular neurons parallel the recovery of vestibulo-ocular reflexes. The correlation of findings from animal studies with observations from vestibular patients offers a perspective on the clinical impact of vestibular disorders. Curthoys’ [pp 207–214] article on vestibular responses to sound strategically bridges the foregoing articles on the vestibular system and the following three articles covering the auditory system. Evidence derived from animal studies which revealed the activation of vestibular receptors by sound has been translated into a new clinical test of human vestibular function using sound as the test stimulus. Malmierca’s [pp 215–229] article summarizes the anatomical and physiological features of the inferior colliculus. This article has put emphasis on the convergence of ascending and descending information. The article of Huetz and Edeline [pp 230–238] focuses on the thalamocortical system. It addresses the information relay system from the perspective of receptive field dynamics and the rate of signal transmission. The article by Zhang et al. [pp 239– 248] summarizes the thalamocortical and corticothalamic interaction in the auditory system. They propose a model in which the corticothalamic system serves as the executive filter for selective attention. The trend of future research in systems neuroscience is highlighted by the last three articles in which specific relay stations are explored as integral parts of a sensory system. With review articles on cognate disciplines in vestibular and auditory research, this special issue portrays the synergism that is necessary for future consolidation of vestibulo-auditory neurobiology.

Prof. Y.S. Chan Department of Physiology, LKS Faculty of Medicine, The University of Hong Kong 21 Sassoon Road Hong Kong, SAR (China) Tel. +852 2819 9263, Fax +852 2855 9730, E-Mail [email protected]

Neuroembryol Aging 2004–05;3:162–170 DOI: 10.1159/000096794

Published online: November 6, 2006

Toward Maturation of the Vestibular System: Neural Circuits and Neuronal Properties Chun-Hong Lai a Yiu-Chung Tse a Daisy Kwok-Yan Shum b, c Ying-Shing Chan a, c Departments of a Physiology and b Biochemistry, and c Research Center of the Heart, Brain, Hormone and Healthy Aging, LKS Faculty of Medicine, The University of Hong Kong, Hong Kong, SAR, China

Key Words Glutamatergic neurotransmission
Pre- and postnatal development
Spatiotemporal information

Abstract This review focuses on efforts to unravel conundrums on the development of the vestibular system. In the first section, maturation of the peripheral vestibular system and the involvement of transcription factors in the patterning of peripheral vestibular components are highlighted. Cell surface and matrix molecules have also been implicated in fasciculation and guidance of axons in the course of vestibular circuit formation. In rodents, the central vestibular neurons continue to develop after birth until they reach maturity in morphology and function. Sequential maturation of neuronal subpopulations within the developing network of the horizontal and vertical otolith systems is also presented. In another section, the expression pattern of glutamate receptor subunits within the developing vestibular nuclear complex is reviewed in relation to their potential role in regulating postnatal function of the vestibular system. Lastly, postnatal changes in the properties of vestibular nuclear neurons and their capability in coding head movement information appear to prime the development of vestibular-related motor functions. Copyright © 2005 S. Karger AG, Basel

© 2005 S. Karger AG, Basel 1661–3406/05/0034–0162$22.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

During development, both the peripheral components and central circuits of the vestibular system undergo progressive changes in structure [1–3], connectivity [4] and functional characteristics [5–7]. The properties of central vestibular neurons are known to be modulated by glutamate receptors [8–11], yet the roles of these receptors in postnatal regulation of vestibular functions have received less attention. In this review, we will also outline the electrophysiological features of central vestibular neurons and their capacity in coding canal and otolith inputs in the course of postnatal development.

Morphological Maturation of the Vestibular System

Peripheral Components The developmental profile of vestibular hair cells and their afferent innervation has been extensively studied in rodents. In the mouse, the vestibular sensory epithelium remains undifferentiated until embryonic day (E) 14 [12]. At E14, nascent hair bundles, which are formed by a group of cilia of the same length and have no precise polarity, can be seen in the immature sensory epithelium, indicating that a few hair cells have begun their differentiation [13, 14]. From E14 to postnatal day (P) 2, terminal mitoses of hair cells and supporting cells occur [12]. By E16, imma-

Prof. Y.S. Chan Department of Physiology, LKS Faculty of Medicine, The University of Hong Kong 21 Sassoon Road Hong Kong, SAR (China) Tel. +852 2819 9263, Fax +852 2855 9730, E-Mail [email protected]

ture hair cells with short stereocilia are evident [15]. Morphologic differentiation into type I and type II hair cells starts at E19 [16]. However, voltage-gated conductance of vestibular hair cells remains undetectable from birth [17– 19] to P4 [20, 21], when the cells progressively acquire voltage-dependent properties of mature hair cells. In the mouse, afferent nerve endings at vestibular receptors are first identified at E17 [22, 23]. Numerous unmyelinated afferent fibers contact the receptor hair cells at their base with the formation of flat and vesiculated terminals. These early hair cell-afferent fiber contacts develop into synapses, characterized by the presence of synaptic bodies (in the form of microvesicles and dense core vesicles) in the presynaptic element [24, 25]. Partial calyx endings were detectable as early as E18 [23], and full calyx endings became detectable in significant numbers by P2–P3 [26]. Efferent nerve terminals at the vestibular hair cells were recognizable as early as E18 in the mouse [22, 23]. At birth, more efferent endings were found to make contact either directly on the hair cells or on nerve calyces. The amount of efferent endings, however, has not reached adult levels [24]. In rats, immunohistochemistry together with confocal microscopy demonstrated that efferent innervation continues to develop until P12 [27, 28]. In this postnatal period, rats exposed to altered gravitational environment showed no change in the organization of the vestibular efferent network within the utricle [29]. In adult rats, the vestibular efferent system, which originates from a few neurons in the brainstem, has been proposed to provide modulatory control over signal transduction between the hair cells and the afferents [28, 30– 32]. Efferent activation in toadfish was found to result in elevation of afferent activity such that the rectified response to acceleration became bidirectionally responsive without change in response sensitivity [30]. An age-dependent decrease in the number of clipped otolith afferents has also been attributed to the maturation of the efferent system [6, 33], which does not fully develop until the 3rd postnatal week in rats [34]. Nonetheless, the contribution of the efferent system to postnatal maturation of vestibular function awaits further study. In addition, inputs to the vestibular efferent neurons have yet to be reported. In cats, Dechesne et al. [35] reported that vestibular efferent neurons did not receive direct inputs from the sensory receptors. Our recent study in adult rats, however, showed direct projections from vestibular afferent terminals to the ipsilateral vestibular efferent nucleus [36], suggesting a short closed loop between vestibular afferent and efferent systems.

Based on studies of knockout mice, development of the vestibular ganglion can be divided into phases [37]. First, cells in the anteroventral lateral region of the otocyst delaminate from the optic epithelium. Then these neuroblasts migrate away and undergo further proliferation before coalescing to form a ganglion that later divides to form the vestibular and spiral ganglia [38]. Numerous transcription factors have been identified to be essential for the ganglion development. For example, the basic helix-loop-helix factors neurogenin 1 (Ngn1) and 2 (Ngn2), which have been shown to be essential for cell fate commitment [39, 40], are expressed in the ganglion placodes as early as E8.5 in mice. Mutations in ngn1 and ngn2 in the early phase abolish the development of the vestibular-cochlear ganglia [39–41]. Likewise, another basic helix-loop-helix family member, BETA2/NeuroD1, is required for neurogenesis of the vestibular-cochlear ganglion [42]. In addition, the homeodomain-containing transcription factor Phox2a has been shown to be required for differentiation and maintenance of ganglion neurons [43]. Gene-targeting analyses have also demonstrated essential roles for the Pit1-Oct1-Unc86 (POU) domain transcription factors Brn3a and Brn3b in supporting the survival of vestibular and cochlear ganglia [41, 44, 45]. Loss of Brn3a affects neuronal differentiation, neural projections and target innervation [46]. Besides, Brn3a controls survival and differentiation of sensory neurons by regulating different downstream genes (such as trkB and trkC). The expression of the neurotrophin receptors TrkA, TrkB, and TrkC in differentiating ganglion neurons marks a late phase of ganglionic development. In rat embryos, neurotrophins are important for the establishment of contacts between ganglion neurons and target cells in the inner ear [47–50]. The survival of these neurons is also dependent on neurotrophins such as brain-derived neurotrophic factor (BDNF) and neurotrophin-3 (NT-3) [37]. BDNF plays a dominant role in the development and survival of mammalian vestibular ganglion neurons [51–53], whereas NT-3 appears to be required for the survival of cochlear ganglion neurons [54]. Knockout of bdnf or its high-affinity receptor, TrkB, results in no innervation of the three cristae and poor innervation of the two maculae [51, 55, 56]. In both embryonic and neonatal rats, NT-3 mRNA is detected in differentiating hair cells and surrounding supporting cells of the cochlear and vestibular sensory epithelia, while BDNF mRNA was found localized exclusively to differentiating hair cells [48]. At the postnatal stage, however, the expression of BDNF and NT-3 is more restricted. NT-3 mRNA is expressed in in-

Embryonic and Postnatal Vestibular Development

Neuroembryol Aging 2004–05;3:162–170

163

ner hair cells of the cochlea and BDNF mRNA in hair cells of the vestibular end organs [48, 57]. Vestibular Neural Circuits In all vertebrates, the embryonic hindbrain neuroepithelium is organized into a series of segments called rhombomeres [58–60]. In larval frogs, for example, vestibular afferent fibers enter the brainstem at rhombomere 4 (r4), then split into ascending and descending fiber bundles that terminate in the dorsal regions of r1–r8, thus laying the scaffold for development into the vestibular nuclear complex of adult frogs [61, 62]. Of the four main vestibular nuclei, the superior vestibular nucleus appears to derive from larval r1/r2, the lateral vestibular nucleus from r3/r4, and the major portions of medial and descending vestibular nuclei from r5–r8 [63, 64]. The development of vestibular neural circuits at embryonic stages largely depends on the ability of axons to fasciculate/defasciculate and to follow their correct trajectories. Fasciculation of axons reduces the complexity of mechanisms required to guide a large array of axons, but defasciculation of axons in the vicinity of the target contributes to proper selection of the synaptic partner. Thus spatial and temporal regulation of axonal fasciculation is critical for the development of precise patterning of neural networks. Different molecules have been implicated in regulating axonal fasciculation and pathfinding in prenatal development of the vestibular system [65–67]. One of these involves the murine SC1-related protein MuSC, a member of the immunoglobulin superfamily. Downregulation of its expression has been implicated in the defasciculation of primary vestibular afferents within the cerebellar primordium [66]. Moreover, Tashiro et al. [65] have highlighted the importance of local cues in the guidance of vestibular afferent axons to their final target. The involvement of matrix molecules in determining the routing of vestibular fibers has recently been reported [67]. And yet, how these molecules regulate the bioavailability of local cues to cell surface receptors for neurite growth and pathfinding in the development of the vestibular nucleus remains to be elucidated. Furthermore, the transcription factor Mash1 was found to play an important role in the migration of r4-derived vestibular efferent neurons in mice [68]. In the rat, the differentiation of vestibular nuclear neurons occurs during E11–E15, following a lateral-tomedial and rostral-to-caudal internuclear gradient [1]. The peak production time is E12 in the lateral vestibular nucleus, E13 in the superior nucleus, E13–E14 in the descending (or spinal) nucleus, and E14 in the medial nu164

Neuroembryol Aging 2004–05;3:162–170

cleus [1]. The sequential differentiation of these nuclei can be correlated with maturation of function attributable to the respective nuclei. For instance, the lateral vestibular nucleus, involved in postural stabilization of neonates, matures earlier than the superior nucleus, which is involved in the control of eye movement that becomes functionally important only by 2 weeks after birth [1]. The temporal sequence of somatic/dendritic growth and synaptic formation within the vestibular nucleus of rats has also been reported [3, 4]. At birth, Deiters’ neurons show small soma and lack extensive dendrites [3]. Rapid increase in dendritic arborization and soma size follows in the 1st postnatal week. Electron-microscopic analysis also indicated that synaptic contacts between vestibular afferents and secondary vestibular neurons were observable as from P6 [4], while structural maturation of this contact was observed in the 2nd postnatal week [3]. The developmental profile of otolith-related vestibular nuclear neurons as revealed by the expression of Fos protein in cell nuclei of functionally activated neurons was also examined in postnatal rats [69, 70]. The coding of horizontal and vertical head positions in Fos-immunoreactive (ir) otolith-related vestibular nuclear neurons was observed as early as P7. These neurons increased steadily with age. When compared with the adults, postnatal Fos expression was concentrated in a centrally located subset of neurons [69–72]. In such brainstem areas as the prepositus hypoglossal nucleus, locus cerulleus and gigantocellular reticular nucleus which receive vestibular inputs, there were also age-dependent increases in Fos-ir neurons after off-vertical axis rotation [69] or vertical linear acceleration [70]. Unlike the aforementioned areas where Fos-ir neurons were first observed at P7, neurons in the dorsal-medial cell column and the
subnucleus of the inferior olive did not show c-fos expression until the 2nd postnatal week. Thus, sequential maturation of neuronal subpopulations within the vestibular circuits provides the cellular substrates for the recognition of gravity-related three-dimensional spatial information.

Maturation of Glutamatergic Neurotransmission in the Vestibular Nuclear Complex

The transmitters involved in neurotransmission and neuromodulation of vestibular nuclear neurons are conveniently classified into excitatory and inhibitory amino acids (e.g. glutamate, -aminobutyric acid and glycine), monoamines (e.g. dopamine and serotonin) and neuroactive peptides [73]. Given that glutamatergic transmisLai /Tse /Shum /Chan

sion is the major excitatory input at the projection fields of vestibular afferents within the vestibular nucleus [73– 75], the expression of different glutamate receptors on vestibular nuclear neurons becomes a central issue in the understanding of vestibular function. The developmental profile of glutamate receptor subunits within the vestibular nuclei is therefore reviewed. Ionotropic Glutamate Receptors Ionotropic glutamate receptors mediate the transmission of vestibular afferent volleys to second-order vestibular neurons. Electrophysiological and pharmacological studies in adult rats and guinea pigs showed that vestibular nuclear neurons are activated by primary vestibular afferent inputs via N-methyl-D-asparate receptor (NMDA) or -amino-3-hydroxyl-5-methyl-4-isoxazolepropinic acid (AMPA) receptors [75–79]. Infusion of NMDA antagonist into the vestibular nuclei of guinea pigs induced postural and oculomotor syndromes similar to those following vestibular deafferentation [80], suggesting that in adult animals, NMDA receptor channels play essential roles in the maintenance of postural symmetry and eye position. In second-order vestibular neurons, both the field potentials and excitatory postsynaptic (EPS) potentials were suppressed by specific AMPA and NMDA antagonists [80–82]. AMPA antagonists suppressed a large part of the field potential and the fast component EPS potentials, whereas NMDA antagonists abolished the remaining field potential and EPS potentials [80]. In brain slice cultures of developing rats (P4–P6), the fast component of EPS current in vestibular nuclear neurons was blocked by 6-cyano-7-nitroquinoxaline-2,3-dione, an AMPA receptor antagonist [81]. The slow component of 6-cyano-7-nitroquinoxaline-2,3-dione-resistant EPS current, however, was abolished by D-2-amino-5-phosphonovaleric acid, an NMDA receptor antagonist. In adult rats, a large proportion of neurons coexpressing NMDA and AMPA receptors were found in the vestibular nucleus [83, 84]. These further corroborate the notion that activation of vestibular nuclear neurons by vestibular afferents involved concomitant participation of AMPA and NMDA receptors [85]. In embryonic and hatching chicks, however, the occurrence of spontaneous excitatory synaptic activity of tangential vestibular neurons was attributed to AMPA receptors [86, 87]. Most native NMDA receptors are composed of NR1 and NR2 subunits. In adult rats, NR1 and NR2 modulate neurotransmission within the medial vestibular nucleus [82]. In the vestibular nuclei of postnatal rats, the differEmbryonic and Postnatal Vestibular Development

ential expression of NMDA receptor subunits [10, 88] may provide for fine-tuning of the properties of voltagegated ion channels. The level of NR1 subunit expression was higher in neonatal rats than in adults. Among the four NR2 receptor subunits, NR2A, NR2B and NR2D were present at P2 while NR2C was not observed until P7. The late occurrence of NR2C in the NMDA receptor complex of vestibular nuclear neurons, as observed in the developing mouse cerebellum, is responsible for a reduction in the sensitivity of NMDA receptors during postnatal development [89, 90]. It is also noteworthy that the latency of NMDA receptor-mediated EPS currents in medial vestibular neurons of adult rats [76] was longer than in younger rats [79]. In the vestibular nuclei, gradual increases in the expression level of NR2A, NR2B and NR2C were shown for the first 3 postnatal weeks [10]. These data were obtained by analyzing the staining density of the entire nucleus and should be interpreted with caution because of the nonspecific sampling of all cells, fibers and neuropil. Using Fos expression in cell nuclei as a marker for functionally activated otolith-related neurons, we recently demonstrated that Fos/NR1 or Fos/NR2A/NR2B double-labeled neurons constituted 75% of the total Fosir neurons in the vestibular nuclei of adult rats [71]. A differential change in postnatal expression of NR2 subunits in Fos-ir vestibular nuclear neurons was reported [91]. In adult rodents, the functional properties of heteromeric AMPA receptors are highly dependent on the abundant expression of GluR2 receptor subunits [92–97] which have low Ca2+ permeability [96]. On the other hand, GluR1, GluR3 and GluR4 subunits have high Ca2+ permeability. In the 1st postnatal week, the functional properties of AMPA receptors in rat hippocampus neurons are mainly determined by GluR1 and GluR4 subunits [98, 99], which correspond to the occurrence of high Ca2+ influx in immature vestibular neurons. This also coincides with the absence of GluR2 subunits at P7 [10]. The progressive decrease in Ca2+ permeability of AMPA receptor channels with age [100, 101] is due to the gradual increase in the expression of GluR2 subunits in the vestibular nuclei [10]. In adult chinchillas, Popper et al. [102] revealed that large vestibular nuclear neurons expressed high levels of GluR2 mRNA but small neurons expressed widely varied levels. Most of the small neurons, especially those that were immunoreactive to -aminobutyric acid decarboxylase, expressed relatively low levels of GluR2, implying that small inhibitory vestibular nuclear neurons have high Ca2+ permeability. Since Ca2+ influx through activated AMPA receptor channels could Neuroembryol Aging 2004–05;3:162–170

165

control intracellular processes that lead to long-lasting changes in synaptic efficacy [103], vestibular nuclear neurons that possess AMPA receptors with high Ca2+ permeability are expected to play a key role in the plasticity of the vestibular system. Metabotropic Glutamate Receptors In addition to the family of ionotropic glutamate receptors, glutamate activates a large family of metabotropic glutamate receptors (mGluRs) that are coupled to second messenger systems through GTP-binding proteins [104, 105]. Group I mGluRs (mGluR1 and mGluR5) play roles in regulating neuronal excitability and in facilitating long-term potentiation and depression (LTP and LTD) at both pre- and postsynaptic levels [106, 107]. On the contrary, group II (mGluR2 and mGluR3) and group III (mGluR4, mGluR6, mGluR7 and mGluR8) are presynaptic autoreceptors controlling glutamate release [106, 108, 109]. In the medial vestibular nucleus of rats, the developmental shift from high-frequency, stimulation-induced LTD to LTP has been attributed to the distinct maturation profiles of glutamate receptors [110]. The progressive decrease in mGluR5-dependent LTD with postnatal development [110] is complemented by the late emergence of LTP that depends on NMDA receptor and mGluR1 [111–113]. Recent evidence from postnatal rats further revealed the different inhibitory mechanisms of mGluR2/mGlu3 and mGluR5 in regulating synaptic transmission and plasticity [9]. mGluR5 is involved in high-frequency, stimulation-induced vestibular LTD, whereas mGluR2/mGlu3 is involved in reducing glutamate release during synaptic overactivation. While this inhibitory influence decreases with maturation, it enhances the threshold for vestibular LTP induction in adults [113].

Postnatal Maturation of Properties of Central Vestibular Neurons

In chicks, spontaneously active neurons were observable in the tangential vestibular nucleus of brain slices from the day of hatching [114]. In brainstem slices of mammals, spontaneously active medial vestibular neurons with mono- or biphasic afterhyperpolarization, characterized as types A and B, respectively [115, 116], were distinguishable as early as P5 when action potentials were still in immature forms [117]. The intrinsic rhythmicity and excitability of type B neurons are regulated by Ca2+-activated K+ current. Furthermore, the afterhyper166

Neuroembryol Aging 2004–05;3:162–170

polarization amplitude of type A neurons progressively increased with postnatal development whereas the early fast afterhyperpolarization of type B neurons was not detectable until P15. Nevertheless, the contribution of these subtypes to the function of the canal and otolith systems is yet to be demonstrated. In adult animals, considerable proportions of the neurons within the vestibular nuclear complex receive convergent inputs from the semicircular canals and otolith end organs [118, 119]. In postnatal animals, the capacity of central vestibular neurons to integrate spatial information arising from the canals and otoliths remains elusive. Central Canal Neurons The postnatal properties of canal-related vestibular nuclear neurons have been studied and characterized in vivo [7, 120]. In very young rats (P4), some of these cells displayed random bursts of spikes. Their firing patterns changed from irregular to regular during the 1st postnatal month. Their resting discharge increased steadily from a very low rate in neonates to a high rate by the end of the 1st month. Likewise, the response sensitivity of these neurons to angular acceleration was low at birth and increased steadily, reaching the adult level, at the end of the 1st month. Central Otolith Neurons Using off-vertical axis rotation, Lai and Chan [6] provided evidence that otolith-related vestibular nuclear neurons of rats displayed position-dependent discharge modulation as early as P7. These otolith-related neurons showed irregular spontaneous activities and as the rats mature more regular activities were observed [6, 121]. A positive correlation between the level of spontaneous activity and the neuronal response gain to natural otolithic stimulation was observed after P14 [6]. In P7 rats, the majority of these otolith-related neurons showed clipped response (silenced in discharge) during parts of each 360° rotary cycle and the remaining neurons showed full-cycle response. The proportion of these clipped neurons decreased progressively from
75 to !25% as the animal matured [6, 122, 123]. In adult mammals, all head orientations on the horizontal plane are encoded within the vestibular nucleus as indicated by the best response orientations of the central otolith neurons [123–128]. In postnatal rats, however, our preliminary data showed that central otolith neurons tuned for anteroposterior direction were not well developed at the end of the 1st postnatal week [129]. This suggests restricted capability of neonatal rats in coding horLai /Tse /Shum /Chan

izontal head positions with respect to gravity [123]. Of particular relevance is the possible correlation between the spatial coding properties of central otolith neurons and motor behavior during postnatal development. While lateral movements of rats could be recorded as early as P2 [130], vertical movement was not observable until P9 when the rat could raise its head for brief periods in response to head-down tilt [131, 132]. These behavioral features therefore appear to correlate with the maturation profile of central otolith neurons in coding gravity-related spatial information.

cues in the guidance of axonal growth and pathfinding as well as pathway refinement during development of the vestibular system are not yet known. Likewise, the contributions of glutamate receptors and other determinants in regulating both the neural circuitry and neuronal plasticity during vestibular development are still controversial. Furthermore, exhaustive studies are required to delineate the plastic capacity of central neurons in coordinating the spatiotemporal information that arises from the semicircular canals and otolith organs as the brain matures.

Acknowledgments

Conclusion and Outlook

Although recent studies have significantly advanced our knowledge of the development of the vestibular system, several important issues await investigation. The

The works of the authors’ laboratories were supported by grants of the Hong Kong Research Grant Council (HKU7380/ 03M, HKU7370/04M and HKU7485/05M).

References 1 Altman J, Bayer SA: Development of the brain stem in the rat. III. Thymidine-radiographic study of the time of origin of neurons of the vestibular and auditory nuclei of the upper medulla. J Comp Neurol 1980; 194: 877–904. 2 Dechesne C, Mbiene JP, Sans A: Postnatal development of vestibular receptor surfaces in the rat. Acta Otolaryngol 1986;101:11–18. 3 Karhunen E: Postnatal development of the lateral vestibular nucleus (Deiters’ nucleus) of the rat. A light and electron microscopic study. Acta Otolaryngol Suppl 1973; 313: 1– 87. 4 Morris RJ, Beech JN, Heizmann CW: Two distinct phases and mechanisms of axonal growth shown by primary vestibular fibres in the brain, demonstrated by parvalbumin immunohistochemistry. Neuroscience 1988; 27:571–596. 5 Curthoys IS: The development of function of horizontal semicircular canal primary neurons in the rat. Brain Res 1979;167:41–52. 6 Lai CH, Chan YS: Spontaneous discharge and response characteristics of central otolith neurons of rats during postnatal development. Neuroscience 2001;103:275–288. 7 Lannou J, Precht W, Cazin C: The postnatal development of functional properties of central vestibular neurons in the rat. Brain Res 1979;175:219–232.

Embryonic and Postnatal Vestibular Development

8 Bolger C, Sansom AJ, Smith PF, Darlington CL: An antisense oligonucleotide to brainderived neurotrophic factor delays postural compensation following unilateral labyrinthectomy in guinea pig. Neuroreport 1999; 10:1485–1488. 9 Grassi S, Frondaroli A, Pettorossi VE: Role of group II metabotropic glutamate receptors 2/3 and group I metabotropic glutamate receptor 5 in developing rat medial vestibular nuclei. Neuroreport 2005;16:1303–1307. 10 Sans NA, Montcouquiol ME, Raymond J: Postnatal developmental changes in AMPA and NMDA receptors in the rat vestibular nuclei. Dev Brain Res 2000;123:41–52. 11 Vidal PP, Vibert N, Serafin M, Babalian A, Muhlethaler M, de Waele C: Intrinsic physiological and pharmacological properties of central vestibular neurons. Adv Otorhinolaryngol 1999;55:26–81. 12 Ruben RJ: Development of the inner ear of the mouse: a radioautographic study of terminal mitoses. Acta Otolaryngol Suppl 1967; 220:1–31. 13 Denman-Johnson K, Forge A: Establishment of hair bundle polarity and orientation in the developing vestibular system of the mouse. J Neurocytol 1999;28:821–835. 14 Mbiene JP, Favre D, Sans A: The pattern of ciliary development in fetal mouse vestibular receptors. Anat Embryol 1984; 170: 229– 238. 15 Nordemar H: Postnatal development of the vestibular sensory epithelium in the mouse. Acta Otolaryngol 1983;95:447–456.

16 Desmadryl G, Sans A: The development of the innervation pattern in the mouse vestibular sensory epithelium. Dev Brain Res 1990; 52:183–189. 17 Eatock RA, Hurley KM: Functional development of hair cells. Curr Top Dev Biol 2003; 57:389–448. 18 Rusch A, Eatock RA: A delayed rectifier conductance in type I hair cells of the mouse utricle. J Neurophysiol 1996;76:995–1004. 19 Rusch A, Lysakowski A, Eatock RA: Postnatal development of type I and type II hair cells in the mouse utricle: acquisition of voltage-gated conductance and differentiated morphology. J Neurosci 1998; 18:7487–7501. 20 Geleoc GS, Risner JR, Holt JR: Developmental acquisition of voltage-dependent conductances and sensory signaling in hair cells of the embryonic mouse inner ear. J Neurosci 2004;24:11148–11159. 21 Lysakowski A: Development of synaptic innervation in the rodent utricle. Ann NY Acad Sci 1999;871:422–425. 22 Mbiene JP, Favre D, Sans A: Early innervation and differentiation of hair cells in the vestibular epithelia of mouse embryo: SEM and TEM study. Anat Embryol 1988; 177: 331–340. 23 Van de Water TR, Anniko M, Nordemar H, Wersäll J: Development of the sensory receptor cells in the utricular macula. Acta Otolaryngol 1978;87:297–305.

Neuroembryol Aging 2004–05;3:162–170

167

24 Dechesne CJ: The development of vestibular sensory organs in human; in Romand R (ed): Development of Auditory and Vestibular Systems 2. Amsterdam, Elsevier, 1992, pp 419–447. 25 Sans A, Scarfone E: Afferent calyces and type I hair cells during development. A new morphofunctional hypothesis. Ann NY Acad Sci 1996;781:1–12. 26 Nordemar H: Embryogenesis of the inner ear. II. The late differentiation of the mammalian crista ampullaris in vivo and in vitro. Acta Otolaryngol 1983;96:1–8. 27 Dememes D, Broca C: Calcitonin gene-related peptide immunoreactivity in the rat efferent vestibular system during development. Dev Brain Res 1998;108:59–67. 28 Dememes D, Seoane A, Venteo S, Desmadryl G: Efferent function of vestibular afferent endings? Similar localization of N-type calcium channels, synaptic vesicle and synaptic membrane-associated proteins. Neuroscience 2000;98:377–384. 29 Dememes D, Dechesne CJ, Venteo S, Gaven F, Raymond J: Development of the rat efferent vestibular system on the ground and in microgravity. Dev Brain Res 2001; 128: 35– 44. 30 Boyle R, Highstein SM: Efferent vestibular system in the toadfish: action upon horizontal semicircular canal afferents. J Neurosci 1990;10:1570–1582. 31 Brichta AM, Goldberg JM: Afferent and efferent responses from morphological fiber classes in the turtle posterior crista. Ann NY Acad Sci 1996;781:183–195. 32 Goldberg JM, Fernandez C: Efferent vestibular system in squirrel monkey: anatomical location and influence on afferent activity. J Neurophysiol 1980; 43:986–1025. 33 Lai CH, Chan YS: Development of the vestibular system. Neuroembryology 2002; 1: 61–71. 34 Clarac F, Vinay L, Cazalets JR, Fady JC, Jamon M: Role of gravity in the development of posture and locomotion in the neonatal rat. Brain Res Rev 1998;28:35–43. 35 Dechesne C, Raymond J, Sans A: The efferent vestibular system in the cat: a horseradish peroxidase and fluorescent retrograde tracers study. Neuroscience 1984; 11: 893– 901. 36 Li C, Zhang YK, Guan ZL, Shum DK, Chan YS: Vestibular afferent innervation in the vestibular efferent nucleus of rats. Neurosci Lett 2005;385:36–40. 37 Fritzsch B, Pirvola U, Ylikoski J: Making and breaking the innervation of the ear: neurotrophic support during ear development and its clinical implications. Cell Tissue Res 1999;295:369–382. 38 Carney PR, Couve E: Cell polarity changes and migration during early development of the avian peripheral auditory system. Anat Rec 1989;225:156–164.

168

39 Fode C, Gradwohl G, Morin X, Dierich A, LeMeur M, Goridis C, Guillemot F: The bHLH protein NEUROGENIN 2 is a determination factor for epibranchial placode-derived sensory neurons. Neuron 1998; 20: 483–494. 40 Ma Q, Anderson DJ, Fritzsch B: Neurogenin 1 null mutant ears develop fewer, morphologically normal hair cells in smaller sensory epithelia devoid of innervation. J Assoc Res Otolaryngol 2000;1:129–143. 41 Fritzsch B: Development of inner ear afferent connections: forming primary neurons and connecting them to the developing sensory epithelia. Brain Res Bull 2003;60:423–433. 42 Liu M, Pereira FA, Price SD, Chu MJ, Shope C, Himes D, Eatock RA, Brownell WE, Lysakowski A, Tsai MJ: Essential role of BETA2/ NeuroD1 in development of the vestibular and auditory systems. Genes Dev 2000; 14: 2839–2854. 43 Morin X, Cremer H, Hirsch MR, Kapur RP, Goridis C, Brunet JF: Defects in sensory and autonomic ganglia and absence of locus coeruleus in mice deficient for the homeobox gene Phox2a. Neuron 1997;18:411–423. 44 Ryan AF: Transcription factors and the control of inner ear development. Semin Cell Dev Biol 1997;8:249–256. 45 Xiang M, Gan L, Zhou L, Klein WH, Nathans J: Targeted deletion of the mouse POU domain gene Brn-3a causes selective loss of neurons in the brainstem and trigeminal ganglion, uncoordinated limb movement, and impaired suckling. Proc Natl Acad Sci USA 1996;93:11950–11955. 46 Huang EJ, Reichardt LF: Neurotrophins: roles in neuronal development and function. Annu Rev Neurosci 2001;24:677–736. 47 Pirvola U, Ylikoski J, Palgi J, Lehtonen E, Arumae U, Saarma M: Brain-derived neurotrophic factor and neurotrophin 3 mRNAs in the peripheral target fields of developing inner ear ganglia. Proc Natl Acad Sci USA 1992;89:9915–9919. 48 Pirvola U, Arumae U, Moshnyakov M, Palgi J, Saarma M, Ylikoski J: Coordinated expression and function of neurotrophins and their receptors in the rat inner ear during target innervation. Hear Res 1994;75:131–144. 49 Wheeler EF, Bothwell M, Schecterson LC, von Bartheld CS: Expression of BDNF and NT-3 mRNA in hair cells of the organ of Corti: quantitative analysis in developing rats. Hear Res 1994;73:46–56. 50 Schecterson LC, Bothwell M: Neurotrophin and neurotrophin receptor mRNA expression in developing inner ear. Hear Res 1994; 73:92–100. 51 Ernfors P, Lee KF, Jaenisch R: Mice lacking brain-derived neurotrophic factor develop with sensory deficits. Nature 1994; 368: 147– 150. 52 Ernfors P, Van De Water T, Loring J, Jaenisch R: Complementary roles of BDNF and NT-3 in vestibular and auditory development. Neuron 1995;14:1153–1164.

Neuroembryol Aging 2004–05;3:162–170

53 Hashino E, Dolnick RY, Cohan CS: Developing vestibular ganglion neurons switch trophic sensitivity from BDNF to GDNF after target innervation. J Neurobiol 1999;38:414– 427. 54 Agerman K, Hjerling-Leffler J, Blanchard MP, Scarfone E, Canlon B, Nosrat C, Ernfors P: BDNF gene replacement reveals multiple mechanisms for establishing neurotrophin specificity during sensory nervous system development. Development 2003; 130: 1479– 1491. 55 Schimmang T, Minichiello L, Vazquez E, San Jose I, Giraldez F, Klein R, Represa J: Developing inner ear sensory neurons require TrkB and TrkC receptors for innervation of their peripheral targets. Development 1995; 121:3381–3391. 56 Bianchi LM, Conover JC, Fritzsch B, DeChiara T, Lindsay RM, Yancopoulos GD: Degeneration of vestibular neurons in late embryogenesis of both heterozygous and homozygous BDNF null mutant mice. Development 1996;122:1965–1973. 57 Ylikoski J, Pirvola U, Moshnyakov M, Palgi J, Arumae U, Saarma M: Expression patterns of neurotrophin and their receptor mRNAs in the rat inner ear. Hear Res 1993; 65: 69– 78. 58 Vaage S: The segmentation of the primitive neural tube in chicken embryo (Gallus domesticus). Adv Anat Embryol Cell Biol 1969; 41:1–88. 59 Lumsden A: The cellular basis of segmentation in the developing hindbrain. Trend Neurosci 1990;13:329–335. 60 Gilland E, Baker R: Conservation of neuroepithelial and mesodermal segments in the embryonic vertebrate head. Acta Anat 1993; 148:110–123. 61 Kuruvilla A, Sitko S, Schwartz IR, Honrubia V: Central projections of primary vestibular fibers in the bullfrog. I. The vestibular nuclei. Laryngoscope 1985;95:692–707. 62 Matesz C, Birinyi A, Straka H, Dieringer N: Location of dye-coupled second order canal and otolith neurons and of efferent vestibular neurons in the frog. Neurobiology 1998; 6:226–227. 63 Maklad A, Fritzsch B: Development of vestibular afferent projections into the hindbrain and their central targets. Brain Res Bull 2003;60:497–510. 64 Straka H, Baker R, Gilland E: Rhombomeric organization of vestibular pathways in larval frogs. J Comp Neurol 2001;437:42–55. 65 Tashiro Y, Miyahara M, Shirasaki R, Okabe M, Heizmann CW, Murakami F: Local nonpermissive and oriented permissive cues guide vestibular axon to the cerebellum. Development 2001;128:973–981. 66 Kawauchi D, Kobayashi H, Sekine-Aizawa Y, Fujita SC, Murakami F: MuSC is involved in regulating axonal fasciculation of mouse primary vestibular afferents. Eur J Neurosci 2003;18:2244–2252.

Lai /Tse /Shum /Chan

67 Lau WK, Kwok JC, Ng TK, Chan YS, Shum DK: Developing vestibular commissural projections display differential patterns with chondroitinase treatment of the hindbrain. Soc Neurosci Abstr 2005, abstract 481.8. 68 Tiveron MC, Pattyn A, Hirsch MR, Brunet JF: Role of Phox2b and Mash1 in the generation of the vestibular efferent nucleus. Dev Biol 2003; 260:46–57. 69 Lai CH, Tse YC, Shum DK, Yung KK, Chan YS: Postnatal expression of Fos in otolith-related brainstem neurons of rats following off-vertical axis rotation. J Comp Neurol 2004;470:282–296. 70 Lai SK, Lai CH, Yung KK, Shum DK, Chan YS: Postnatal expression of Fos in otolith-related brainstem neurons of rats following off-vertical axis rotation. Eur J Neurosci 2006;23:2431–2446. 71 Chen LW, Lai CH, Law HY, Yung KK, Chan YS: Quantitative study of the coexpression of Fos and N-methyl-D aspartate (NMDA) receptor subunits in otolith-related vestibular nuclear neurons of rats. J Comp Neurol 2003; 460:292–301. 72 Zhang FX, Lai CH, Tse YC, Shum DK, Chan YS: Expression of Trk receptors in otolithrelated neurons in the vestibular nucleus of rats. Brain Res 2005;1062:92–100. 73 de Waele C, Muhlethaler M, Vidal PP: Neurochemistry of the central vestibular pathways. Brain Res Rev 1995;20:24–46. 74 Cochran SL, Kasik P, Precht W: Pharmacological aspects of excitatory synaptic transmission to second-order vestibular neurons in the frog. Synapse 1987;1:102–123. 75 Lewis MR, Phelan KD, Shinnick-Gallagher P, Gallagher JP: Primary afferent excitatory transmission recorded intracellularly in vitro from rat medial vestibular neurons. Synapse 1989;3:149–153. 76 Kinney GA, Peterson BW, Slater NT: The synaptic activation of N-methyl-D-aspartate receptors in the rat medial vestibular nucleus. J Neurophysiol 1994; 72:1588–1595. 77 Knopfel T: Evidence for N-methyl-D-aspartic acid receptor-mediated modulation of the commissural input to central vestibular neurons of the frog. Brain Res 1987; 426: 212– 224. 78 Straka H, Debler K, Dieringer N: Size-related properties of vestibular afferent fibers in the frog: differential synaptic activation of Nmethyl-D-aspartate and non-N-methyl-Daspartate receptors. Neuroscience 1996; 70: 697–707. 79 Takahashi Y, Takahashi MP, Tsumoto T, Doi K, Matsunaga T: Synaptic input-induced increase in intraneuronal Ca 2+ in the medial vestibular nucleus of young rats. Neurosci Res 1994;21:59–69. 80 Vidal PP, Babalian A, de Waele C, Serafin M, Vibert N, Muhlethaler M: NMDA receptors of the vestibular nuclei neurones. Brain Res Bull 1996;40:347–352.

Embryonic and Postnatal Vestibular Development

81 Doi K, Tsumoto T, Matsunaga T: Actions of excitatory amino acid antagonists on synaptic inputs to the rat medial vestibular nucleus: an electrophysiological study in vitro. Exp Brain Res 1990;82:254–262. 82 Sakai N, Ujihara H, Ishihara K, Sasa M, Tanaka C: Electrophysiological and pharmacological characteristics of ionotropic glutamate receptors in medial vestibular nucleus neurons: a whole cell patch clamp study in acutely dissociated neurons. Jpn J Pharmacol 1996;72:335–346. 83 Chan YS, Chen LW, Lai CH, Shum DK, Yung KK, Zhang FX: Receptors of glutamate and neurotrophin in vestibular neuronal functions. J Biomed Sci 2003; 10:577–587. 84 Chen LW, Yung KK, Chan YS: Co-localization of NMDA receptors and AMPA receptors in neurons of the vestibular nuclei of rats. Brain Res 2000;884:87–97. 85 Chan YS, Lai CH, Shum DK: Spatial coding capacity of central otolith neurons. Exp Brain Res 2006;173:205–214. 86 Shao M, Hirsch JC, Giaume C, Peusner KD: Spontaneous synaptic activity in chick vestibular nucleus neurons during the perinatal period. Neuroscience 2004;127:81–90. 87 Shao M, Hirsch JC, Peusner KD: Emergence of action potential generation and synaptic transmission in vestibular nucleus neurons. J Neurophysiol 2006; 95:1215–1226. 88 Wenzel W, Fritschy JM, Mohler H, Benke D: NMDA receptor heterogeneity during postnatal development of the rat brain: differential expression of the NR2A, NR2B, and NR2C subunit proteins. J Neurochem 1997; 68:469–478. 89 Ebralidze AK, Rossi DJ, Tonegawa S, Slater NT: Modification of NMDA receptor channels and synaptic transmission by targeted disruption of the NR2C gene. J Neurosci 1996;16:5014–5025. 90 Takahashi T, Feldmeyer D, Suzuki N, Onodera K, Cull-Candy SG, Sakimura K, Mishina M: Functional correlation of NMDA receptor subunits expression with the properties of single-channel and synaptic currents in the developing cerebellum. J Neurosci 1996;16:4376–4382. 91 Tse YC, Lai SK, Zhou H, Yung WH, Chan YS: AMPA and NMDA receptors properties in spinal vestibular neurons of developing rats. Neurosci Lett, in press. 92 Bettler B, Mulle C: Review: neurotransmitter receptors. II. AMPA and kainate receptors. Neuropharmacology 1995; 34:123–139. 93 Dingledine R, Borges K, Bowie D, Traynelis SF: The glutamate receptor ion channels. Pharmacol Rev 1999;51:7–61. 94 Geiger JR, Melcher T, Koh DS, Sakmann B, Seeburg PH, Jonas P, Monyer H: Relative abundance of subunit mRNAs determines gating and Ca 2+ permeability of AMPA receptors in principal neurons and interneurons in rat CNS. Neuron 1995;15:193–204.

95 Hollmann M, Heinemann S: Cloned glutamate receptors. Annu Rev Neurosci 1994; 17:31–108. 96 Jonas P, Burnashev N: Molecular mechanisms controlling calcium entry through AMPA-type glutamate receptor channels. Neuron 1995;15:987–990. 97 Myers SJ, Dingledine R, Borges K: Genetic regulation of glutamate receptor ion channels. Annu Rev Pharmacol Toxicol 1999;39: 221–241. 98 Shi S, Hayashi Y, Esteban JA, Malinow R: Subunit-specific rules governing AMPA receptor trafficking to synapses in hippocampal pyramidal neurons. Cell 2001; 105: 331–343. 99 Zhu JJ, Esteban JA, Hayashi Y, Malinow R: Postnatal synaptic potentiation: delivery of GluR4-containing AMPA receptors by spontaneous activity. Nat Neurosci 2000;3: 1098–1106. 100 Durand GM, Zukin RS: Developmental regulation of mRNAs encoding rat brain kainate/AMPA receptors: a northern analysis study. J Neurochem 1993; 61: 2239– 2246. 101 Pellegrini-Giampietro DE, Bennett MV, Zukin RS: Are Ca 2+-permeable kainate/ AMPA receptors more abundant in immature brain? Neurosci Lett 1992;144:65–69. 102 Popper P, Rodrigo JP, Alvarez JC, Lopez I, Honrubia V: Expression of the AMPA-selective receptor subunits in the vestibular nuclei of the chinchilla. Mol Brain Res 1997;44:21–30. 103 Jonas P, Racca C, Sakmann B, Seeburg PH, Monyer H: Differences in Ca 2+ permeability of AMPA-type glutamate receptor channels in neocortical neurons caused by differential GluR-B subunit expression. Neuron 1994;12:1281–1289. 104 Conn PJ, Pin JP: Pharmacology and functions of metabotropic glutamate receptors. Annu Rev Pharmacol Toxicol 1997;37:205– 237. 105 Pin JP, Duvoisin R: The metabotropic glutamate receptors: structure and functions. Neuropharmacology 1995;34:1–26. 106 Anwyl R: Metabotropic glutamate receptors: electrophysiological properties and role in plasticity. Brain Res Rev 1999; 29: 83–120. 107 Schwartz NE, Alford S: Physiological activation of presynaptic metabotropic glutamate receptors increases intracellular calcium and glutamate release. J Neurophysiol 2000;84:415–427. 108 Bradley SR, Marino MJ, Wittmann M, Rouse ST, Awad H, Levey AI, Conn PJ: Activation of group II metabotropic glutamate receptors inhibits synaptic excitation of the substantia nigra pars reticulate. J Neurosci 2000;20:3085–3094. 109 Dube GR, Marshall KC: Activity-dependent activation of presynaptic metabotropic glutamate receptors in locus coeruleus. J Neurophysiol 2000; 83:1141–1149.

Neuroembryol Aging 2004–05;3:162–170

169

110 Puyal J, Grassi S, Dieni C, Frondaroli A, Dememes D, Raymond J, Pettorossi VE: Developmental shift from long-term depression to long-term potentiation in the rat medial vestibular nuclei: role of group I metabotropic glutamate receptors. J Physiol 2003;553:427–443. 111 Capocchi G, Della Torre G, Grassi S, Petorossi VE, Zampolini M: NMDA receptormediated long term modulation of electrically evoked field potentials in the rat medial vestibular nuclei. Exp Brain Res 1992;90:546–550. 112 Grassi S, Pettorossi VE: Synaptic plasticity in the medial vestibular nuclei: role of glutamate receptors and retrograde messengers in rat brainstem slices. Prog Neurobiol 2001;64:527–553. 113 Grassi S, Frondaroli A, Pettorossi VE: Different metabotropic glutamate receptors play opposite roles in synaptic plasticity of the rat medial vestibular nuclei. J Physiol 2002;543:795–806. 114 Shao M, Hirsch JC, Peusner KD: Maturation of firing pattern in chick vestibular nucleus neurons. Neuroscience 2006; 141: 711–726. 115 Serafin M, de Waele C, Khateb A, Vidal PP, Muhlethaler M: Medial vestibular nucleus in the guinea-pig. I. Intrinsic membrane properties in brainstem slices. Exp Brain Res 1991;84:417–425. 116 Johnston AR, MacLeod NK, Dutia MB: Ionic conductances contributing to spike repolarization and after-potentials in rat medial vestibular nucleus neurones. J Physiol 1994;481:61–77.

170

117 Dutia MB, Johnston AR: Development of action potentials and apamin-sensitive after-potentials in mouse vestibular nucleus neurones. Exp Brain Res 1998; 118: 148– 154. 118 Angelaki DE, Dickman JD: Spatiotemporal processing of linear acceleration: primary afferent and central vestibular neuron responses. J Neurophysiol 2000; 84: 2113– 2132. 119 Uchino Y, Sasaki M, Sato H, Bai R, Kawamoto E: Otolith and canal integration on single vestibular neurons in cats. Exp Brain Res 2005;164:271–285. 120 Lannou J, Cazin L, Hamann KF: Responses of central vestibular neurons to horizontal linear acceleration in the rat. Pflügers Arch 1980;385:123–129. 121 Lai CH, Chan YS: Spontaneous activity of otolith-related vestibular nuclear neurons in the decerebrate rat. Brain Res 1996; 739: 322–329. 122 Chan YS, Lai CH, Shum DK: Bilateral otolith contribution of spatial coding in the vestibular system. J Biomed Sci 2002; 9: 574–586. 123 Lai CH, Chan YS: Properties of otolith-related vestibular nuclear neurons in response to bidirectional off-vertical axis rotation of the rat. Brain Res 1995; 693: 39– 50. 124 Chan YS: The coding of head orientations in neurons of bilateral vestibular nuclei of cats after unilateral labyrinthectomy: response to off-vertical axis rotation. Exp Brain Res 1997;114:293–303.

Neuroembryol Aging 2004–05;3:162–170

125 Chan YS, Cheung YM, Hwang JC: Effect of tilt on the response of neuronal activity within the cat vestibular nuclei during slow and constant velocity rotation. Brain Res 1985;345:271–278. 126 Chan YS, Cheung YM, Hwang JC: Response characteristics of neurons in the cat vestibular nuclei during slow and constant velocity off-vertical axes rotations in the clockwise and counterclockwise directions. Brain Res 1987;406:294–301. 127 Chan YS, Shum DK, Lai CH: Neuronal response sensitivity to bidirectional off-vertical axis rotations: a dimension of imbalance in the lateral vestibular nuclei of cats after unilateral labyrinthectomy. Neuroscience 1999;94:831–843. 128 Chan YS, Lai CH, Shum DK: Response properties of Y group neurons to crossed otolith inputs in the cat. Neuroreport 2003; 14:729–733. 129 Lai CH, Chan YS: Signaling capacity of peripheral and central otolith neurons in coding horizontal head movement in postnatal rats. J Vestib Res 2004;14:262–263. 130 Geisler HC, Westerga A, Gramsbergen A: Development of posture in the rat. Acta Neurobiol Exp (Wars) 1993;53:517–523. 131 Petrosini L, Molinari M, Gremoli T: Hemicerebellectomy and motor behaviour in rats. I. Development of motor function after neonatal lesion. Exp Brain Res 1990;82: 472–482. 132 Brocard F, Vinay L, Clarac F: Development of hindlimb postural control during the first postnatal week in the rat. Dev Brain Res 1999;117:81–89.

Lai /Tse /Shum /Chan

Neuroembryol Aging 2004–05;1:171–182 DOI: 10.1159/000096795

Published online: November 6, 2006

What Peripheral Vestibular Manipulations Reveal about the Function and Plasticity in the Primate Oculomotor System Shawn D. Newlands a Dora E. Angelaki b a b

Department of Otolaryngology, University of Texas Medical Branch, Galveston, Tex., and Department of Anatomy and Neurobiology, Washington University, St. Louis, Mo., USA

Key Words Electrical stimulation
Eye movement
Labyrinthectomy
Lesion
Primate
Vestibulo-ocular reflex

Abstract Central to our understanding of the function and dysfunction of the vestibular system are results from experiments where distinct manipulations of the peripheral vestibular apparatus have been used to characterize deficits and recovery of neural and reflexive responses. Using the primate oculomotor system as a focus, we summarize here recent advances where peripheral vestibular manipulations have revealed important properties of vestibular function and adaptability. Copyright © 2005 S. Karger AG, Basel

Peripheral Manipulations: Their Physiological Effects and Limitations

Introduction

Vestibular signals, often coupled with other extravestibular cues, contribute to the maintenance of visual acuity, balance, and spatial orientation. Being relatively well preserved throughout evolution [1], vestibular dysfunction leads to deficits in vestibular reflexes and equilibrium. Because of their frontally placed eyes and anatomical/functional similarities, nonhuman primates are ideal as an animal model for the treatment of vestibular-re-

© 2005 S. Karger AG, Basel Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

lated diseases in humans. A unique advantage of nonhuman primates is the ability to manipulate the peripheral vestibular system. Because of the strong multisensory nature of most balance functions, it is often difficult to dissociate, either psychophysically or by other non-invasive methods, contributions of various peripheral components to central functions. For this reason, the ability to selectively activate or inhibit either canal or otolith signals and regular or irregular afferents (unilaterally or bilaterally) has proven efficacious in understanding fundamental aspects of vestibular physiology. Here we summarize both the advantages and disadvantages of the various methods of peripheral vestibular manipulations and highlight recent important contributions to our understanding of vestibular function.

Accessible online at: www.karger.com/nba

Labyrinthectomy Labyrinthectomy is the best-studied peripheral vestibular lesion. It involves either the chemical or surgical destruction of the vestibular labyrinth. Complete labyrinthectomy involves destruction or removal of all of the vestibular neuroepithelium. Without the vestibular neuroepithelium, the eighth-nerve vestibular afferents lack their sole source of excitatory drive and are silenced.

Dr. Dora Angelaki Department of Anatomy and Neurobiology, Box 8108 Washington University School of Medicine, 660 South Euclid Avenue St. Louis, MO 63110 (USA) Tel. +1 314 747 5529, Fax +1 314 747 4370, E-Mail [email protected]

Selective Destruction of End Organs or Nerves Selective sectioning of nerves to semicircular ampullae, particularly the lateral canal and the posterior canal, has been successfully performed in a number of species including monkeys [2–5]. The main advantage of this technique is the selective lesion of a single nerve or nerve branch. In fact, these lesions are interpreted as equivalent to selective labyrinthectomy, and they are characterized by both dynamic and static deficits. Care must be taken in interpreting findings from such studies, as careful controls must be performed to assure that other portions of the labyrinth have not been injured. In the primate, it has been particularly problematic to lesion the utricle selectively. Such selective lesions have been demonstrated by Igarashi et al. [6] in squirrel monkeys. This technique involves an intralabyrinthine approach, and thus concern about damage to the membranous labyrinth must be considered when interpreting results using such methods. Semicircular Canal Plugging Mechanical occlusion of the semicircular canals involves mechanically blocking endolymph motion in the membranous labyrinth, generally through placement of a bony or wax plug through fenestration in bony labyrinth [7, 8]. This lesion is selective for the semicircular canal that is plugged [9]. Theoretically, this lesion does not result in damage of the neuroepithelium of the semicircular canals, and canal plugging preserves resting activity in the vestibular nerve [10]. Practically, the absence of nystagmus at rest after unilateral canal plugging (UCP) supports the notion that the neuroepithelium is not disrupted by the procedure, and anatomical data support the idea that cupular function is likewise preserved after the procedure [10, 11]. While canal plugging was previously accepted as a complete deactivation of canal functionality, careful studies in the toadfish, and subsequently other animals, have demonstrated both in practice and mathematically that occlusion of a semicircular canal does not completely inactivate the canal. The ability of the plugged semicircular canal to transduce head motion is compromised only at low frequencies [11, 12]. Indeed the dynamics of the plugged semicircular canal are changed such that the canal is functional at high frequencies but with substantially increased phase leads [12]. The particular dynamics of the canal after plugging depend upon a number of geometric factors including the radius of the canal, the diameter of the canal lumen, and the stiffness of the round window and bony labyrinth [12]. Theory predicts that there would be a 190% attenuation of rotational vestibulo-ocular reflex (RVOR) in rhesus 172

Neuroembryol Aging 2004–05;3:171–182

monkeys at frequencies below 1 Hz but that attenuation above 1 Hz would be less. These predictions have been proven experimentally [11, 13–15]. Electrical Stimulation Constant galvanic stimulation of the peripheral labyrinth can relatively selectively activate or inhibit the most irregularly firing vestibular afferents [16–18]. These findings have enabled a number of studies of the vestibular system during the application of such galvanic stimulation to discern the effects of either regular or irregular afferents in normal vestibular reflexes [16, 19] and their relative input to central vestibular neurons and pathways [18, 20]. As with canal plugging, the selective ablation of irregular afferents by hyperpolarization represents a functional inactivation (rather than a classic lesion). While classified as regular, irregular, or intermediate, vestibular afferents represent a continuum reflected in fiber size, firing regularity, sensitivity to rotation or translation, and sensitivity to electrical current [17, 21–23]. Despite the limitation in the specificity of the galvanic inhibition of irregular afferents, the effect of such galvanic ‘ablation’ of irregular afferents on central vestibular and vestibule-ocular function can provide information that is difficult to discern in other ways. Gentamicin Transtympanic application of buffered gentamicin solution to the middle ear cleft has become commonplace in otologic practice as a measure to control dizziness associated with Ménière’s disease that is refractory to more conservative measures [24]. There has been a great deal of research concerning the exact mechanism whereby gentamicin affects the vestibular periphery. While the exact mechanisms and cellular targets of aminoglycoside antibiotics are not known, these drugs have been demonstrated to induce apoptosis in vestibular hair cells [25]. It has been well demonstrated in chinchillas that gentamicin application to the middle ear selectively destroys type 1 hair cells in preference to type 2 hair cells in the ampulla [26]. Additionally, it is believed that the remaining type 2 hair cells are relatively ineffective in transducing rotational stimuli to the nerve [26]. However, there appears to be continued release of neurotransmitter from these remaining but damaged hair cells, as evidenced by continuation of spontaneous firing in the vestibular nerve. Thus, another mechanism of action for gentamicin effects might be disruption of transduction mechanisms in preserved but ineffective hair cells [26]. To our Newlands /Angelaki

knowledge, gentamicin has yet to be utilized in primates, but one expects that peripheral gentamicin effects in primates will be investigated in the future.

Rotational Vestibulo-Ocular Reflex The behavioral effects of acute labyrinthectomy have been documented in numerous species, including primates. In one of the early pioneering studies of macaque deficits following UL, Fetter and Zee [27] reported that RVOR gains recovered from approximately 0.5 acutely after the lesion to near normal values during low velocity rotations. During high velocity rotations, gains were lower, particularly towards the side of the lesion. The recovery was fast during the first few days after the lesion but slower thereafter. In general, the behavioral deficits documented after labyrinthine lesion can be divided into two groups, static and dynamic, based upon whether they are observed with the subject at rest (static signs, such as spontaneous nystagmus) or are only revealed with head movement (dynamic signs, such as asymmetric RVOR gain). Static signs are believed to be related to asymmetry in the resting activity between central neurons populating the bilateral vestibular nuclei. These signs recover more quickly and more completely than do the dynamic signs, presumably through the restoration of balance in neuronal discharge bilaterally.

In contrast, dynamic signs are believed to be a manifestation of asymmetric inputs for ipsilesional versus contralesional rotation. Although each vestibular nerve can respond to inhibitory and excitatory rotational directions through modulation of the resting rate, the responses of primary afferents are asymmetric favoring excitation (Ewald’s second law). Additionally, many afferents, particularly sensitive ones, rectify or silence in the off direction, such that at high accelerations the amount of data in the inhibitory direction is further reduced. Thus, rotation of an animal with a unilateral peripheral lesion will provide asymmetric information to the central nervous system. There may also be central mechanisms that filter the signal in such a way that high frequency information is lost through the commissural system [28]. For these reasons, it is not surprising that the RVOR shows both a reduction in overall gain and develops asymmetry after a unilateral labyrinthine lesion. Fetter et al. [29] reported that the reduction in spontaneous nystagmus seen in the first few days after UL also occurred in monkeys who recovered in the dark or monkeys who were previously blinded by occipital lobectomy. However, compensation of RVOR gain was dependent upon cortical vision, as monkeys blinded by occipital lobectomy after compensation lost their RVOR gain improvement. It might be then expected that compensation in canalplugged animals would be identical to that in labyrinthectomized animals, absent the static signs. In primates, several studies allow comparison between UCP versus UL compensation. For low acceleration sinusoidal and step rotations, symmetric, nearly complete recovery of the RVOR is seen after both UL [27, 30, 31] and UCP [10, 14]. In particular, using high-frequency (up to 15 Hz) but low-velocity (20°/s) sinusoidal stimuli and equivalent step stimuli, Lasker et al. [14, 31] found few differences between compensation in UCP and UL in squirrel monkeys. The primary finding was a slower recovery of VOR gain and symmetry in UL compared to UCP animals for velocity step (impulse) stimuli. Figure 1 shows averaged responses for step rotation towards and away from the side of lesion 1 and 10 days after UCP. The asymmetry in the responses is greater on day 1 but improved by day 10. These manipulations allowed the authors to define two afferent pathways with different involvement in vestibular compensation. A linear pathway recovered for both ipsilesional and contralesional rotation at low velocities, while a non-linear pathway mainly recovered during high acceleration and velocity stimuli for contralesional rotations. In contrast, for ipsilesional rotation,

Primate Vestibular System

Neuroembryol Aging 2004–05;3:171–182

Bilateral versus Unilateral Lesions All of the previously mentioned lesions can be applied either bilaterally or unilaterally. In general, unilateral lesions have been more commonly utilized when one is interested in recovery over time. For example, unilateral labyrinthectomy (UL) or UCP would be used to investigate how the system recovers from unilateral pathology. UCP and UL have been the predominant models for looking at plasticity in the vestibular system as one considers adaptation from a bilaterally balanced system to information obtained from only one side. Unilateral lesions have also revealed information about which processes require input from both labyrinths. Bilateral lesions, on the other hand, are commonly utilized to completely eliminate vestibular function. For instance, bilateral labyrinthectomy is the most common model for complete loss of rotary and translational signals. In the next sections, we summarize results from recent primate VOR and gaze shift studies following bilateral and UCP and UL lesions.

Vestibular Compensation

173

Horizontal eye velocity (º/s)

movement responses to yaw velocity step impulse stimuli (whole body rotation) in a squirrel monkey. a, b Day 1 responses (animal kept in the dark) after plugging of the three semicircular canals on one side. c, d Day 10 responses (9 days in light) after the plugging. a, c Rotations towards the plugged side (ipsilesional). b, d Contralesional rotations. Each panel shows the average (white line) and one standard deviation (shaded area) of 10 trials. Fast-phase (saccadic) eye movements were removed. Dashed line: head velocity. The horizontal eye velocity, which is in the compensatory direction, is shown inverted for ease of comparison to the head velocity in the figure. Modified and used with permission from Lasker et al. [14].

50

150

0

100

–50

50

–100

0

–150

–50

–200 0.2

0.3

a

Horizontal eye velocity (º/s)

Fig. 1. Effects of UCP on the RVOR. Eye

200

0.5

0.6

0.7

0.2

0.3

b

Time (s)

0.4

0.5

0.6

0.7

0.6

0.7

Time (s)

200

50

150

0

100

–50

50

–100

0

–150 –200

–50 0.2

0.3

c

the non-linear pathway (arising from the intact labyrinth) was silenced and did not contribute to compensation. This lack of recovery accounts for the persistent, uncompensated asymmetries seen in both UCP and UL preparations with high-acceleration (13,000°/s2) rotations. Paige [10] reported similar results with animals that only had the lateral canal plugged, though his study focused on low-frequency stimulation (!4 Hz). Both Paige [10] and previous studies [30, 32] have also reported long-lasting gain saturation for ipsilesional rotation at high rotational velocities (1240°/s) and substantial increases in the low-frequency RVOR phase leads. As a result, the RVOR time constant was shortened after labyrinthectomy and recovered little over time [27]. The velocity storage component of the optokinetic system was also impaired by labyrinthectomy but partially recovered over time. A number of behavioral studies, including those discussed here, have led to predictions about the behavior of central neurons in compensated monkeys [14, 25, 27, 29, 31, 33–36]. None of these models of the central mechanisms of compensation have been tested directly at the cellular level in primates. Studies in anesthetized guinea pigs [37–39], decerebrate gerbils [40–42], anesthetized 174

0.4

Neuroembryol Aging 2004–05;3:171–182

0.4

0.5

Time (s)

0.6

0.2

0.7

d

0.3

0.4

0.5

Time (s)

rats [43], cats [28], and awake guinea pigs [44] have revealed common patterns of recovery of neuronal responses in the vestibular nuclei at various stages of compensation, but data in well-characterized primate neurons will be a welcome addition. Translational Vestibulo-Ocular Reflex The well-described peripheral anatomy of the otolith organs is such that each saccule or utricle includes hair cells that respond to linear accelerations in all directions through the plane of the organ. Thus, in contrast to the semicircular canals, one otolith organ encodes both excitatory and inhibitory information for any direction of stimulation. While in the canal system redundancy is in the form of ‘push-pull’ complementary responses between canal pairs, the otolith organs have three levels of redundancy for translational motion: between the opposite sides of the striola in one peripheral organ, between the bilateral utricles and saccules, and between the saccules and the utricles along the naso-occipital axis. For this reason, asymmetry in the translational VOR (TVOR) after UL is not necessarily a priori given, as one vestibular nerve has afferents responding to both directions of linear movement. Newlands /Angelaki

40 cm

20 cm

10 cm 20º/s

Horizontal eye velocity

Before labyrinthectomy

10 ms

20º/s 10 ms 1 month after right labyrinthectomy 20º/s

Horizontal eye velocity

Fig. 2. Effects of UL on the TVOR. Horizontal eye velocity of the right and left eyes during transient lateral linear motion and different viewing distances (40, 20, and 10 cm) in a rhesus monkey before and after UL (means ± SD). Within each plot, the dashed line shows zero velocity, and the small solid line at the beginning of the trace shows the initial portion of the stimulus acceleration. Positive eye movements are in response to translation towards the side of the lesion, and negative eye movements are in response to translation towards the intact labyrinth. Used with permission from Angelaki et al. [45].

Horizontal eye velocity

1 week after right labyrinthectomy

Right eye Left eye

10 ms

Study of the TVOR has trailed the study of the RVOR, primarily due to difficulty in the application of controlled translational stimulation as opposed to rotation, the higher frequency dynamics of the TVOR, and the added complication of controlling ocular vergence. The relative paucity of TVOR research has been primarily true for responses after peripheral vestibular lesion. In rhesus macaques, Angelaki et al. [45] found that after UL, TVOR responses are disrupted in complicated ways, dependent upon frequency of stimulation, the axis of movement relative to the head, the direction of gaze, and vergence angle. During sinusoidal lateral motion oscillations (4– 12 Hz), the TVOR amplitude and its viewing distance dependence were initially (3 days after UL) reduced and responses showed an inappropriate vertical eye movement component.

Transient lateral displacements in the UL animals further revealed response asymmetries with lower gains for ipsilesional motion and, as seen in the sinusoidal data, a reduction in the viewing distance dependence. Figure 2 demonstrates the severe and persistent TVOR asymmetry seen in response to translation, especially for near target (10 cm) viewing. Perhaps the most striking deficit was the kinematically inappropriate eye movements during fore-aft motion. Specifically, although compensatory eye movements were typically normal when looking contralaterally to the side of the lesion, the eyes moved in an anti-compensatory direction for targets located ipsilateral to the lesion side. It is important to note that all of these deficits did not compensate completely several months after UL. Although TVOR gain and its viewing distance dependence during sinusoidal lateral motion

Primate Vestibular System

Neuroembryol Aging 2004–05;3:171–182

175

improved with time, there was little recovery in the TVOR during fore-aft motion. Thus, the deficits seen after UL were not due to loss of spontaneous drive to central neurons, but probably due to the loss of the dynamic responsiveness of the lesioned labyrinth. Thus, despite a complete representation of movement directions in the remaining intact labyrinth, these results demonstrate that both labyrinths are necessary for a compensatory TVOR [45]. Similar findings were reported for animals that had first undergone plugging of all six canals and subsequently had a UL [46]. In these monkeys, as with the normal animals that underwent UL, the translational motion only stimulated the otolith organs on one side. However, there were differences in the two preparations, with deficits during lateral motion being larger in animals that had previously adapted from canal plugging. It is possible, as will be further discussed in the next section, that canal and otolith signals interact in such a way that otolith-related responses are changed after adaptation from canal lesions. Given the great deal of convergence in vestibular nuclei neurons [47] (including those involved in the disconjugate pathways of the TVOR [48]), it is not surprising that compensation following canal plugging impacts TVOR responses.

Otolith/Canal Interactions

Tilt/Translation Detection Tilt relative to gravity and translation through space affect the otolith hair cells and primary afferents identically. As a result, based entirely upon signals from otolith afferents, gravitational and translational (inertial) accelerations represent identical stimuli (Einstein’s equivalence principle [49]). The equivalence (ambiguity) of translational and gravitational accelerations has been reported in several studies, where it was shown that the signal conveyed to the brain by primary otolith afferents is the vector sum of the translational and gravitational components of linear acceleration [50–54]. However, the behavioral responses by the organism to tilt and translation are different, so the ability to distinguish the two stimuli is paramount to normal function. Indeed, it was shown that the monkey TVOR appropriately distinguishes roll tilts and lateral translations [55]. The major findings are summarized in figure 3. The head was either rolled 821.8° or translated 837.6 cm at 0.5 Hz. Both stimuli produced the same net acceleration along the interaural axis of the head (0.37 g), but the roll stimulus 176

Neuroembryol Aging 2004–05;3:171–182

produced gravitational acceleration and the translation produced inertial acceleration. The stimuli were combined either in or out of phase to produce a net acceleration of 0 or 0.74 g (tilt – translation and tilt + translation stimulus conditions, respectively). In normal animals, translational motion along the intra-aural axis produced horizontal eye movements, whose amplitude was inversely proportional to vergence angle. This eye movement response was present even when the animal was simultaneously translated and rolled in the same direction, such that the combined stimulation (tilt and translation) produced no net acceleration (fig. 3, column 3). After plugging all six semicircular canals, such a combined roll and translation stimulus did not generate an eye movement because both intra-aural translation and roll tilt produced the same (but opposite) horizontal eye movements. Thus, in the absence of canal signals, all otolith stimulation was interpreted as translation during both sinusoidal (0.16–1.0 Hz) and transient step stimuli with higher frequency content. These results demonstrated that canal signals are critical in the discrimination of tilt from translation, at least for frequencies 10.16 Hz. Mathematically, this discrimination occurs because velocity information from the semicircular canals can be used to keep track of the change in the gravity vector relative to the head [56]. The logical extension of those experiments was to investigate the role of the semicircular canals in tilt/translation discrimination at the single neuron level. Angelaki et al. [57] undertook such an investigation to determine whether they could discern at the cellular level the presence of neural correlates for the terms of the equations necessary to solve the ambiguity question. Specifically, neurons that encode the net acceleration signal and/or an internal estimate of the gravitational component along the same axis were found in the rostral fastigial and vestibular nuclei of rhesus macaques [57]. These areas receive primary vestibular afferent input in monkeys [58, 59] and are known to be involved in the processing of vestibular information. Using similar combinations of tilt and translation as in the behavioral studies, it was shown that central neurons carry the intermediate signals for computing translation. Definitive evidence for the role of the semicircular canals in the neural estimation of the gravitational vector came in a follow-up study, when the same experiments were repeated in animals after all six semicircular canals were plugged [60]. In the fastigial and vestibular nuclei of plugged animals, neurons no longer encoded the gravitational component of linear acceleration. All of the units in the canal-plugged animals instead responded like priNewlands /Angelaki

Translation only IAA 0.37 g

Roll tilt only IAA 0.37 g

Roll tilt + translation IAA 0.74 g

Roll tilt – translation IAA 0 g

Horizontal eye Torsional eye velocity velocity

Horizontal eye Torsional eye velocity velocity

Normal animal

(60º/s)

(30º/s)

After bilateral canal plugging (60º/s)

(30º/s) 10 s

Fig. 3. Tilt-translation discrimination in a rhesus monkey before and acutely after canal plugging. Torsional and horizontal slow-phase eye velocity of the right eye during lateral translation and/or roll tilt at 0.5 Hz in complete darkness. Dotted lines are zero eye velocity. IAA = Interaural acceleration. Modified and used with permission from Angelaki et al. [55].

mary afferents, essentially encoding only net acceleration (the combination of gravitational and inertial acceleration). Otolith-Driven Tilt Responses Help RVOR after Compromised Canal Function The tilt-translation system demonstrates adaptation over time. As discussed above, acutely after canal plugging, both roll tilt and lateral translation elicited horizontal eye movement responses. These eye movements were compensatory to translational motion but not to roll motion, since all linear accelerations without a change in the gravity vector orientation were interpreted as translation. However, over time the system also adapted to having the canals plugged so that the eye movement responses with either roll tilt or interaural translation elicited both torsional and horizontal eye movements [61]. In this study, the roll and pitch RVOR at 0.1–0.5 Hz and translational

lateral and fore-aft VOR were followed over 3 months in monkeys after plugging all six semicircular canals. As seen in figure 4, the roll and pitch VOR gain during rotation about an earth horizontal axis (i.e. with the animal upright) increased over time. In contrast, no improvement was seen during rotation about an earth vertical axis (i.e. with the monkey lying on his side for pitch rotation or on his back for roll rotation), which is a stimulus that does not activate the otolith organs. An important consequence of this adaptation is that, with recovery after canal plugging, the improvement in RVOR causes degradation of the TVOR (e.g. inappropriate torsional eye movements). It is important to note that the increase in RVOR gain was only seen at frequencies above approximately 0.1 Hz, where it is functionally important for visual acuity. As will be discussed in the next section, otolith-driven velocity storage responses, which dominate at low frequencies, deteriorated rather than improved over time [62].

Primate Vestibular System

Neuroembryol Aging 2004–05;3:171–182

177

Gain

1.0

1.0

0.1

0.1

0.01

0.01

Fig. 4. Functional adaptation of the RVOR after canal plugging. Roll (a, b) and pitch VOR (c, d) gains from rhesus monkeys A (a, c) and B (b, d) acutely and at different times after bilateral canal plugging. Rotations were about an earth-horizontal axis with the animal upright. Dotted lines illustrate the earth-vertical axis (EVA) response 3 months after plugging. Used with permission from Angelaki et al. [61].

Gain

a

0.1

b 1.0

0.1

0.1

Neuroembryol Aging 2004–05;3:171–182

0.1

1.0

0.1

1.0

0.01

0.01 0.1 Frequency (Hz)

Spatial Orientation – Velocity Storage Canal-otolith interactions after canal plugging were further tested by examining the low-frequency RVOR and off-vertical axis rotation (OVAR) responses [62]. In contrast to the increase in mid- and high-frequency RVOR described in the previous section, the low-frequency otolith contributions to the RVOR and OVAR decreased and deteriorated over time after plugging of the six semicircular canals. For example, figure 5 illustrates the response to yaw rotation about an earth horizontal axis (‘barbeque spit’ rotation). The bias velocity component, which is the steady-state slow-phase horizontal eye velocity that is proportional to the velocity of rotation, was reduced 4 days after the operation and absent by 2 months after canal plugging. Plugging the canals acutely resulted in a 34% decrease in the bias velocity, and after 2 months, the decrement was over 70%. In these same animals, the gain of the low-frequency (0.01–0.1 Hz) RVOR during sinusoidal rotations in the plane of the plugged canals also decreased over time after surgery. These results argue against velocity storage as a purely oculomotor phenomenon, as it is now generally believed that canal, otolith, somatosensory, and visual inputs all interact in this central process – suggesting that ‘velocity storage’ is the manifestation of a larger, global system of spatial orientation. Functional ablation studies have 178

1.0

1.0

c

Control Acute 4 days 2 weeks 1 month 2 months 3 months 3 months-EVA

1.0

d

Frequency (Hz)

shown that irregularly firing afferents are instrumental in otolith-generated rotational signals during constantvelocity OVAR [19]. Ablating (silencing) the irregular afferents with anodal current bilaterally reduced OVAR responses by roughly 65%, while not influencing optokinetic nystagmus or optokinetic after nystagmus.

Head-Eye Coordination

Coordination of head and eye movements such that gaze (eye position in space) is stabilized while the head is moving is paramount to normal visual function when the head is not stationary. Gaze stabilization has been shown in several studies to be normally dependent upon the VOR [63–65]. A standard horizontal gaze shift includes an eye and head shift that initiate at roughly the same time (fig. 6a). Gaze shifts are typically hypometric and ultimately reach the target with a corrective saccade. Between the primary eye saccade and the correction saccade, the gaze is stabilized by a counterrotation (CR) of the eyes that is mediated by a compensatory RVOR. The CR is in the opposite direction of the head movement, such that gaze remains fixed in space. Early work investigating gaze stabilization during head-eye combined horizontal gaze shifts utilized bilatNewlands /Angelaki

Before lateral canal plugging

Hyaw

(360º)

Torsional (20º/s)

Vertical Horizontal

a

20 s 4 days after lateral canal plugging (360º)

Hyaw

Torsional Vertical

(20º/s)

Horizontal

b

20 s 2 months after lateral canal plugging Hyaw

Fig. 5. Functional adaptation of OVAR re-

sponses after canal plugging. Torsional, vertical, and horizontal slow-phase eye velocity elicited during constant velocity yaw rotation about an earth-horizontal axis (angular head velocity: 58°/s). Responses are shown before (a), and 4 days (b) and 2 months (c) after bilateral inactivation of the lateral semicircular canals. Hyaw = head position (turntable potentiometer output reset every 360°). Dotted lines: zero eye velocity. Modified and used with permission from Angelaki et al. [62].

(360º)

Torsional

(20º/s)

Vertical

Horizontal

c

20 s

eral labyrinthectomy to investigate the role the vestibular system played in the compensatory gaze shift [65]. After bilateral labyrinthectomy, horizontal gaze shifts were initially hypermetric because there was no compensatory CR during the time between the end of the eye saccade and the end of the head saccade. With time, however, the

CR, i.e. the eye-head interaction between the end of the eye saccade and the end of the head saccade, recovered to near normal levels [65]. The observation that, despite abnormal passive vestibular responses (e.g. RVOR), primates with compensated but abnormal vestibular function make gaze shifts

Primate Vestibular System

Neuroembryol Aging 2004–05;3:171–182

179

0

Target

0

Gaze position

60 Gaze position

Target

Gaze position

Target 0

–60 –60

Eye position

20

20

20 0

Eye position

–20

0

–10 –20

Eye position 40

Head position

Head position

Head position 100 ms

–50

100 ms

400

100 ms

–60

0

500

0 Gaze velocity

Gaze velocity Gaze velocity

0

–500

0 400

500

0 Eye velocity

Eye velocity 0

Eye velocity

0 100

–100

0 CR

b

CR

Head velocity

0

Head velocity

0

a

–500

200

Head velocity

c

CR

Fig. 6. Effects of UL on head-free gaze shifts. Examples of 60° centripetal (towards midline) gaze shifts in a rhesus monkey before (a) and 2 days after (b, c) UL. The gaze shift shown in b is towards the lesion (ipsilateral), and the shift in c is away from the lesion (contralateral). Position calibration is in degrees. The velocity calibration is in °/s. The CR period is marked from the end of the eye saccade to the beginning of the correction saccade. Used with permission from Newlands et al. [66].

that are indistinguishable from normal responses has also recently been reported after either UL [66] or bilateral canal plugging [67]. An example after UL is shown in figure 6. After surgery, the CR for a gaze shift towards the side of the lesion is not compensatory, resulting in a gaze shift that slides towards the target (fig. 6b). In contrast, the gaze shift away from the lesion does not slide; instead, the eye velocity exceeds head velocity as a result of the spontaneous nystagmus that is still present 2 days after UL (fig. 6c). Acute deficits were found in both the CR gain (an ‘active’ RVOR response) and in the passive RVOR, both of which were asymmetric, favoring contralesional rotation. Within 2 weeks, the CR gains returned 180

Neuroembryol Aging 2004–05;3:171–182

to normal [66], while recovery of the passive RVOR was incomplete (see sections above). In both of these studies, we found that while the RVOR is normally responsible for gaze stabilization during active head movements, other mechanisms, likely including neural signals representing head movement, may substitute or facilitate a deficient vestibular signal. However, neither cervico-ocular reflexes, passive orbital properties, nor visual feedback could account for the recovery of the CR gain after bilateral canal plugging [67]. When the head was arrested during the gaze shift in the compensated animals, the eye CR continued despite the loss of any sensory feedback, indicating continued head moveNewlands /Angelaki

ment. The data were interpreted as indicating that the recovery of CR gain is not due to substitution of another sensory signal, but rather an internal mechanism, such as an internal neural copy of the intended head movement, substitutes the lost vestibular signal [67].

Conclusion

Peripheral vestibular manipulations continue to represent valuable tools for probing the central organization and functional compensation of vestibular system functions. Such experiments have more recently focused on

probing spatial orientation functions, beyond the VORs [68, 69]. These experiments might be even more necessary since spatial orientation functions are strongly multimodal, such that the contributions of vestibular signals can only be addressed using peripheral vestibular manipulations. Such studies will continue to be valuable in order to understand the many vestibular signal contributions to diverse spatial perception functions. Acknowledgments This work was supported by grants from NASA (NNA04 CC77G) and NIH (R01 EY12814 and R01 DC04260).

References 1 Newlands SD, Perachio AA: Central projections of the vestibular nerve: a review and single fiber study in the Mongolian gerbil. Brain Res Bull 2003;60:475–495. 2 Cohen B, Suzuki JI, Raphan T: Role of the otolith organs in generation of horizontal nystagmus: effects of selective labyrinthine lesions. Brain Res 1983;276:159–164. 3 Yakushin SB, Raphan T, Buttner-Ennever JA, Suzuki J, Cohen B: Spatial properties of central vestibular neurons of monkeys after bilateral lateral canal nerve section. J Neurophysiol 2005;94:3860–3871. 4 Filippone MV, Igarashi M, Miyata H, Coats AC, Alford BR: Superior vestibular nerve sectioning. Experimental studies in squirrel monkeys. Arch Otolaryngol 1975; 101: 241– 245. 5 Igarashi M, Kato Y: Effect of different vestibular lesions upon body equilibrium function in squirrel monkeys. Acta Otolaryngol Suppl 1975;330:91–99. 6 IgarashiM, Takahashi M, Kubo T, Alford BR, Wright WK: Effect of off-vertical tilt and macular ablation on postrotatory nystagmus in the squirrel monkey. Acta Otolaryngol 1980;90:93–99. 7 Money KE, Scott JW: Functions of separate sensory receptors of non-auditory labyrinth of the cat. Am J Physiol 1962;202:1211–1220. 8 Ewald R (ed): Physiologische Untersuchungen über das Endorgan des Nervus octavus. Wiesbaden, Bergman, 1892. 9 Suzuki J, Kodama A, Cohen B, Henn V: Canal-plugging in the rhesus monkey. A tool to study the contribution of individual canals to nystagmus generation. Acta Otolaryngol Suppl 1991;481:91–93. 10 Paige GD: Vestibuloocular reflex and its interactions with visual following mechanisms in the squirrel monkey. II. Response characteristics and plasticity following unilateral inactivation of horizontal canal. J Neurophysiol 1983;49:152–168.

Primate Vestibular System

11 Hess BJ, Lysakowski A, Minor LB, Angelaki DE: Central versus peripheral origin of vestibuloocular reflex recovery following semicircular canal plugging in rhesus monkeys. J Neurophysiol 2000;84:3078–3082. 12 Rabbitt RD, Boyle R, Highstein SM: Influence of surgical plugging on horizontal semicircular canal mechanics and afferent response dynamics. J Neurophysiol 1999;82: 1033–1053. 13 Yakushin SB, Raphan T, Suzuki J, Arai Y, Cohen B: Dynamics and kinematics of the angular vestibulo-ocular reflex in monkey: effects of canal plugging. J Neurophysiol 1998; 80:3077–3099. 14 Lasker DM, Backous DD, Lysakowski A, Davis GL, Minor LB: Horizontal vestibuloocular reflex evoked by high-acceleration rotations in the squirrel monkey. II. Responses after canal plugging. J Neurophysiol 1999; 82:1271–1285. 15 Angelaki DE, Hess BJ, Suzuki J: Differential processing of semicircular canal signals in the vestibulo-ocular reflex. J Neurosci 1995; 15:7201–7216. 16 Minor LB, Goldberg JM: Vestibular-nerve inputs to the vestibulo-ocular reflex: a functional-ablation study in the squirrel monkey. J Neurosci 1991;11:1636–1648. 17 Goldberg JM, Smith CE, Fernandez C: Relation between discharge regularity and responses to externally applied galvanic currents in vestibular nerve afferents of the squirrel monkey. J Neurophysiol 1984; 51: 1236–1256. 18 Goldberg JM, Highstein SM, Moschovakis AK, Fernandez C: Inputs from regularly and irregularly discharging vestibular nerve afferents to secondary neurons in the vestibular nuclei of the squirrel monkey. I. An electrophysiological analysis. J Neurophysiol 1987;58:700–718.

19 Angelaki DE, Perachio AA, Mustari MJ, Strunk CL: Role of irregular otolith afferents in the steady-state nystagmus during offvertical axis rotation. J Neurophysiol 1992; 68:1895–1900. 20 Highstein SM, Goldberg JM, Moschovakis AK, Fernandez C: Inputs from regularly and irregularly discharging vestibular nerve afferents to secondary neurons in the vestibular nuclei of the squirrel monkey. II. Correlation with output pathways of secondary neurons. J Neurophysiol 1987;58:719–738. 21 Goldberg JM, Fernandez C: Physiology of peripheral neurons innervating semicircular canals of the squirrel monkey. 3. Variations among units in their discharge properties. J Neurophysiol 1971; 34:676–684. 22 Fernandez C, Goldberg JM: Physiology of peripheral neurons innervating semicircular canals of the squirrel monkey. II. Response to sinusoidal stimulation and dynamics of peripheral vestibular system. J Neurophysiol 1971; 34:661–675. 23 Goldberg JM, Fernandez C: Physiology of peripheral neurons innervating semicircular canals of the squirrel monkey. I. Resting discharge and response to constant angular accelerations. J Neurophysiol 1971; 34: 635– 660. 24 Blakley BW: Update on intratympanic gentamicin for Meniere’s disease. Laryngoscope 2000;110:236–240. 25 Nakagawa T, Yamane H, Shibata S, Nakai Y: Gentamicin ototoxicity induced apoptosis of the vestibular hair cells of guinea pigs. Eur Arch Otorhinolaryngol 1997; 254:9–14. 26 Hirvonen TP, Minor LB, Hullar TE, Carey JP: Effects of intratympanic gentamicin on vestibular afferents and hair cells in the chinchilla. J Neurophysiol 2005; 93: 643– 655. 27 Fetter M, Zee DS: Recovery from unilateral labyrinthectomy in rhesus monkey. J Neurophysiol 1988;59:370–393.

Neuroembryol Aging 2004–05;3:171–182

181

28 Farrow K, Broussard DM: Commissural inputs to secondary vestibular neurons in alert cats after canal plugs. J Neurophysiol 2003; 89:3351–3353. 29 Fetter M, Zee DS, Proctor LR: Effect of lack of vision and of occipital lobectomy upon recovery from unilateral labyrinthectomy in rhesus monkey. J Neurophysiol 1988; 59: 394–407. 30 Wolfe JW, Kos CM: Nystagmic responses of the rhesus monkey to rotational stimulation following unilateral labyrinthectomy: final report. Trans Am Acad Ophthalmol Otolaryngol 1977;84:38–45. 31 Lasker DM, Hullar TE, Minor LB: Horizontal vestibuloocular reflex evoked by high-acceleration rotations in the squirrel monkey. III. Responses after labyrinthectomy. J Neurophysiol 2000;83:2482–2496. 32 Wolfe JW, Kos CM: Nystagmic responses of the rhesus monkey to rotational stimulation following unilateral labyrinthectomy: a preliminary report. Trans Sect Otolaryngol Am Acad Ophthalmol Otolaryngol 1976; 82: ORL60–ORL69. 33 Broussard DM, Bhatia JK, Jones GE: The dynamics of the vestibulo-ocular reflex after peripheral vestibular damage. I. Frequencydependent asymmetry. Exp Brain Res 1999; 125:353–364. 34 Broussard DM, Bhatia JK, Hong JA: The dynamics of the vestibulo-ocular reflex after peripheral vestibular damage. II. Comparison with dynamics after optically induced learning. Exp Brain Res 1999;125:365–374. 35 Galiana HL, Flohr H, Jones GM: A reevaluation of intervestibular nuclear coupling: its role in vestibular compensation. J Neurophysiol 1984;51:242–259. 36 Galiana HL, Smith HL, Katsarkas A: Modelling non-linearities in the vestibulo-ocular reflex (VOR) after unilateral or bilateral loss of peripheral vestibular function. Exp Brain Res 2001;137:369–386. 37 Curthoys IS, Harris RA, Smith PF: Effect of unilateral labyrinthectomy on neural activity in the guinea pig vestibular nuclei; in Graham MD, Kemink JL (eds): The Vestibular System: Neurophysiologic and Clinical Research. New York, Raven, 1987, pp 633– 642. 38 Smith PF, Curthoys IS: Neuronal activity in the ipsilateral medial vestibular nucleus of the guinea pig following unilateral labyrinthectomy. Brain Res 1988;444:308–319. 39 Smith PF, Curthoys IS: Neuronal activity in the contralateral medial vestibular nucleus of the guinea pig following unilateral labyrinthectomy. Brain Res 1988;444:295–307. 40 Newlands SD, Perachio AA: Effect of T2 spinal transection on compensation of horizontal canal related activity in the medial vestibular nucleus following unilateral labyrinth ablation in the decerebrate gerbil. Brain Res 1991;541:129–133.

182

41 Newlands SD, Perachio AA: Compensation of horizontal canal related activity in the medial vestibular nucleus following unilateral labyrinth ablation in the decerebrate gerbil. II. Type II neurons. Exp Brain Res 1990; 82: 373–383. 42 Newlands SD, Perachio AA: Compensation of horizontal canal related activity in the medial vestibular nucleus following unilateral labyrinth ablation in the decerebrate gerbil. I. Type I neurons. Exp Brain Res 1990; 82: 359–372. 43 Hamann KF, Lannou J: Dynamic characteristics of vestibular nuclear neurons responses to vestibular and optokinetic stimulation during vestibular compensation in the rat. Acta Otolaryngol Suppl 1987; 445:1–19. 44 Ris L, Godaux E: Neuronal activity in the vestibular nuclei after contralateral or bilateral labyrinthectomy in the alert guinea pig. J Neurophysiol 1998; 80:2352–2367. 45 Angelaki DE, Newlands SD, Dickman JD: Primate translational vestibuloocular reflexes. IV. Changes after unilateral labyrinthectomy. J Neurophysiol 2000; 83: 3005– 3018. 46 Angelaki DE, McHenry MQ, Newlands SD, Dickman JD: Functional organization of primate translational vestibulo-ocular reflexes and effects of unilateral labyrinthectomy. Ann NY Acad Sci 1999;871:136–147. 47 Dickman JD, Angelaki DE: Vestibular convergence patterns in vestibular nuclei neurons of alert primates. J Neurophysiol 2002; 88:3518–3533. 48 Chen-Huang C, McCrea RA: Effects of viewing distance on the responses of vestibular neurons to combined angular and linear vestibular stimulation. J Neurophysiol 1999; 81: 2538–2557. 49 Einstein A: Über das Relativitätsprinzip und die aus demselben gezogenen Folgerungen. Jahrb Radioakt 1908;4:411–462. 50 Loe PR, Tomko DL, Werner G: The neural signal of angular head position in primary afferent vestibular nerve axons. J Physiol 1973;230:29–50. 51 Fernandez C, Goldberg JM: Physiology of peripheral neurons innervating otolith organs of the squirrel monkey. I. Response to static tilts and to long-duration centrifugal force. J Neurophysiol 1976; 39:970–984. 52 Si X, Angelaki DE, Dickman JD: Response properties of pigeon otolith afferents to linear acceleration. Exp Brain Res 1997; 117: 242–250. 53 Dickman JD, Angelaki DE, Correia MJ: Response properties of gerbil otolith afferents to small angle pitch and roll tilts. Brain Res 1991;556:303–310. 54 Angelaki DE, Dickman JD: Spatiotemporal processing of linear acceleration: primary afferent and central vestibular neuron responses. J Neurophysiol 2000; 84: 2113– 2132.

Neuroembryol Aging 2004–05;3:171–182

55 Angelaki DE, McHenry MQ, Dickman JD, Newlands SD, Hess BJ: Computation of inertial motion: neural strategies to resolve ambiguous otolith information. J Neurosci 1999;19:316–327. 56 Goldstein H: Classical Mechanics. Reading, Addison-Wesley, 1980. 57 Angelaki DE, Shaikh AG, Green AM, Dickman JD: Neurons compute internal models of the physical laws of motion. Nature 2004; 430:560–564. 58 Kotchabhakdi N, Walberg F: Primary vestibular afferent projections to the cerebellum as demonstrated by retrograde axonal transport of horseradish peroxidase. Brain Res 1978;142:142–146. 59 Newlands SD, Vrabec JT, Purcell IM, Stewart CM, Zimmerman BE, Perachio AA: Central projections of the saccular and utricular nerves in macaques. J Comp Neurol 2003; 466:31–47. 60 Shaikh AG, Green AM, Ghasia FF, Newlands SD, Dickman JD, Angelaki DE: Sensory convergence solves a motion ambiguity problem. Curr Biol 2005; 15:1657–1662. 61 Angelaki DE, Newlands SD, Dickman JD: Inactivation of semicircular canals causes adaptive increases in otolith-driven tilt responses. J Neurophysiol 2002; 87:1635–1640. 62 Angelaki DE, Merfeld DM, Hess BJ: Low-frequency otolith and semicircular canal interactions after canal inactivation. Exp Brain Res 2000;132:539–549. 63 Bizzi E, Kalil R, Tagliasco V: Eye-head coordination in monkeys: evidence for centrally patterned organization. Science 1971; 173: 452–454. 64 Lefevre P, Bottemanne I, Roucoux A: Experimental study and modeling of vestibuloocular reflex modulation during large shifts of gaze in humans. Exp Brain Res 1992; 91: 496–508. 65 Dichgans J, Bizzi E, Morasso P, Tagliasco V: Mechanisms underlying recovery of eyehead coordination following bilateral labyrinthectomy in monkeys. Exp Brain Res 1973;18:548–562. 66 Newlands SD, Hesse SV, Haque A, Angelaki DE: Head unrestrained horizontal gaze shifts after unilateral labyrinthectomy in the rhesus monkey. Exp Brain Res 2001;140:25– 33. 67 Newlands SD, Ling L, Phillips JO, Siebold C, Duckert L, Fuchs AF: Short- and long-term consequences of canal plugging on gaze shifts in the rhesus monkey. I. Effects on gaze stabilization. J Neurophysiol 1999; 81: 2119– 2130. 68 Wei M, Li N, Newlands SD, Dickman JD, Angelaki DE: Deficits and recovery in visuospatial memory using head motion after bilateral labyrinthine lesion. J Neurophysiol 2006;96:1676–1682. 69 Angelaki DE, Gu Y, Newlands SD, DeAngelis GC: Role of area MSTd in heading perception based on vestibular signals. Soc Neurosci Abstr, 2006, Program 12.7.

Newlands /Angelaki

Neuroembryol Aging 2004–05;3:194–206 DOI: 10.1159/000096797

Published online: November 3, 2006

Recovery after Vestibular Lesions: From Animal Models to Patients P.P. Vidal H. Straka N. Vibert L.E. Moore C. de Waele Laboratoire de Neurobiologie des Réseaux Sensorimoteurs, Centre National de la Recherche Scientifique, Unité Mixte de Recherche 7060, Université René Descartes, Paris, France

Key Words Compensation
Intrinsic membrane properties
Oculomotor control
Postural control
Proprioception

Abstract Gaze and postural stabilization is the result of a complex multisensory integration, which can be defined as the process of matching multiple internal representations of an external event (head and/or trunk rotation), obtained from different sensory modalities, into a unique intrinsic frame of reference in which appropriate motor commands can be coded. Despite the importance of the labyrinths, their lesions are not rare. They result in static and dynamic deficits. What is remarkable is that some of these deficits recover in mammalian species, including humans: the vestibular compensation process refers to the complete or partial normalization of the static and dynamic postural and ocular motor deficits. In the following, only two aspects of vestibular compensation will be discussed: (1) the relationship between the recovery of the vestibulo-ocular reflexes and the postlesional plasticity of the intrinsic membrane properties of the medial vestibular nucleus, and (2) the relationship between the postural recovery and the vestibular control of the skeletal geometry at large. Subsequently, the results of investigations in animal models will be related to some clinical findings. While such an exercise is highly speculative, it points to new directions for clinical research and vestibular rehabilitation. Copyright © 2005 S. Karger AG, Basel

© 2005 S. Karger AG, Basel 1661–3406/05/0034–0194$22.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

Introduction

For vertebrates, maintaining their body equilibrium in the gravitational field and being capable of orienting themselves in their environment are mandatory for survival. It requires a permanent control of their skeletal geometry, i.e. not only the control of the head and trunk position in space but also of the head in relation to the trunk, as well as the trunk in relation to the limbs. In this task, three sensory modalities are significantly involved, namely the visual, proprioceptive and vestibular system. Gaze and postural stabilization is therefore the result of a complex multisensory integration. This integration can be defined as the process of matching multiple internal representations of an external event (head and/or trunk rotation), obtained from different sensory modalities, into a unique intrinsic frame of reference in which appropriate motor commands can be coded. Previous studies have demonstrated that well-defined neuronal networks in the central nervous system (CNS) implement these complex sensorimotor transformations known as the vestibular, cervical and optokinetic synergies. In essence, the non-linear properties of several classes of brainstem neurons such as the first- and second-order vestibular neurons, the reticular and prepositus hypoglossi neurons combine in vivo with the emerging properties of these networks to achieve gaze and postural control. The internal representations of head and trunk movements processed by the vestibular nuclei [1–4] are also used as in-

P.P. Vidal LNRS, CNRS UMR 7060, Université René Descartes (Paris 5) 45 rue des Saints-Pères FR–75270 Paris Cedex 06 (France) Tel. +33 142 86 33 97, Fax +33 142 86 33 99, E-Mail [email protected]

puts for various cortical areas and are at the origin of the perception of egomotion [5]. Despite the crucial importance of the labyrinths, their lesions are not rare. They can be progressive, e.g. in case of a slowly growing neurinoma of the eighth nerve or during the course of Ménière’s disease, or more drastically, e.g. during labyrinthitis, following a traumatic event, or after surgery for a small neurinoma. As an unwanted demonstration of the importance of the labyrinth for motor control, its acute lesion leads to a spectacular postural and oculomotor syndrome. This can be divided into two components: static and dynamic deficits. In order to understand the origins of these deficits, it is necessary to summarize the cellular events that trigger them. In humans, about 18,000 first-order vestibular neurons abruptly stop discharging following a lesion in the labyrinth from which these fibers receive their signals [6, 7]. Using the resting discharge of the primate first-order vestibular neurons as an index (100 action potentials per second), this means that about 1,800,000 action potentials per second abruptly cease to depolarize the population of second-order vestibular neurons localized in the vestibular nuclei on the side of the lesion. Hence, these cells are massively disfacilitated. In contrast, the secondorder vestibular neurons localized in the vestibular nuclei on the intact side are disinhibited due to commissural inhibitory fibers that interconnect the vestibular nuclei on both sides of the brainstem. As a result, the mass discharges of the second-order vestibular neurons between the ipsi- and contralesional sides of the brainstem become very asymmetric. In the guinea pig, most secondorder vestibular neurons on the side of the lesion cease discharging [8], however, on the contralesional side [9] they increase their discharge by about 15%. The asymmetry of the mass discharge of vestibular neurons mimics a physiological signal indicating a passive body rotation towards the contralesional side. Quite naturally, that illusion of movement triggers a counterrotation of the eyes and a readjustment of the posture towards the ipsilesional side. These multiple events are referred to as the static vestibular syndrome, which includes a spontaneous nystagmus, yaw- and roll-tilts of the head, asymmetric tones in the extensor muscles of the anterior and posterior limbs and finally, a tendency to fall towards the side of the lesion. What is remarkable is that some of these deficits recover in about a week in a number of mammalian species, including humans. By contrast the dynamic syndrome, which is an impairment of the vestibulo-ocular and vestibulo-spinal reflexes after unilateral labyrinthectomy [10–12], takes several weeks to im-

prove and the final compensation is much more limited compared to the static syndromes both for animals [13– 18] and humans [19–22]. In fact, some of the dynamic deficits never recover and can be seen even 2 years after the lesion [14, 15]. The term vestibular compensation usually refers to the complete or partial normalization of the static and dynamic postural and ocular motor deficits. Hypothetically, once a labyrinth is destroyed, the first-order vestibular neurons could still recover a resting discharge due to plastic cellular pacemaker properties. However, this never occurs [23, 24], in the case of postganglionic lesions [25–27], the afferent fibers even degenerate. Hence, the neuronal mechanisms underlying vestibular compensation must take place within the CNS, which explains that it is one of the oldest and most attractive models of postlesional plasticity of the CNS. To briefly summarize the compensation process, the improvement in the postural and oculomotor syndromes is quite logically related to the disappearance of the asymmetry in the mass resting discharge rates of central vestibular neurons, which caused them in the first place. During the 1st week after the lesion, the ipsilesional medial vestibular nucleus neurons (MVNn) progressively recover a normal spontaneous activity in rodents, while the discharge of contralesional MVN neurons decreases [8, 9]. On the other hand, some components of the behavioral recovery might not be directly correlated with the restoration of a symmetrical mass discharge of the vestibular neurons [28]. For instance, the ocular nystagmus and the head nystagmus disappear well before any recovery of the spontaneous activity in central vestibular neurons can be observed on the ipsilesional side [28]. Clearly then, other events than the discharge recovery taking place in the vestibular nuclei occur elsewhere in the CNS during the vestibular compensation process. They include a reorganization of synaptic inputs onto the deafferented vestibular neurons [29, 30], changes in the sensitivity of the neurons to excitatory and inhibitory transmitters, as well as changes at various levels of the CNS, including in the spinal cord [31, 32]. Given the limited scope of that review, it is not possible to summarize all the studies devoted to that topic. Therefore, we refer the interested reader to the number of reviews available on vestibular compensation [21, 30, 33–42]. In the present paper, we will discuss only two aspects of the vestibular compensation process: first the relationship between the recovery of the vestibuloocular reflex (VOR) and the postlesional plasticity of the intrinsic membrane properties of the MVNn, and second, the relationship between the postural recovery and

Implications for Clinical Research in Vestibular Compensation

Neuroembryol Aging 2004–05;3:194–206

195

the vestibular control of the skeletal geometry at large. In doing so, we will try to link the results of investigations in animal models to some clinical findings. While such an exercise is highly speculative by nature, it may point to new directions for clinical research and vestibular rehabilitation.

Recovery of VOR and Postlesional Plasticity of the Intrinsic Membrane Properties of the MVNn

Intrinsic Membrane Properties of the MVNn Following pioneering studies in the rat by the group of Gallagher et al. [43] and Lewis et al. [44, 45], intracellular recordings in guinea pig brainstem slices [46, 47, for review see 41] led us to propose that the MVNn could be classified into two main classes according to their intrinsic membrane properties: the type A and B cells [see also 48 for lateral vestibular nucleus neurons]. Since then, type A and B MVNn have also been recorded in mouse and chick MVN [49, 50]. The description of these two types of neurons will be based on our description in the guinea pig [46, 47]. However, two important points should first be mentioned. As advocated by du Lac and Lisberger [50], a segregation in classes may be quite artificial considering that the MVNn membrane properties may be distributed as a continuum between two stereotyped schemes corresponding to the canonical A and B cell types. On the other hand, as proposed by Johnston et al. [51], the great variability in the size of the fast and slow after-hyperpolarizations (AHP) in type B MVNn as well as their dependence on the resting membrane potential require averaging of several successive action potentials to identify unambiguously the class of a given cell. Once this is done, MVNn can be unambiguously segregated into essentially two classes: the type A and B MVNn. In the guinea pig, type A MVNn are characterized by rather wide action potentials (11 ms at threshold) followed by a single deep AHP. They also exhibit a transient rectification of the membrane potential which delays the firing of the cell at the break of hyperpolarizing current steps and is likely due to the activation of an IA-like current. Noteworthy, in rats, the true IA conductance is present in both type A and B MVNn, although being larger in type A cells [51]. Finally, type A MVNn display small, high-threshold calcium spikes. Type B MVNn are characterized by shorter duration action potentials followed by a biphasic AHP, with an early fast and small component followed by a slower delayed one. The overall amplitude of this AHP is smaller than in 196

Neuroembryol Aging 2004–05;3:194–206

type A MVNn. Type B MVNn also display potent persistent subthreshold sodium plateau potentials, high-threshold calcium spikes and prolonged calcium-dependent plateau potentials. Finally, about one quarter of B MVNn, namely type B + LTS MVNn, show low-threshold calcium spikes (LTS) that provide bursting properties. For more details, we refer the interested reader to a recent and extensive review on that topic [41]. The MVNn Membrane Properties Play a Critical Role in Their Network Dynamic Properties The vestibulo-ocular and vestibulo-spinal network provide the ability to hold gaze fixed on an object during passive head movement. Within these networks, most of the second-order neurons of the MVN compute internal representations of head movement velocity in the horizontal plane. In vivo recordings in the cat have led to a classification of these neurons into tonic and kinetic neurons [52]. This separation into subpopulations with different response dynamics might be related to the wide frequency spectrum of head accelerations (DC to 20 Hz), which occurs physiologically, for instance in humans [53–55]. On the other hand, in vitro slice studies have shown that the same MVNn can also be grouped into two broad categories, namely type A and B neurons, depending on their intrinsic membrane properties. Again, the latter classification may encompass a continuum of cells more than two strictly defined classes. Based on a study in the in vitro whole brain of the guinea pig [56], we have proposed that type A and B neurons might correspond to the tonic and phasic cells identified in vivo, respectively. In other words, the MVNn membrane properties could contribute to specify the dynamic properties of these neurons independently of their connectivity, which in turn would segregate the vestibulo-ocular and vestibulospinal pathways in frequency-tuned channels [see discussion in 57]. In order to test that hypothesis theoretically, we performed two studies aimed at investigating whether the MVNn intrinsic membrane properties indeed constrain the dynamics of their responses to head accelerations of various amplitudes and frequencies [58, 59]. Biophysical models of the type A and B MVNn were developed. Two major factors were found to determine tonic and phasic firing activity: the activation of the delayed potassium current and the calcium conductance kinetic behaviour. The ratio between the densities of different potassium currents also determined the different forms of action potentials observed experimentally. The two models (type A and B cells) were then tested with depolarizing Vidal/Straka/Vibert/Moore/de Waele

stimulations, random noise, step, ramp and sinusoidal inputs. Altogether, these theoretical results supported the hypothesis that intrinsic membrane properties of MVNn (and by extension those of other cell types of the vestibular network at large) could indeed help specify their dynamic properties and therefore contribute to the wide range of dynamic responses, which characterize the vestibulo-ocular and vestibulo-spinal network. Experimental Results In order to test this hypothesis experimentally, in vitro intracellular injections of small amplitude sinusoidal currents (^10 mV) were then performed in MVNn. It induced a modulation of the MVNn resting discharge in a linear way, which reflected the underlying modulation of the membrane potential [60–62]. The modulation of the discharge augmented together with the frequency of the injected current up to a resonant frequency. Interestingly, the resonance was observed at peak frequencies ranging from 4 to 10 Hz [60, 62], which were higher in type A than in type B MVNn. Above that frequency, the modulation of the discharge decreased for both types of neurons. As a rule, the modulation of the firing rate was of higher amplitude, and the peak discharge was better synchronized with the peak of current in type B than in type A MVNn. The input resistance of type A and type B MVNn being equal, the higher sensitivity of type B cells was related to a higher sensitivity to a given modulation in their membrane potential. Hence, synaptic inputs should be more efficient to trigger the discharge of type B than type A MVNn [62]. Injection of ramp-like currents induced a linear response in type A MVNn. Type B MVNn were more sensitive and displayed a non-linear overshoot preceding the steady-state plateau [60, 62]. Altogether, the experimental results corroborated some of the predictions made by the theoretical studies. Our conclusion, as stated in a larger review on that topic [41], is that the synaptic responses of type B MVNn are more likely to trigger non-linear firing behavior in response to relatively small amplitude, but high-frequency inputs. Thus, the differential behavior of the two types of neurons suggests that type B neurons are moderately tuned active filters that promote high-frequency responses, whereas type A neurons are more like low-pass filters well suited to maintain and modulate at low frequency the resting activity. Consequently, the MVNn membrane properties appear indeed to contribute to the specification of the dynamic properties of these neurons independently of their connectivity.

Implications for Clinical Research in Vestibular Compensation

MVNn Membrane Properties following a Unilateral Vestibular Lesion The long-term consequences of unilateral labyrinthectomy were investigated by characterizing the static and dynamic membrane properties of ipsi- and contralesional vestibular neurons recorded intracellularly in guinea pig brainstem slices. We compared the responses of type A and type B MVNn identified in vitro to current steps and ramps, and to sinusoidal currents of various frequencies. All ipsilesional vestibular neurons were depolarized by 6–10 mV at rest compared to the cells recorded from control slices [62]. Both their average resting membrane potential and firing threshold were increased, which suggests that changes in active conductances compensated for the loss in excitatory afferents. The AHP and discharge regularity of the type B neurons were increased, which should help stabilize the recovered tonic discharge in the disfacilitated MVNn on the ipsilesional side. Furthermore, these neurons became more sensitive to current injections over a large range of frequencies (0.2– 30 Hz), which might underlie the partial compensation of the dynamic vestibular reflexes observed in vivo. This was associated with an increase of the peak frequency of linear response of type B neurons, from 4–6 to 12– 14 Hz. This might reduce their ability to act as non-linear signal detectors at high frequency, and thus may explain why responses to high-amplitude velocity steps stay permanently impaired in lesioned animals. Altogether, these results suggest that long-term vestibular compensation involves major changes in the membrane properties of vestibular neurons on the ipsilesional side. However changes in membrane properties of MVNn occur also on the contralesional side, as indicated by an increase in the proportion of type B neurons among contralesional MVNn [63]. The sensitivity of their firing rate to current injections and variations of the membrane potential was increased. This was associated with a shift of the responses towards larger phase leads, reflecting a more active behavior of the neurons. The response dynamics of contralesional type A MVNn were subjected to the largest modifications and became similar to that of type B MVNn. The general shift of contralesional MVNn towards more phasic properties would explain the full recovery of the vestibular responses to high-acceleration impulses directed towards the contralesional side. Following unilateral labyrinthectomy, the membrane properties of MVNn located on the ipsi- and contralesional sides of the brainstem are differentially modified. The shift of ipsilesional MVNn towards more ‘A-like’ memNeuroembryol Aging 2004–05;3:194–206

197

brane properties should stabilize the tonic activity recovered by the deafferented neurons, while the increased sensitivity and shift of contralesional MVNn towards more ‘B-like’ properties should increase the efficiency of the vestibular synergies triggered by the remaining sensory afferent inputs. These results are likely to be partly explained by structural changes taking place at the level of the vestibular neurons both on the ipsi- and contralesional sides [64, 65]. Impairment and Recovery of the VOR following a Unilateral Vestibular Lesion Immediately following unilateral labyrinthectomy, the VOR gain not only decreases during rotation towards the lesioned side but also towards the intact one. It is likely that this decreased excitability of the intact side contributes to reduce both the nystagmus and the asymmetry of the VOR. It takes several weeks in guinea pigs to observe an amelioration of the VOR gain [13, 14] and the same holds for primates [17, 66] and vestibular patients [19, 20, 67–73]. The ‘compensation’ of the dynamic syndrome is by far less complete than the static syndrome. For instance, the VOR gain during rotation towards the lesioned side remains permanently decreased between about 0.01 Hz and 0.1 Hz, in patients [19, 20, 67–70, 72, 74, 75], as well as in guinea pigs [13, 76]. Above 0.1 Hz, due to the compensation process, the VOR gain improves progressively to achieve quasi-normal values around 1–2 Hz. Above 3 Hz, sinusoidal rotations have not yet been used in animal models and of course not in patients. However, velocity steps are routinely used to test the high-frequency responses of the VOR following unilateral lesion, which turned out to be impaired. Both in guinea pigs [77] and monkeys [17, 78], head rotations at high velocities towards the lesioned side revealed a lack of compensation for high frequencies of head stimulation. The same result was observed in patients following passive rapid head rotations stimulating all pairs of semicircular canals [79–84]. Since the frequency of natural head movements extends from very low frequencies up to 20 Hz in humans and the VOR compensates neither at very low nor at high frequencies, patients are bound to have some problems following unilateral vestibular lesions. In other words, VOR compensation is likely to have a bell-shaped curve (no compensation at high and very low frequencies), and this curve is likely to also differ from patient to patient. In this context, it could be interesting to quantify the compensation process with a wider range of stimulation fre198

Neuroembryol Aging 2004–05;3:194–206

quencies to determine the upper and lower limits of the bell-shaped curve of the compensation of dynamic deficits in each vestibular loss patient. Indeed, some patients who complain of enduring instability and dizziness following a vestibular lesion may well present abnormalities in the dynamic compensation process. For instance, they could compensate over an abnormally narrow range of frequencies of head movements. Incidentally, another reason to test these patients with a wide range of frequencies may be to avoid a wrong diagnosis. Indeed, a patient can present a complete abolition of the caloric response, which corresponds to a very-low-frequency stimulation of the intact labyrinth, while having a very decent VOR gain around 1–2 Hz. For the moment, it is technically difficult to establish the causal relationship between the cellular events recorded in vitro and the behavioral observations performed in vivo. Nevertheless, given the similarities of vestibular compensation in human and animal models, it is tempting to speculate about the cellular basis of the compensation process. In particular, many of the modifications described above concerning the intrinsic membrane properties of the MVNn could well promote the partial recovery of the horizontal VOR. For instance, the shift of ipsilesional type B MVN neurons towards more A-like electrophysiological properties contributes to stabilize the resting discharge of these cells. This is important, since a regular and stable resting discharge of the vestibular neurons on the ipsilesional side is a prerequisite for an efficient modulation of their activity by their remaining afferent inputs, e.g. the inhibitory input from the contralateral MVNn, the excitatory input from the cervical afferents and the visual input. However, this shift towards more type A-like properties implies also an impairment in their non-linear response behavior. Therefore, it may explain the absence of recovery of compensatory eye movements to high frequency-/high-acceleration head movements towards the ipsilesional side. By the same token, the shift towards more B-like properties on the intact side may help to recover a full dynamic range of the vestibular reflexes when the head is turned towards the intact side, despite the disappearance of the commissural inhibition coming from the ipsilesional side. Also, clinical research can suggest new direction for investigations in animal models. A study by Fish [85] using the caloric test has shown that following a unilateral vestibular lesion, the initial hypo-excitability of the intact side lasted several months. However and quite surprisingly, it was then replaced by several months of hyperexcitability. To the best of our knowledge, no one has Vidal/Straka/Vibert/Moore/de Waele

studied vestibular compensation after such a long delay following the initial lesion. Moreover, it suggests that the compensation process in patients continues for several months, even years, which is obviously of practical importance and would deserve more clinical investigations. Sensory Substitution and Behavioral Strategies Since the VOR is severely impaired at low and high stimulus frequencies, other sensory modalities might come into play to stabilize gaze. This has been well demonstrated in animal models and in vestibular patients. Visual substitution was shown to compensate for the vestibular deficit at low frequencies of stimulation [86–89]. Proprioceptive afferents also play a very important role. The gain of the cervico-ocular reflex (COR) rises from negligible values to 0.9 following unilateral labyrinthectomy in the monkey [90]. Conversely, in vivo [91] and in vitro studies [29] have demonstrated that the synaptic strength of the proprioceptive synaptic inputs reaching the central vestibular neurons increased on the side of the lesion and decreased on the intact one. Similar result were observed in patients using vibrations of the neck muscles [92–95] and measurements of the COR [96, 97]. The COR gain increased from 0.1 to reach 0.9 for trunk rotations at 0.3 Hz and then decreased with increasing frequencies. Hence, the COR can substitute for missing vestibular signals at low frequency. Interestingly for rehabilitation, this compensatory process was very variable from patient to patient. Finally, the VOR deficit can be overcome using various behavioral strategies. Patients may simply stop making rapid head movements. They can also replace deficient slow phases by succession of small saccades [98]. Finally, bilaterally labyrinthectomized monkeys have been shown to perform pseudo-slow phases [90], as demonstrated by abruptly stopping head movement during a rapid gaze shift. These eye movements were presumably of cortical origin [99]. This last strategy has never been investigated in vestibular patients following unilateral or bilateral deficits.

Skeletal Geometry and Vestibular Lesion

Balance is a complex biological function dependent upon sensory inputs from visual, proprioceptive and vestibular systems [100]. Using cineradiographic methods, we have studied for the past 10 years the tridimensional geometry of the skeletal system in several species of quadrupedal vertebrates during various types of motor beImplications for Clinical Research in Vestibular Compensation

haviors [101–106] and compared the results with those obtained in humans. Two major conclusions have been made: first, the resting postures, in alert unrestrained subjects, present some common characteristics across all investigated mammalian species (mice, rats, guinea pigs, rabbits, cats, monkeys and humans). Amongst other invariants, the cervical column is held vertically and the plane of the horizontal semicircular canals is tilted up by about 10–20°, which brings the utricular plane approximately in the earth-horizontal plane. Second, this resting posture severely constrains the number of solutions available for the movement of the head-neck ensemble. This reduction in the degrees of freedom likely simplifies the computational work of the CNS during various motor tasks, e.g. gaze shift, gaze stabilization, feeding behavior and locomotion. Using a naval analogy, we have proposed that what is important for postural control is to be in good trim [107]. Mammals are thought as lying in the gravity field as a boat in water. The underlying assumption is that what is controlled is a particular skeletal geometry in the frontal, sagittal and horizontal plane. In that view, the vertical and straight-ahead directions are considered as the perceptual consequence of a seemingly correct symmetrical posture rather than the critical variables to be controlled. An obvious candidate to control the trim is the vestibular system. Despite its name, the vestibular nuclear complex is essentially elaborating internal tridimensional representations of head, trunk and probably leg movements in space on the basis of various sensorial inputs: vestibular, visual and proprioceptive inputs converge at this level. This is a bilaterally organized system. Afferents from each end organ of the labyrinth and their spatially matched visual and proprioceptive inputs converge on specific populations of central vestibular neurons to generate these internal representations. These neurons relay their signals monosynaptically (vestibulospinal tracts) and disynaptically (reticulospinal tracts) to several motoneuronal pools controlling distinct groups of axial and distal muscles and therefore the position of various segments of the body. Therefore, a good trim would be an emerging property of the resting discharge and of the dynamics of such hard-wired neuronal circuits (as seen above, the heterogeneity of the membrane properties of the reticular and vestibular neurons provide a large dynamic range). These were hypotheses and in theory, one should be able to test them by stimulating selectively the vestibular sensors.

Neuroembryol Aging 2004–05;3:194–206

199

Selective Lesions of the Vestibular Sensors in an Animal Model By stimulating unilaterally each of the nerve branches innervating the different vestibular end organs, one should induce a new trim of the posture. The new skeletal geometry should be the result of the activation of the hard-wiring pattern linking a given vestibular end organ to a specific motoneuronal pool. In theory, this approach would provide a way to define which body segments are controlled by each of the semicircular canals including the utricular and saccular end organs, as well as in what plane of space are the orientation and the motility of these segments organized. However, because we are dealing with posture, a study on fully alert, totally unrestrained animal is required. In that condition, the use of a selective stimulation of each of the peripheral subdivisions of the vestibular nerve innervating each end organ is technically out of reach. So we have used selective lesions of the vestibular sensors. Our hypothesis was that the destruction of a given vestibular end organ would give a mirror image of the modifications of the skeletal geometry induced by its selective stimulation [102]. In order to test that hypothesis, we have taken advantage of the fact that, contrary to selective stimulations of the vestibular end organs, the stimulation of the whole vestibular nerve was technically feasible using chronically implanted electrodes in totally unrestrained guinea pigs. The guinea pig was chosen as a model because the anatomy of its middle ear makes it particularly suitable for surgical approaches. As a result, mirror images of the changes in the skeletal geometry induced by a unilateral lesion of the labyrinth and by its unilateral stimulation with galvanic current were obtained [Vidal, unpubl. results]. A somewhat similar result was also obtained in humans by comparing the effects of a galvanic stimulation of the labyrinths in healthy subjects with those of an unilateral vestibular lesion in patients [108]. Hence, we concluded that the selective lesions of each end organ or the corresponding subdivision of the vestibular nerve could give at least an approximate idea of the effects of a selective stimulation of that end organ. Selective lesions were performed for the horizontal ampullary nerve, the superior ampullary nerve, the utricular and saccular macula, the saccular macula and the posterior semicircular canal ampulla. Guinea pigs were X-rayed 4–7 h after full recovery from anesthesia and the posture was quantified [102]. Results of the Canalar Lesions Unilateral lesion of the horizontal semicircular canal produced a rotation of the head directed ipsilaterally to 200

Neuroembryol Aging 2004–05;3:194–206

the side of the lesion in the earth-horizontal plane. The cervical column was still held vertically in the frontal and sagittal plane. Our previous biomechanical studies showed that the rotation induced by the lesion was produced by a rotation of the cervical column about a vertical axis between C1 and C5, of which 50% occurred between C1 and C2. This result is in good agreement with previous electrophysiological studies. Indeed, the excitatory drive from the horizontal semicircular canal is strictly contralateral, and these signals are relayed mostly to the upper cervical cord, where they activate motoneurons of the obliquus capitis, splenius, rectus and complexus muscle, which act as rotators of the head in the earth-horizontal plane. In contrast, unilateral lesion of an anterior or posterior semicircular ampulla did not induce changes in the overall head-neck configuration at rest. In other words, guinea pigs could maintain a normal posture in the sagittal plane with one anterior or posterior canal instead of two. This implies that the excitatory and inhibitory inputs from the intact anterior and posterior semicircular canals are conveyed in a symmetric way to spinal motoneurons on both sides of the spinal cord. Again, this result is in good agreement with previous electrophysiological studies, which demonstrated that the second-order vestibular neurons receiving a vertical semicircular canal input project bilaterally to the spinal cord. The convergence of signals from the two posterior and anterior semicircular ampullae also occurs at the interneuronal level of the spinal cord. Altogether then, our behavioral results nicely fit with the pattern of monosynaptic connections of the different categories of second-order vestibular neurons, which relay the information transduced by the three semicircular canals to the spinal motoneurons. This is quite surprising in view of the complex network, which links the vestibular receptor hair cells of each semicircular canal ampulla to these cells via reticular neurons and several spinal interneurons. These findings have two consequences: first, they suggest that monosynaptic connections are indeed of strong functional importance, which was not previously accepted, and second, the sensory to motor transformation necessary to convert head velocity into a command within the appropriate motor frame of reference for the axial musculature could indeed be embedded in the hard-wired networks involving the second-order vestibular neurons, such as those used for the stabilization of the eyes. Regarding the compensation process, the ocular nystagmus and the rotation of the head induced by the horizontal semicircular canal nerve section disappears in a matter of 6 h, which is well before a recovery of a symVidal/Straka/Vibert/Moore/de Waele

metrical discharge by the vestibular nuclei neurons. In rodents and guinea pigs, the fovea is poorly developed compared to humans and primates. Hence it is unlikely that the nystagmus and the head deviation are cancelled by eye fixation as in the latter species. Cerebellar loops would play a role as well [5, 109–111]. Results of the Otolithic Lesions The otolith system is a good candidate to assure a good trim of skeletal geometry at rest and of the head-neck ensemble in particular, since it is sensing both linear head acceleration and orientation with respect to gravity [102]. Unilateral otolithic lesions sparing the semicircular canals have shown that the head-neck ensemble was inclined towards the side of the lesion. The tilt of the headneck ensemble in the frontal plane was caused by rotations of the thoracic and lumbar vertebrae, which took place at the cervicothoracic junction. Quite logically there was no rotation around the longitudinal axis of the cervical column because the horizontal semicircular nerve was intact. It is as if the animal was still looking straight ahead in what used to be the earth-horizontal plane before the lesion. Performing a selective manual lesion of the utricle is almost technically impossible. However, following lesioning of the utricles on both sides by centrifugation, the neck pitched forward. In contrast to these latter utricular effects, a selective lesion of the saccule did not produce much effect. Altogether these results strongly suggest that the utricle controls the neck orientation in both the sagittal and frontal planes, which also is logical considering the orientation vectors of the sensory cells on the utricular maculae. On the other hand, these results also suggest a fundamental difference in the control of the skeletal geometry in the frontal and sagittal planes by the utricle, despite the fact that the orientation vectors of the sensory cells of the utricular macula on both side of the striola appear to indicate the contrary. Indeed, no deviation in the head-neck ensemble was observed in the sagittal plane following a unilateral otolithic lesion, while the head was pitched forward following a bilateral utricular lesion. Therefore, following a unilateral lesion, the intact utricle could control a normal neck orientation in the sagittal plane. This result implies that excitatory and inhibitory inputs coming from the intact utricle were conveyed in a symmetrical way to motoneuron pools on both sides of the spinal cord which control the head orientation in the sagittal plane. In contrast, unilateral lesion of the otoliths induced a gross deviation of the head-neck ensemble in the frontal plane. This indicates that the intact utricle Implications for Clinical Research in Vestibular Compensation

cannot control the orientation of the head in the frontal plane by itself. This result implies that excitatory inputs from the intact utricle are only conveyed to the motoneuron pools on the contralateral side and could therefore not substitute the silenced afferents from the severed utricle. In summary, while the utricular control of the head in the sagittal plane was organized in a synergistic mode between the two utricles. It was organized in a push-pull manner in the frontal plane. As expected from the above results, a complete unilateral lesion of the labyrinth mainly consisted of the additive effects of a lesion of the horizontal semicircular canal and the utricle. The head was therefore rotated ipsilaterally to the side of the lesion about the axis of the cervical column, and the head-neck ensemble was inclined in the frontal plane about the cervicothoracic junction. Finally, in contrast to the head rotation induced by the lesion of the horizontal semicircular nerve, the inclination of the head-neck ensemble induced by the otolithic lesion takes much longer to compensate. This takes about 1 week and the postural recovery nicely parallels the progressive reappearance of a normal resting discharge of the second-order vestibular neurons on the ipsilesional side. However, and very interestingly for what will follow concerning the vestibular patients, the bascule of the head-neck ensemble never fully recovers. A residual head tilt was observed in all the vertebrate species tested so far. Postural Control in Vestibular Patients At the acute stage following a unilateral lesion of the labyrinth, a discrete rotation of the head of the patients can be observed towards the side of the lesion. If we consider the results described above in the animal models, it is presumably due to the lesion of the horizontal semicircular canal. As for the horizontal nystagmus, it rapidly disappears in the light. However, as stated above, the compensation of the lesion of the horizontal semicircular canal is likely to be different in humans, monkeys and rodents. Indeed, in patients and in primates, the nystagmus decreases in about 10 days in the light [112] but it persists in the dark for several weeks [17, 66]. Our hypothesis would be that when patients (and monkeys) are tested in the light, foveation of a point of interest would cancel the postlesional nystagmus, as is the case for the VOR in control subjects [10]. The enduring nystagmus in the dark suggests that in patients, contrary to what happens in rodents, a persistent asymmetry in the mass discharge of the MVNn can be seen for several weeks after the lesion. In other words, the cancellation of the asymNeuroembryol Aging 2004–05;3:194–206

201

metry of the mass discharge of vestibular neurons would be more context dependent in primates than in rodents and could be under cortical control [99]. The otolithic lesion is at the origin of what has been called the ocular tilt reaction (OTR). The OTR is a triad composed of a vertical strabismus, a tonic deviation of the eye and a lateral head tilt. It is observed either after a peripheral [113–115] or a central lesion [114, 116–119]. OTR induces a deviation of the subjective vertical. As expected from what we have described for the static syndrome, most of the signs of the OTR usually recover in about 1 week [72, 120]. As in animals, the head of the patients is tilted laterally and not pitched forward following a unilateral vestibular lesion. This can be explained in view of the results of the selective lesion of the otolith system in the animal model described above. Therefore, by analogy, we propose that in patients, it is the utricular lesion which causes the head tilt and the postural syndrome in the frontal plane. However, if so, it remains to be explained why the utricular lesion causes an inclination of the head-neck ensemble, which is far more dramatic in quadrupedal mammals than in patients. It would be simply due to the fact that the skeletal geometry has changed when a bipedal stance was adopted in humans. The cervical column became aligned with the thoracic column and hence the degree of freedom of the cervical column with respect to the thoracic column was modified. In particular, it was largely restricted to the frontal plane. Accordingly, the utricular control of the cervicothoracic junction in the frontal plane became obsolete. Instead, the utricular information regulates complex head-neck-backlimbs synergies in order to secure balance in the frontal plane. Two arguments render this hypothesis not too speculative: First, studies using multidirectional perturbations have shown that postural control in both humans [121–124] and animals [125] is distinct in the mediolateral and anteroposterior planes. Second, patients indeed fall to the side at the acute stage of a unilateral vestibular lesion and not forward. In other words, in humans, by analogy with what we have described in quadrupeds, the control of the whole posture in the sagittal plane is organized in a synergistic mode from the two utricles, however it is organized in a push-pull manner in the frontal plane. Everything described up to now in animal models and in patients are the consequences of brisk and brutal unilateral vestibular lesions. On the other hand, we have very little insight of what is going on when the vestibular system slowly decays, as it is the case in aging. In that context, it is intriguing that numerous falls in elderly people 202

Neuroembryol Aging 2004–05;3:194–206

occur sideways and often result in hip fracture. It can be questioned whether this is linked to the degeneration of the vestibular hair cells in the two utricles over time. Skeletal Geometry and Posturography in Patients with Peripheral Vestibular Lesions We have speculated in the paragraph above that vestibular patients with a unilateral lesion of the labyrinth encounter problems in regulating their posture essentially due to the fact that they have a utricular lesion, which compromises their postural control in the frontal plane. It is clear from clinical observations that such a deficit, whatever its origin, is very prominent immediately after the lesion, but then fully recovers in a matter of a week. This is surprising because the inclination of the headneck ensemble towards the side of the lesion never fully recovers in quadrupeds following a unilateral labyrinthectomy. This opens an intriguing question: Why is the postural syndrome fully compensated in human patients while the compensation seems to be only partial in nonprimate vertebrates? One possibility is that, in humans, an adaptive strategy could be involved to solve the problem of the residual tilt by using context-specific visual and proprioceptive information as substitutes for the missing vestibular inputs. Such a hypothesis can be tested because it has two predictable consequences: first, postural control in patients when tested in the absence of proper visual and proprioceptive information should be specifically more impaired in the frontal than in the sagittal plane due to the residual utricular deficit. Second, this deficit should last for a very long time following the vestibular lesion, if not for ever as in quadrupeds. In order to test these two hypotheses, the postural instability of 11 patients with bilateral and 88 with unilateral vestibular loss (vestibular neurotomy, acoustic neurinoma resection, chemical labyrinthectomy and vestibular neuritis) was investigated using static and dynamic posturography at various times following the lesion [108]. Caloric and head impulse tests, vestibular evoked myogenic potentials evoked by short tone burst and short duration galvanic currents were used to select the patients. Our aim was actually threefold: first, to determine the best indicators to distinguish vestibular loss patients from controls; second, for reasons explained above, to compare body sway motion in the sagittal and frontal plane on a moving platform for both eyes-open and eyesclosed conditions, and third, to assess any recovery of abnormal postural performance over time associated with the vestibular compensation process. Except in bilateral vestibular loss patients, static posturography was unable Vidal/Straka/Vibert/Moore/de Waele

to distinguish unilateral vestibular loss patients from control subjects in either eyes-open or eyes-closed conditions. In contrast, in the eyes-closed condition, dynamic posturography using a non-motorized, low-cost, seesaw platform allowed one to separate bilateral and unilateral loss patients from control subjects: bilateral loss patients were unable to stand up without falling at any time following the lesion, and unilateral loss patients were unable to stand up in either the sagittal or frontal planes during the 1st week after the lesion. By the 2nd week, unilateral loss patients could with difficulty maintain balance in the sagittal plane, but not in the frontal plane. After 2 months, most of the patients could maintain their equilibrium in both planes but all the measured instability values were higher than those in controls, and higher for the frontal than the sagittal plane. Total energy and sway area were found to be the best indicators. After 1 year, values were closer to normal in both planes but remained higher than normal in the frontal plane. These latter results led to three conclusions: first, from a practical point of view, dynamic posturography on a seesaw platform can be a valuable tool for clinical diagnosis and quantitative analysis of imbalance in patients suffering from a unilateral and/or bilateral vestibular loss up to 1 year after the lesion (longest delay investigated). Second, our data strengthen the hypothesis that an enduring utricular deficit is instrumental in the deterioration of the posture in the frontal plane in patients with a unilateral vestibular loss. Third, it suggests that this deficit, while well compensated 1 week after the lesion using an adaptive strategy, is in fact a long-lasting problem, which is obviously of importance for the rehabilitation of these patients [112, 126]. Two previous findings in unilateral vestibular loss patients are in good agreement with these conclusions: the deviation in the subjective vertical of these patients could last up to 4 years [127, 128]. Also, their trunk stabilization in the roll plane remained impaired up to 3 months after the lesion, during a natural task of knee bends [87] and complex gait tasks [129].

gies during the first hours following the lesion. Second, extra- and intracellular recordings in the in vitro whole brain between the 3rd and 7th day following the lesion show that the compensation process of the static oculomotor and postural syndromes mainly relies on synaptic changes in the vestibular-related networks embedding the deafferented and intact vestibular neurons. Third, extra- and intracellular recordings of MVNn in slice preparations indicate that, beginning a few days, maybe a few hours [131] after the lesion and evolving for at least a month, vestibular compensation of both the static and dynamic syndromes must increasingly rely on plastic changes in the intrinsic membrane properties of the vestibular nucleus neurons, and of neurons elsewhere in the CNS such as in the spinal cord. In particular, on the side of the lesion, it would progressively allow MVNn to regain a normal resting discharge in the absence of tonic vestibular excitatory drive. On the intact side, we have uncovered an increase in the MVNn sensitivity. It would allow them to maintain a normal resting activity and a correct sensitivity in the middle frequency range, in the absence of commissural inhibition. Comparison of various animal models in the literature and clinical research tends to show that the respective time course and importance of each of these three mechanisms could be species dependent. In particular, while sensory substitution and adaptive context-specific behavioral strategies appear to be common at the early stage following a vestibular loss, whatever the species considered, we suspect that they play a more important role at later stages in humans and primates than in non-primate mammalian or non-mammalian species. Encephalization would favor the appearance of various strategies to improve the recovery from the lesion-induced deficits. For instance, a fovea promotes the possibility to cancel by fixation the postlesional nystagmus. A succession of small saccades can substitute the missing slow phase at high velocities of head rotation. A sudden block of head movement reveals that a pseudoslow phase can be generated. This would explain and underline the importance of an adequate rehabilitation in vestibular loss patients.

Conclusion

Based on the results obtained in vivo, in slice and in the in vitro whole brain [29, 56], we proposed that vestibular compensation followed a top-down strategy [130]. First, in vivo recordings in the alert guinea pig show that vestibular compensation relies initially on the external cues given by the intact proprioceptive and visual systems. They are used to elaborate alternative motor strateImplications for Clinical Research in Vestibular Compensation

Neuroembryol Aging 2004–05;3:194–206

203

References 1 Gdowski GT, McCrea RA: Integration of vestibular and head movement signals in the vestibular nuclei during whole-body rotation. J Neurophysiol 1999; 82:436–449. 2 Gdowski GT, McCrea RA: Neck proprioceptive inputs to primate vestibular nucleus neurons. Exp Brain Res 2000;135:511–526. 3 Roy JE, Cullen KE: A neural correlate for vestibulo-ocular reflex suppression during voluntary eye-head gaze shifts. Nat Neurosci 1998;1:404–410. 4 Roy JE, Cullen KE: Selective processing of vestibular reafference during self-generated head motion. J Neurosci 2001;21:2131–2142. 5 Fukushima M, Kitahara T, Takeda N, Saika T, Uno A, Kubo T: Role of cholinergic mossy fibers in medial vestibular and prepositus hypoglossal nuclei in vestibular compensation. Neuroscience 2001;102:159–166. 6 Colletti V, Fiorino FG, Carner M: Investigations in electrophysiology and morphology of the VIII cranial nerve. Rev Laryngol Otol Rhinol (Bord) 1993;114:107–111. 7 Kitamura K, Takahashi K, Kakoi H, Sugimoto T, Silverstein H: A morphological and morphometric study of the peripheral process of the human vestibular nerve following posterior cranial fossa neurectomy. J Laryngol Otol 1999; 113:967–972. 8 Ris L, de Waele C, Serafin M, Vidal PP, Godaux E: Neuronal activity in the ipsilateral vestibular nucleus following unilateral labyrinthectomy in the alert guinea pig. J Neurophysiol 1995;74:2087–2099. 9 Ris L, Godaux E: Neuronal activity in the vestibular nuclei after contralateral or bilateral labyrinthectomy in the alert guinea pig. J Neurophysiol 1998; 80:2352–2367. 10 McKinley PA, Peterson BW: Voluntary modulation of the vestibulo-ocular reflex in humans and its relation to smooth pursuit. Exp Brain Res 1985;60:454–464. 11 Paige GD: Nonlinearity and asymmetry in the human vestibulo-ocular reflex. Acta Otolaryngol 1989;108:1–8. 12 Haslwanter T, Curthoys IS, Black RA, Topple AN, Halmagyi GM: The three-dimensional human vestibulo-ocular reflex: response to long-duration yaw angular accelerations. Exp Brain Res 1996;109:303–311. 13 Vibert N, de Waele C, Escudero M, Vidal PP: The horizontal vestibulo-ocular reflex in the hemilabyrinthectomized guinea-pig. Exp Brain Res 1993;97:263–273. 14 Gilchrist DP, Curthoys IS, Cartwright AD, Burgess AM, Topple AN, Halmagyi M: High acceleration impulsive rotations reveal severe long-term deficits of the horizontal vestibulo-ocular reflex in the guinea pig. Exp Brain Res 1998;123:242–254. 15 Hamann KF, Reber A, Hess BJ, Dieringer N: Long-term deficits in otolith, canal and optokinetic ocular reflexes of pigmented rats after unilateral vestibular nerve section. Exp Brain Res 1998;118:331–340.

204

16 Broussard DM, Bhatia JK, Jones GE: The dynamics of the vestibulo-ocular reflex after peripheral vestibular damage. I. Frequencydependent asymmetry. Exp Brain Res 1999; 125:353–364. 17 Lasker DM, Hullar TE, Minor LB: Horizontal vestibuloocular reflex evoked by high-acceleration rotations in the squirrel monkey. III. Responses after labyrinthectomy. J Neurophysiol 2000; 83:2482–2496. 18 Murai N, Tsuji J, Ito J, Mishina M, Hirano T: Vestibular compensation in glutamate receptor delta-2 subunit knockout mice: dynamic property of vestibulo-ocular reflex. Eur Arch Otorhinolaryngol 2004; 261: 82– 86. 19 Allum JHJ, Yamane M, Pfaltz CR: Long-term modifications of vertical and horizontal vestibulo-ocular reflex dynamics in man. Acta Otolaryngol 1988;105:328–337. 20 Allum JH, Ledin T: Recovery of vestibuloocular reflex-function in subjects with an acute unilateral peripheral vestibular deficit. J Vestib Res 1999;9:135–144. 21 Curthoys IS, Halmagyi GM: Vestibular compensation. Adv Otorhinolaryngol 1999; 55: 82–110. 22 Brandt T: Management of vestibular disorders. J Neurol 2000;247:491–499. 23 Jensen DW: Reflex control of acute postural asymmetry and compensatory symmetry after a unilateral vestibular lesion. Neuroscience 1979;4:1059–1073. 24 Sirkin DW, Precht W, Courjon JH: Initial, rapid phase of recovery from unilateral vestibular lesion in rat not dependent on survival of central portion of vestibular nerve. Brain Res 1984;302:245–256. 25 Schuknecht HF: Behavior of the vestibular nerve following labyrinthectomy. Ann Otol Rhinol Laryngol Suppl1982; 91:16–32. 26 Kunkel A, Dieringer N: Morphological and electrophysiological consequences of unilateral pre- versus postganglionic vestibular lesions in the frog. J Comp Physiol 1994; 174: 621–632. 27 Li H, Godfrey DA, Rubin AM: Comparison of surgeries for removal of primary vestibular inputs: a combined anatomical and behavioral study in rats. Laryngoscope 1995; 105:417–424. 28 Ris L, Capron B, de Waele C, Vidal PP, Godaux E: Dissociations between behavioural recovery and restoration of vestibular activity in the unilabyrinthectomized guinea-pig. J Physiol 1997;500:509–522. 29 Vibert N, Babalian A, Serafin M, Gasc JP, Mühlethaler M, Vidal PP: Plastic changes underlying vestibular compensation in the guinea-pig persist in isolated, in vitro whole brain preparations. Neuroscience 1999; 93: 413–432. 30 Dieringer N: Activity-related postlesional vestibular reorganization. Ann NY Acad Sci 2003;1004:50–60.

Neuroembryol Aging 2004–05;3:194–206

31 Llinás R, Walton K: Vestibular compensation: a distributed property of the central nervous system; in Asanuma H, Wilson VJ (eds): Integration in the Nervous System. Tokyo, Igaku Shoin, 1979, pp 145–166. 32 Straka H, Dieringer N: Spinal plasticity after hemilabyrinthectomy and its relation to postural recovery in the frog. J Neurophysiol 1995;73:1617–1631. 33 Smith PF, Curthoys IS: Mechanisms of recovery following unilateral labyrinthectomy: a review. Brain Res Rev 1989; 14: 155– 180. 34 Curthoys IS, Halmagyi GM: Vestibular compensation: a review of the oculomotor, neural and clinical consequences of unilateral vestibular loss. J Vestib Res 1995;5:67–107. 35 Dieringer N: ‘Vestibular compensation’: neural plasticity and its relations to functional recovery after labyrinthine lesions in frogs and other vertebrates. Prog Neurobiol 1995;46:97–129. 36 Vibert N, de Waele C, Serafin M, Babalian A, Mühlethaler M, Vidal PP: The vestibular system as a model of sensorimotor transformations. A combined in vivo and in vitro approach to study the cellular mechanisms of gaze and posture stabilization in mammals. Prog Neurobiol 1997;51:243–286. 37 Kitahara T, Takeda N, Kiyama H, Kubo T: Molecular mechanisms of vestibular compensation in the central vestibular system – review. Acta Otolaryngol Suppl 1998; 539: 19–27. 38 Vidal PP, de Waele C, Vibert N, Mühlethaler M: Vestibular compensation revisited. Otolaryngol Head Neck Surg 1998;119:34–42. 39 Curthoys IS: Vestibular compensation and substitution. Curr Opin Neurol 2000;13:27– 30. 40 Darlington CL, Dutia MB, Smith PF: The contribution of the intrinsic excitability of vestibular nucleus neurons to recovery from vestibular damage. Eur J Neurosci 2002; 15: 1719–1727. 41 Straka H, Vibert N, Vidal PP, Moore LE, Dutia MB: Intrinsic membrane properties of vertebrate vestibular neurons: function, development and plasticity. Prog Neurobiol 2005;76:349–392. 42 Dutia MB: Mechanisms of vestibular compensation; in Luxon L, Davies R (eds): Handbook of Vestibular Rehabilitation. London, Whurr Press, 2005. 43 Gallagher JP, Lewis MR, Shinnick-Gallagher P: An electrophysiological investigation of the rat medial vestibular nucleus in vitro; in Correia MJ, Perachio AA (eds): Progress in Clinical and Biological Research. New York, Liss, 1985, vol 176, pp 293–304. 44 Lewis MR, Gallagher JP, Shinnick-Gallagher P: An in vitro brain slice preparation to study the pharmacology of central vestibular neurons. J Pharmacol Methods 1987; 18: 267– 273.

Vidal/Straka/Vibert/Moore/de Waele

45 Lewis MR, Phelan KD, Shinnick-Gallagher P, Gallagher JP: Primary afferent excitatory synaptic transmission recorded intracellularly in vitro from rat medial vestibular neurons. Synapse 1989;3:149–153. 46 Serafin M, de Waele C, Katheb A, Vidal PP, Mühlethaler M: Medial vestibular nucleus in the guinea pig. I. Intrinsic membrane properties in brainstem slices. Exp Brain Res 1991;84:417–425. 47 Serafin M, de Waele C, Katheb A, Vidal PP, Mühlethaler M: Medial vestibular nucleus in the guinea-pig. II. Ionic basis of the intrinsic membrane properties in brainstem slices. Exp Brain Res 1991;84:426–433. 48 Uno A, Idoux E, Beraneck M, Hachemaoui M, Vidal PP, Moore LE, Wilson VJ, Vibert N: Static and dynamic membrane properties of lateral vestibular nucleus neurons in guinea pig brainstem slices. J Neurophysiol 2003;90: 1689–1703. 49 Dutia MB, Lotto RB, Johnston AR: Post-natal development of tonic activity and membrane excitability in mouse medial vestibular nucleus neurones. Acta Otolaryngol Suppl 1995; 520:101–104. 50 du Lac S, Lisberger SG: Membrane and firing properties of avian medial vestibular nucleus neurons in vitro. J Comp Physiol 1995; 176:641–651. 51 Johnston AR, MacLeod NK, Dutia MB: Ionic conductances contributing to spike repolarization and after-potentials in rat medial vestibular nucleus neurones. J Physiol 1994; 481:61–77. 52 Shimazu H, Precht W: Tonic and kinetic responses of cat’s vestibular neurons to horizontal angular acceleration. J Neurophysiol 1965;28:991–1013. 53 Grossman GE, Leigh RJ, Abel LA, Lanska DJ, Thurston SE: Frequency and velocity of rotational head perturbations during locomotion. Exp Brain Res 1988;70:470–476. 54 Pozzo T, Berthoz A, Lefort L, Vitte E: Head stabilization during various locomotor tasks in humans. II. Patients with bilateral peripheral vestibular deficits. Exp Brain Res 1991; 85:208–217. 55 Pozzo T, Berthoz A, Lefort L: Head stabilization during various locomotor tasks in humans. I. Normal subjects. Exp Brain Res 1990;82:97–106. 56 Babalian A, Vibert N, Serafin M, Mühlethaler M, Vidal PP: Central vestibular networks: functional characterization in the isolated, in vitro whole brain of guinea-pig. Neuroscience 1997;81:405–426. 57 Straka H, Beraneck M, Rohregger M, Moore LE, Vidal PP, Vibert N: Second-order vestibular neurons form separate populations with different membrane and discharge properties. J Neurophysiol 2004; 92:845–861. 58 Av-Ron E, Vidal PP: Intrinsic membrane properties and dynamics of medial vestibular neurons: a simulation. Biol Cybern 1999; 80:383–392.

Implications for Clinical Research in Vestibular Compensation

59 Moore LE, Hachemaoui M, Idoux E, Vibert N, Vidal PP: The linear and non-linear relationships between action potential discharge rates and membrane potential in model vestibular neurons. Nonlinear Stud 2004; 11: 423–447. 60 Ris L, Capron B, Vibert N, Vidal PP, Godaux E: Modification of the pacemaker activity of vestibular neurones in brain stem slices during vestibular compensation in the guineapig. Eur J Neurosci 2001;13:2234–2240. 61 Sekirnjak C, du Lac S: Intrinsic firing dynamics of vestibular nucleus neurons. J Neurosci 2002;22:2083–2095. 62 Beraneck M, Hachemaoui M, Idoux E, Ris L, Uno A, Godaux E, Vidal PP, Moore LE, Vibert N: Long-term plasticity of ipsilesional medial vestibular nucleus neurons after unilateral labyrinthectomy. J Neurophysiol 2003;90:184–203. 63 Beraneck M, Idoux E, Uno A, Vidal PP, Moore LE, Vibert N: Unilateral labyrinthectomy modifies the membrane properties of contralesional vestibular neurons. J Neurophysiol 2004;92:1668–1684. 64 Gacek RR, Lyon MJ, Schoonmaker JE: Ultrastructural changes in contralateral vestibulo-ocular neurons following vestibular neurectomy in the cat. Acta Otolaryngol Suppl 1991;477:1–14. 65 Gacek RR, Schoonmaker JE: Transneuronal down regulation of vestibulo-ocular neurons following vestibular ablation. Acta Otolaringol 1997;117:810–814. 66 Fetter M, Zee DS: Recovery from unilateral labyrinthectomy in rhesus monkey. J Neurophysiol 1988;59:370–393. 67 Gresty MA, Hess K, Leech J: Disorders of the vestibulo-ocular reflex producing oscillopsia and mechanisms compensating for loss of labyrinthine function. Brain 1977; 100: 693– 716. 68 Kasai T, Zee DS: Eye-head coordination in labyrinthine-defective human beings. Brain Res 1978;144:123–141. 69 Baloh RW, Honrubia V, Yee RD, Hess K: Changes in the human vestibulo-ocular reflex after loss of peripheral sensitivity. Ann Neurol 1984;16:222–228. 70 Jenkins HA: Long-term adaptive changes of the vestibulo-ocular reflex in patients following acoustic neuroma surgery. Laryngoscope 1985;95:1224–1234. 71 Katsarkas A, Galiana HL, Smith HL: Vestibulo-ocular reflex (VOR) biases in normal subjects and patients with compensated vestibular loss. Acta Otolaryngol 1995;115:476– 483. 72 Fetter M, Dichgans J: Adaptive mechanisms of VOR compensation after unilateral peripheral vestibular lesions in humans. J Vestib Res 1990;1:9–22. 73 Kubo T, Doi K, Koizuka I, Takeda N, Dugiyama N, Yamada K, Kohmura E, Hayakawa T: Assessment of auditory and vestibular functions after vestibular neurectomy for Meniere’s disease. Acta Otolaryngol 1995; 115:149–153.

74 Brantberg K, Fransson PA, Magnusson M, Johansson R, Bergenius JR: Short vestibuloocular reflex time-constant in complete unilateral vestibular lesions. Am J Otol 1995;16: 787–792. 75 Brantberg K, Magnusson M: The dynamics of the vestibulo-ocular reflex in patients with vestibular neuritis. Am J Otolaryngol 1990;11:345–351. 76 Escudero M, de Waele C, Vibert N, Berthoz A, Vidal PP: Saccadic eye movements and the horizontal vestibulo-ocular and vestibulocollic reflexes in the intact guinea-pig. Exp Brain Res 1993;97:254–262. 77 Curthoys IS, Topple AN, Halmagyi GM: Unilateral vestibular deafferentation (UVD) causes permanent asymmetry in the gain of the yaw VOR to high acceleration head impulses in guinea pigs. Acta Otolaryngol Suppl 1995;520:59–61. 78 Minor LB, Lasker DM, Backous DD, Hullar TE: Horizontal vestibuloocular reflex evoked by high-acceleration rotations in the squirrel monkey. I. Normal responses. J Neurophysiol 1999;82:1254–1270. 79 Halmagyi GM, Aw ST, Cremer PD, Todd MJ, Curthoys IS: The human vertical vestibuloocular reflex in response to high-acceleration stimulation after unilateral vestibular neurectomy. Ann NY Acad Sci 1992; 656: 732–738. 80 Aw ST, Haslwanter T, Halmagyi GM, Curthoys IS, Yavor RA, Todd MJ: Three-dimensional vector analysis of the human vestibuloocular reflex in response to high-acceleration head rotations – I. Responses in normal subjects. J Neurophysiol 1996; 76: 4009– 4020. 81 Aw ST, Haslwanter T, Halmagyi GM, Curthoys IS, Yavor RA, Todd MJ: Three-dimensional vector analysis of the human vestibuloocular reflex in response to high-acceleration head rotations – II. Responses in subjects with unilateral vestibular loss and selective semicircular canal occlusion. J Neurophysiol 1996; 76:4021–4030. 82 Aw ST, Halmagyi GM, Black RA, Curthoys IS, Yavor RA, Todd MJ: Head impulses reveal loss of individual semicircular canal function. J Vestib Res 1999;9:173–180. 83 Cremer PD, Halmagyi GM, Aw ST, Curthoys IS, McGarvie LA, Todd MJ, Black RA, Hannigan IP: Semicircular canal plane head impulses detect absent function of individual semicircular canals. Brain 1998; 121: 699– 716. 84 Schmid-Priscoveanu A, Straumann D, Böhmer A, Obzina H: Vestibulo-ocular responses during static head roll and three-dimensional head impulses after vestibular neuritis. Acta Otolaryngol 1999;119:750–757. 85 Fish U: The vestibular responses following unilateral vestibular neurectomy. Acta Otolaryngol 1973;76:229–238. 86 Borel L, Lacour M: Functional coupling of the stabilizing eye and head reflexes during horizontal and vertical linear motion in the cat. Exp Brain Res 1992;91:191–206.

Neuroembryol Aging 2004–05;3:194–206

205

87 Borel L, Harlay F, Magnan J, Chays A, Lacour M: Deficits and recovery of head and trunk orientation and stabilization after unilateral vestibular loss. Brain 2002; 125:880–894. 88 Borel L, Harlay F, Lopez C, Magnan J, Chays A, Lacour M: Walking performance of vestibular-defective patients before and after unilateral vestibular neurotomy. Behav Brain Res 2004;150:191–200. 89 Lacour M, Barthelemy J, Borel L, Magnan J, Xerri C, Chays A, Ouaknine M: Sensory strategies in human postural control before and after unilateral vestibular neurotomy. Exp Brain Res 1997;115:300–310. 90 Dichgans J, Bizzi E, Morasso P, Tagliasco V: Mechanisms underlying recovery of eyehead coordination following bilateral labyrinthectomy in monkeys. Exp Brain Res 1973;18:548–562. 91 Hikosaka O, Maeda M: Cervical effects on abducens motoneurons and their interaction with vestibulo-ocular reflex. Exp Brain Res 1973;18:12–30. 92 Kobayashi Y, Yagi T, Kamio T: Cervico-vestibular interaction in eye movements. Auris Nasus Larynx Suppl 1986; 13:S87–S95. 93 Yagi T, Ohyama Y: Three-dimensional analysis of nystagmus induced by neck vibration. Acta Otolaryngol 1996;116:167–169. 94 Yagi T, Hatano G, Morizono T: Role of dorsal neck proprioceptive inputs to vestibular compensation in humans. Nippon Ika Daigaku Zasshi 1998;65:291–297. 95 Strupp M, Arbusow V, Dieterich M, Sautier W, Brandt T: Perceptual and oculomotor effects of neck muscle vibration in vestibular neuritis. Ipsilateral somatosensory substitution of vestibular function. Brain 1998; 12:677–685. 96 Bronstein AM: Plastic changes in the human cervico-ocular reflex. Ann NY Acad Sci 1992:656:708–715. 97 Ivanenko YP, Viaud-Delmon I, Semont A, Gurfinkel VS, Berthoz A: Eye movements during torso rotations in labyrinthine-defective subjects. J Vestib Res 1999; 9: 401– 412. 98 Berthoz A: The role of gaze in compensation of vestibular dysfunction: the gaze substitution hypothesis. Prog Brain Res 1988;76:411–420. 99 Fukushima K: Cortico-vestibular interactions: anatomy, electrophysiology, and functional considerations. Exp Brain Res 1997;117:1–16. 100 Nashner LM, McCollum G: The organization of human postural movements: a formal basis and experimental synthesis. Behav Brain Sci 1985;8:35–172. 101 Vidal PP, Graf W, Berthoz A: The orientation of the cervical vertebral column in unrestrained awake animals. I. Resting position. Exp Brain Res 1986;61:549–559. 102 de Waele C, Graf W, Josset P, Vidal PP: A radiological analysis of the postural syndromes following global and selective lesions in the guinea pig. Exp Brain Res 1989; 77:166–182.

206

103 Graf W, de Waele C, Vidal PP: Skeletal geometry in vertebrates and its relation to the vestibular endorgans; in Berthoz A, Graf W, Vidal PP (eds): The Head-Neck SensoryMotor System. Oxford, Oxford University Press, 1992, pp 129–134. 104 Graf W, de Waele C, Vidal PP: Functional anatomy of the head-neck movement system of quadrupedal and bipedal mammals. J Anat 1995;186:55–74. 105 Graf W, de Waele C, Vidal PP, Wang DH, Evinger C: The orientation of the cervical vertebral column in unrestrained awake animals. II. Movement strategies. Brain Behav Evol 1995;45:209–231. 106 Vidal PP, Degallaix L, Josset P, Gasc JP, Cullen KE: Postural and locomotor control in normal and vestibularly deficient mice. J Physiol 2004;559:625–638. 107 Vidal PP, Wang DH, Graf W, de Waele C: Vestibular control of skeletal geometry in the guinea pig: a problem of good trim? Prog Brain Res 1993;97:229–243. 108 Mbongo F, Patko T, Vibert N, Tran Ba Huy P, Vidal PP, de Waele C: Postural control in patients with unilateral vestibular lesions is more impaired in the frontal than in the sagittal plane: a static and dynamic posturography study. Audiol Neurootol 2005; 10:291–302. 109 Kitahara T, Takeda N, Saika T, Kubo T, Kiyama H: Role of the flocculus in the development of vestibular compensation: immunohistochemical studies with retrograde tracing and flocculectomy using Fos expression as a marker in the rat brainstem. Neuroscience 1997;76:571–580. 110 Kitahara T, Fukushima M, Takeda N, Saika T, Kubo T: Effects of pre-flocculectomy on Fos expression and NMDA receptor-mediated neural circuits in the central vestibular system after unilateral labyrinthectomy. Acta Otolaryngol 2000;120:866–871. 111 Johnston AR, Seckl JR, Dutia MB: Role of the flocculus in mediating vestibular nucleus neuron plasticity during vestibular compensation in the rat. J Physiol 2002; 545: 903–911. 112 Strupp M, Arbusow V, Maag KP, Gall C, Brandt T: Vestibular exercises improve central vestibulospinal compensation after vestibular neuritis. Neurology 1998; 51: 838–844. 113 Curthoys IS, Dai MJ, Halmagyi GM: Human ocular torsional position before and after unilateral vestibular neurectomy. Exp Brain Res 1991;85:218–225. 114 Wolfe GI, Taylor L, Flamm ES, Gray LG, Raps EC, Galetta SL: Ocular tilt reaction resulting from vestibuloacoustic nerve surgery. Neurosurgery 1993;32:417–420. 115 Safran AB, Vibert D, Issoua D, Hausler R: Skew deviation after vestibular neuritis. Am J Ophthalmol 1994;118:238–245. 116 Brandt T: Man in motion – historical and clinical aspects of vestibular function – a review. Brain 1991;114:2159–2174.

Neuroembryol Aging 2004–05;3:194–206

117 Brandt T: Vertigo – Its Multisensory Syndromes. London, Springer, 1991. 118 Brandt T, Dieterich M: Pathological eyehead coordination in roll: tonic ocular tilt reaction in mesencephalic and medullary lesions. Brain 1987;110:649–666. 119 Brandt T, Dieterich M: Vestibular syndromes in the roll plane: topographic diagnosis from brainstem to cortex. Ann Neurol 1994; 36:337–347. 120 Taguchi K, Sasaki O, Sato K, Nezu K, Sakaguchi M: Subjective vertical and vestibular lesion. Acta Otolaryngol Suppl 1995; 519: 201–203. 121 Maki BE, Holliday PJ, Topper AK: A prospective study of postural balance and risk of falling in an ambulatory and independent elderly population. J Gerontol 1994; 2: M72–M74. 122 Carpenter MG, Allum JHJ, Honegger F: Directional sensitivity of stretch reflexes and balance corrections for normal subjects in the roll and pitch planes. Exp Brain Res 1999;129:93–113. 123 Matjacic Z, Voigt M, Popovic D, Sinkjaer T: Functional postural responses after perturbations in multiple directions in a standing man: a principle of decoupled control. J Biomech 2001;34:187–196. 124 Allum JHJ, Adkin AL, Carpenter MG, Held-Ziolkowska M, Honneger F, Pierchala K: Trunk sway measures of postural stability during clinical balance tests: effects of a unilateral vestibular deficit. Gait Posture 2001;14:227–237. 125 Rushmer DS, Moore SP, Windus SL, Russell CJ: Automatic postural responses in the cat: responses of hindlimb muscles to horizontal perturbations of stance in multiple directions. Exp Brain Res 1988;71:93–102. 126 Lacour M: Relearning and critical postoperative period in the restoration of nerve function. Example of vestibular compensation and clinical implications (in French). Ann Otolaryngol Chir Cervicofac 1984; 101:177–187. 127 Vibert D, Hausler R, Safran AB: Subjective visual vertical in peripheral unilateral vestibular diseases. J Vestib Res 1999; 9: 145– 152. 128 Vibert D, Hausler R: Long-term evolution of subjective visual vertical after vestibular neurectomy and labyrinthectomy. Acta Otolaryngol 2000;120:620–622. 129 Allum JHJ, Adkin AL: Improvements in trunk sway observed for stance and gait tasks during recovery from an acute unilateral peripheral vestibular deficit. Audiol Neurootol 2003;8:286–302. 130 Vibert N, Bantikyan A, Babalian A, Serafin M, Mühlethaler M, Vidal PP: Post-lesional plasticity in the central nervous system of the guinea-pig: a ‘top-down’ adaptation process? Neuroscience 1999;94:1–5. 131 Cameron SA, Dutia MB: Cellular basis of vestibular compensation: changes in intrinsic excitability of MVN neurones. Neuroreport 1997;8:2595–2599.

Vidal/Straka/Vibert/Moore/de Waele

Neuroembryol Aging 2004–05;3:207–214 DOI: 10.1159/000096798

Published online: November 3, 2006

Peripheral Vestibular Responses to Sound Ian S. Curthoys Vestibular Research Laboratory, School of Psychology, University of Sydney, Sydney, Australia

Key Words Bone conduction
Labyrinth
Myogenic potential
Otolith
Saccule
VEMP
Vibration

Abstract As well as activating the cochlea, sound also activates some vestibular receptors and afferents. That is true for both airand bone-conducted sound and this review briefly summarizes the neural evidence for vestibular activation by sound in mammals and considers the implications of this result. These neural results have been rapidly translated into clinical use and underpin a new clinical test of human vestibular function using sound as the test stimulus. Copyright © 2005 S. Karger AG, Basel

Introduction

There is excellent evidence from the frog that some afferent neurons, originating from sensory regions in the inner ear which are generally classified as being vestibular, respond to sound – either air-conducted (AC) or bone-conducted (BC) sound. In some cases these neurons are quite exquisitely sensitive to BC sound, showing clear modulation to accelerations as small as 10–6 g [1, 2]. However, in the frog labyrinth, the distinction between auditory and vestibular sensory regions is not straightforward so there remains a question as to whether the responses really can be classed as activation of vestibular

© 2005 S. Karger AG, Basel 1661–3406/05/0034–0207$22.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

sensory regions by sound. In mammals the distinction between auditory and vestibular sensory regions is more clear-cut and the evidence is unequivocal; some vestibular receptors and afferents in mammals do respond to sound at physiological levels. In normal healthy animals with normal intact labyrinths it can be shown that sound activates specific categories of vestibular neurons, in some cases at very low intensities. If the bony wall of the labyrinth is opened (a fenestration or dehiscence), then the categories of activated vestibular neurons are much larger and the intensities of sound required are smaller. This rather unusual fenestration procedure has its origins in the historical development of the area where an early investigator of the behavioral responses to sound, Tullio, found that such behavioral responses could only be demonstrated after an opening had been made in the bony wall of the labyrinth so that there was an artificial opening into the perilymphatic space, as described by Russolo [3]. It appears that such a procedure effectively removes a natural ‘block’ which prevents many vestibular neurons from responding to sound.

Vestibular Primary Neurons

Vestibular primary afferents can be set along a continuum according to the regularity of their resting discharge (defined by the coefficient of variation of interspike intervals) and there are many functional differences which are related to the regularity of afferent discharge

Ian S. Curthoys Vestibular Research Laboratory, School of Psychology University of Sydney Sydney, NSW 2006 (Australia) Tel. +61 2 9351 3570, Fax +61 2 9036 5223, E-Mail [email protected]

[4, for review see Goldberg et al. 5]. Irregular afferent neurons tend to be large diameter fibers which originate predominantly from the amphora-shaped type I receptors with large calyx endings predominantly in the central striolar region of the vestibular sensory regions, and they show a clear preference for changes in acceleration stimuli (jerk), rather than for maintained accelerations. Regular afferent neurons tend to be smaller diameter fibers which originate from more peripheral areas in the vestibular sensory regions, predominantly from small bouton endings on the barrel-shaped type II receptors, and these neurons tend to respond to maintained accelerations. However, there is a continuum of regularity of spontaneous discharge and this simplification of cells as ‘irregular’ and ‘regular’ is used simply to contrast the two extremes of this continuum, whilst paying due homage to it. Extracellular recordings from single primary vestibular neurons in the mouse, guinea pig, chinchilla, cat and squirrel monkey have shown, in all of these species, that primary neurons from sensory regions, which in mammals are acknowledged to be vestibular (e.g. the cristae of the semicircular canals, and the utricular and saccular maculae), respond to sound – clicks, tones and short tone bursts [6–14]. Two major questions must be answered: (1) These sound-sensitive primary vestibular neurons travel in the VIIIth nerve together with cochlear neurons. How can we be certain these really are vestibular neurons and not cochlear neurons? (2) How can we be sure the response was to stimuli of physiological intensities and was not some aberrant response to unphysiologically intense stimuli? The answer to the first question is that vestibular neurons travel in a distinctly different bundle within the VIIIth nerve compared to cochlear neurons. Vestibular afferents have their cell bodies located in Scarpa’s ganglion which is dorsal to and separate from the bundle of auditory nerve fibers of the VIIIth nerve. The neurons which have been reported to be activated by sound are located in the vestibular division of the VIIIth nerve and not the auditory division of the VIIIth nerve. In some studies, these neurons were additionally confirmed as being vestibular by virtue of their response to natural vestibular stimulation such as angular and linear accelerations. In a few cases, sound-activated neurons have been backfilled with horseradish peroxidase or neurobiotin and the axons traced back to their site of origin in vestibular sensory regions. Evoked responses to sound can still be recorded after auditory receptors have been de208

Neuroembryol Aging 2004–05;3:207–214

stroyed by aminoglycoside antibiotics [15]. All of these lines of evidence support the identification of the neurons as vestibular and not cochlear. [It is conceivable that these neurons may be efferent fibers but that remote possibility is excluded because the neurons are activated at such short latencies after the onset of the sound stimulus (less than 1 ms for AC stimuli and around 1–1.5 ms for BC stimuli)]. The answer to the second question is that the response of many of these neurons occurs at stimulus intensities which are well within the physiological range. That has been demonstrated by relating the auditory stimulus intensity for eliciting the response from a vestibular neuron to the animal’s own averaged auditory brainstem response (ABR) threshold for the same auditory stimulus. So for AC sound the usual thresholds of responding by vestibular neurons are about 70 dB above the animal’s own ABR threshold. For BC sound the thresholds may be only 10–15 dB above the animal’s own ABR threshold. In both cases, the response is well within the physiological range – these are not artifactual neural responses to abnormally intense stimuli. In fact, for BC sound the standard transducer itself (the Radioear B-71 bone oscillator) poses a physical limit on the largest stimuli which can be delivered and the clear result is that some vestibular neurons (irregular otolithic neurons) can be evoked at very low intensities by that device, whereas other neurons (e.g. most semicircular canal neurons) cannot be driven even by the largest stimuli which the B-71 can deliver (about 88 g). Absolute measures of intensity in dB or g units are not as meaningful as measures referred to a biological reference for that animal since the biological reference shows the functional effectiveness of the stimulus. For example, if there is a conductive hearing loss the absolute level of an AC stimulus is attenuated before it even reaches the receptors.

Physiological Evidence – Recordings from Animals with Intact Labyrinths

Recordings in the cat and guinea pig have shown that in animals with bony labyrinth intact, it is the irregular otolithic neurons which preferentially show an increase in firing in response to AC sound [8, 9, 12]. Regular otolithic neurons and also semicircular canal neurons (be they regular or irregular) show little evidence of such activation by sound up to the maximum intensities used in those studies. Physiological and anatomical evidence Curthoys

from both the cat and guinea pig identify the saccular macula as being the vestibular sensory region preferentially activated by AC sound [8–12, 16]. It is irregular saccular neurons which are responsive to AC sound, and given other evidence about the site of origin of irregular neurons (see Goldberg [4] for a summary) these irregular neurons probably originate predominantly from the striola of the saccule. Information about the particular vestibular sensory region of origin is provided by the division of the vestibular nerve from which the cell originated – afferents from the utricular macula course in the superior division and afferents from the saccular macula course in both the inferior and partly in the superior division of the nerve [17]. Combining this anatomical evidence with the evidence about response to natural vestibular stimulation allows us to assign neural responses to particular vestibular sensory regions. The most conclusive evidence comes from staining either into, or very close to, sound-responsive neurons and demonstrating the site of origin of such identified neurons is from the striola of the saccular or utricular maculae [11, 14]. With the bony casing of the labyrinth intact (i.e. the membranous labyrinth is encased in bone – as is normally the case), the vestibular sensory regions activated by AC sound are different from the regions activated by BC sound. AC sound preferentially activates irregular saccular neurons [8, 9], whereas BC sound appears to activate irregular neurons from both the saccular and utricular maculae [14]. It has been assumed that the bulk of neurons in the inferior vestibular nerve responding to static pitch tilts probably originates from the saccular macula and that neurons in the superior vestibular nerve responding to static roll tilts probably originate from the utricular macula [18]. These assignments to the saccular and utricular macula are not hard and fast since the anatomical configuration of the planes of the utricular and saccular maculae in the guinea pig [19] show that the saccular macula does have a projection into the frontal plane so some saccular neurons could show responses to static roll stimuli. Similarly, the utricular macula has a projection into the coronal plane, and some utricular neurons exhibit responses to static pitch stimuli [18]. Finally, some saccular afferents (from the ‘hook’ of the saccular macula) course in the superior division of the vestibular nerve. On the other hand, up to the maximum levels used in the study, none of the 137 regular canal neurons which were tested could be activated by BC sound, and only a small percentage of irregular canal neurons could be acPeripheral Vestibular Responses to Sound

Fig. 1. Summary of results from a recent study of guinea pig pri-

mary vestibular neurons to show the preferential activation of irregular otolithic neurons by BC sound; both regular () ) and irregular ($ ), activated neurons (%) – defined as showing a detectable increase in firing rate to BC sound – are depicted. A total of 189 canal neurons and 157 otolithic neurons were tested.

tivated, i.e. show a detectable increase in discharge rate to BC sounds (fig. 1). These irregular otolithic neurons which are activated fire at very short latencies. The average latency for AC clicks was around 0.9 ms. The latency for activation for BC sound is more difficult to give a simple value to, since a BC click or 7-ms 500-Hz tone burst causes the skull to ring with a damped oscillation (see accelerometer trace in fig. 2). A neuron may be activated on the first peak of that damped oscillation or on the second peak (1 ms or more later), so the measured latency may show a (misleading) quantal jump of 1 ms or so if the neuron is not activated by the first peak. Nevertheless, there is evidence from many neurons showing that the minimum latency for activation by BC sound is very similar to that for AC sound at around 1 ms. Some primary vestibular neurons in animals with normal intact labyrinths show sensitive responses to sound, especially to BC sound, although not as sensitive as neurons in the frog. The responsiveness of the various vestibular sensory regions is not uniform, with otolithic neurons tending to be activated at lower sound levels than neurons from other areas. The exact profile of the vestibular sensory regions activated by sound and the exact characteristics of the optimal stimuli, e.g. tuning curves or intensity-activation functions, remain to be determined because there have been so few studies, and those that have taken place have used different means of quantifying the stimulus intensity and different criteria as to Neuroembryol Aging 2004–05;3:207–214

209

50 μV

1g

0

10

20 Time (ms)

30

40

Fig. 2. An example of the activation of an otolithic irregular neu-

ron to BC sound – in this case a short tone burst of 500 Hz presented for 7 ms. Traces from 7 stimulus presentations are superimposed to show the activation of the neuron and the very tight locking (synchronization) of the neural response to the peaks in the stimulus. The spike waveform can be seen after the activation. The regularities in the baseline response are cochlear microphonics to this stimulus – the unit recording site is close to the auditory nerve (although not in it) and so the cochlear microphonic is picked up as well as the action potential from the neuron. The accelerometer trace shows that the BC stimulus causes a ringing of the skull which lasts far longer than the nominal 7-ms stimulus duration (from Curthoys et al. [14]).

what constitutes neural activation. The upper limit of the frequency activating otolithic irregular neurons to BC sound is still not clear. A few neurons show an increased firing for tones up to 1,000 Hz, but the frequency which seems to optimally activate most neurons is around 500 Hz. Vestibular afferents can show synchronization up to high frequencies (fig. 2), which implies that the receptors are responding systematically up to high frequencies. The clear result from the study by Murofushi et al. [12] on responses of vestibular neurons to AC sound has been confirmed with BC sound, and the conclusion is that under normal conditions semicircular canal neurons show very little activation for AC and BC stimuli at these reasonable stimulus intensities. Those few neurons which are activated are irregular neurons. 210

Neuroembryol Aging 2004–05;3:207–214

One of the conclusions is that the mechanoreceptor operation of the individual hair cell receptors can mediate responses to auditory frequencies even when those hair cells are located in the vestibular – ‘non-auditory’ – labyrinth. The implication is that there must exist in at least some vestibular receptor hair cells (probably the type I receptors) membrane mechanisms which can permit such a high frequency response. Irregular primary otolithic neurons show pronounced tuning for rapid accelerations – changes in linear acceleration (jerk) [4] – and such preferential tuning for dynamic stimuli may be responsible for the selective activation of these neurons by sound. One especially interesting aspect is that the irregular otolithic neurons which are preferentially activated by BC stimuli likely come from the type I receptors at the striola which are especially vulnerable to ototoxic antibiotics. Testing using sound may be one way of specifically monitoring the functional state of these receptors.

Criterion for Neural Activation

One major study [7] used an unusual criterion of neural activation – that of synchronization of action potentials to the stimulus waveform – rather than the criterion which is usual in the vestibular literature, that of a stimulus-locked detectable increase in firing. Synchronization is a standard index in auditory studies, but as far as the generation of vestibulo-ocular or vestibulo-spinal responses is concerned, changes in synchronization are probably of minor importance, as Young et al. [7, p. 358] noted in the squirrel monkey. They reported that sound stimuli changed the exact timing of the spikes in the spontaneously active primary vestibular neurons from all vestibular sensory regions and that this effect occurred well below the stimulus levels necessary to elicit a detectable increase in neural firing.

Physiological Evidence – Recordings from Animals with Labyrinth Fenestrations or Dehiscences

There is one rather unusual variable which affects the profile of vestibular sensory regions which respond to sound – whether the labyrinth is in its normal state, encased in bone, or whether an opening into the bony casing of the labyrinth (called a fenestration or a dehiscence) has been made, with the result that the perilymphatic space of the labyrinth has been opened. Tullio explored Curthoys

the eye movements and postural responses evoked by AC sound in animals but in order to elicit these responses he had to open the perilymphatic space of the labyrinth. Intense sounds have been known to generate eye, head and postural responses of alert animals since the early 1920s by the work of the Italian physiologist Tullio (described by Russolo [3]; see also Halmagyi et al. [20]). Vestibular behavioral and perceptual responses to sound have come to be referred to as Tullio phenomena. Obviously sound will activate cochlear receptors and neurons, so how can one be sure that the responses Tullio observed really are vestibular and not cochlear? The reasoning has been that the fast projections from the vestibular receptors to eye muscles and the vestibulospinal projections are potent and if vestibular receptors can be activated, even by what may be regarded as an inadequate stimulus (i.e. sound), then these same responses should occur. That does not exclude the possibility of direct auditory pathways to drive the responses, but some of the experimental evidence reviewed below suggests that auditory pathways are not the cause of the observed responses to sound. Mikaelian [6] showed the probable cause of the Tullio effect by recording from single primary vestibular neurons and demonstrating that they are indeed activated by AC sound, after the labyrinth has been opened. To solve the problem as to whether the responses could be of cochlear origin, Mikaelian [6] used mutant mice which had no cochleas but which still had an apparently normal vestibular system. He found that when the labyrinth was intact (sealed), vestibular afferents were not activated by sound even at high stimulus levels. However, when he carried out a dehiscence or fenestration so that the perilymphatic space was opened, then the previously unresponsive vestibular neurons showed clear increases in firing in response to AC pure tone stimuli. For the majority of afferent neurons this does not normally happen since the intact bony casing of the membranous labyrinth acts to prevent the hair cells being sufficiently stimulated by sound. This neural response after labyrinth dehiscence has recently been confirmed and extended by Carey et al. [13] in the chinchilla. In chinchillas with intact labyrinths there was little evidence of activation of semicircular canal neurons by AC sound. However after the labyrinth had been opened, Carey et al. [13] found that horizontal and superior semicircular canal afferent neurons showed strong increases in firing in response to AC sound. This evidence from Mikaelian [6] and Carey et al. [13] provides the neural basis for the behavioral effects which Tullio observed and the analogous semicircular canal de-

hiscence syndrome seen in human patients. Examples of this are patients with a thinning of the bony wall of the superior semicircular canal, where sound causes vestibular sensation (e.g. vertigo) and corresponding eye movements [21]. Such an opening changes the mode of operation of the human semicircular canal system so semicircular canal receptors become responsive to sound. This opening forms a ‘third window’ (in addition to the round and oval windows), and its presence results in canal neurons increasing their firing rate during a maintained AC tone. It seems that the firing of these canal neurons is synchronized to the stimulus waveform which suggests that the movement of the stapes during the tonal stimulus causes fluid movement which results in crista hair cell deflection at the tonal frequency. This then seems to be a mechanism which would account for some patient responses – after labyrinthine opening, loud AC sounds activate the previously unresponsive semicircular canal receptors and result in sensations and eye movements which are appropriate for semicircular canal stimulation but are inappropriate responses to sound stimulation [20]. Such an opening of the bony casing of the membranous labyrinth seems to change the whole mode of operation of the vestibular divisions of the labyrinth, so that once ‘the seal’ is broken by making such a fenestration then vestibular sensory regions, which are normally unresponsive to sound even at high intensities, respond to sound stimulation at low intensities [6, 13]. The implication is that the seal imposed by the bony casing of the labyrinth is an important factor in determining the normal operation of the vestibular sensory system.

Peripheral Vestibular Responses to Sound

Neuroembryol Aging 2004–05;3:207–214

The Receptor Hair Cell

Sound, vibration, angular head acceleration and head position are very different stimuli which pose considerable problems for transduction. It seems as if an ideal design for a mechanoreceptor transducer element has evolved – the receptor hair cell – and specialized sensory machinery has developed to allow these very different stimuli to bend the hair cell. So hair cells in semicircular canals respond sensitively to head angular acceleration by virtue of the fact that the duct and cupula function to deflect the cilia of the receptor hair cells during an angular acceleration; those in the otoliths respond sensitively to linear accelerations because the movement of the dense otoconia results in hair cell deflections. Obviously, there are differences in hair cells in different regions of the labyrinth – for example the cochlear hair cells have no kino211

cilium, whereas the vestibular receptors do; the height of the receptor hair cells varies across the macula surface; the efferent fibers from the brainstem have different patterns of innervation on cochlear and vestibular receptor cells, and even within the cochlea the pattern of innervation differs for inner as opposed to outer hair cells. Nevertheless, the basic structure and apparently the basic operation of the receptor hair cell is similar across labyrinthine sensory regions. By working in these various specialized structures such as the cochlea, the otoliths and the semicircular canals, the one basic receptor element can transduce all these various forces, given the labyrinthine machinery which has evolved to allow these various forces to optimally stimulate this ‘ideal’ receptor.

Vestibular Evoked Myogenic Potentials

Already the finding that auditory stimuli can be used to probe peripheral vestibular function has had clinical impact in that it provides the physiological basis for a new clinical test. Bickford et al. [22] reported that sound caused a short latency response at the inion which was affected by the tonic activation of the underlying neck muscles. Later recordings showed a large clear-cut myogenic response from the sternocleidomastoid muscle (SCM), and this change in potential has been called the vestibular evoked myogenic potential (VEMP) [23, 24]. Activation of vestibular afferents by sound (a brief click or short tone burst) can be used to generate a short latency (13 ms) inhibitory potential in tonically stretched SCM. The presence of this response confirms that the vestibular (and probably saccular) receptors in the stimulated ear are functional. The absence of the response is more difficult to interpret – it may be because of absent saccular function, but other factors, such as conductive hearing loss, may also prevent the response from being generated. The VEMP was shown to be vestibular and not cochlear since totally deaf subjects with remaining vestibular function still had a preserved VEMP while patients who had preserved hearing after selective surgical section of the vestibular nerve did not have a remaining VEMP.

The Neural Basis for the VEMP Test

The VEMP response is a very basic response since it has been demonstrated in guinea pigs, monkeys and humans, so the evidence from electrophysiology in the cat 212

Neuroembryol Aging 2004–05;3:207–214

Fig. 3. Schematic illustration of the major projections which un-

derlie the VEMP response. For clarity, only the receptors and connections for the left saccular macula are shown. The standard convention is that excitatory neurons (hexagons) and synapses (triangles) are unfilled whereas inhibitory neurons and synapses are filled (black). Primary afferent fibers from receptors in the saccular macula are excitatory and synapse on second order vestibular neurons in the ipsilateral vestibular nuclei. Some of these secondary neurons are inhibitory and project in the ipsilateral medial vestibulospinal tract to cervical motoneurons, specifically inhibiting motoneurons projecting to the ipsilateral SCM. However, some second order neurons in the vestibular nucleus receiving primary saccular afferent input are excitatory and project in the contralateral medial vestibulospinal tract to the contralateral extensor motoneurons. In this way, the same input from the saccular afferents (e.g. from an intense click or short tone burst) results in a short latency inhibitory response in ipsilateral SCM and a short latency excitatory response in contralateral neck extensor neurons.

and guinea pig can be used with a high degree of confidence to understand the probable neural pathways for the human response. The response in animals occurs in response to comparable acoustic stimuli and intensities to those used in human studies. That comparability is ensured by relating the intensity of the click or tone burst to the ABR threshold for each animal and ensuring that the intensities delivered to animals are a comparable number of dB above ABR threshold for the different species.

Curthoys

Uchino et al. [25] have established part of the connectivity of the neural pathways from the saccular macula by electrical stimulation of the saccular macula and recording second and third order neurons in the cat including the vestibulospinal projections of the saccular afferents. A recent review of the many experiments by Uchino et al. [25] in this area provides the basis for figure 2. There are short, fast disynaptic pathways from the saccular receptors to the motoneurons controlling the ipsilateral SCM (fig. 3). These consist of second order neurons in the ipsilateral vestibular nuclei, some of which are inhibitory and project to the ipsilateral SCM motoneurons. Other second order neurons are excitatory and project to the contralateral extensor motoneurons. It is unlikely that the ipsilateral and contralateral projections from the saccular macula are of equal potency, and indeed clinical measures typically show a strong short latency (13 ms) ipsilateral inhibitory response and a weaker contralateral excitatory response to the same click stimulus or to a short tone burst to one ear. Is the VEMP response due to activation of auditory receptors and pathways? This is unlikely. Firstly, as the stimulus frequency of the short tone burst increases up to 1,000 Hz and beyond, the audibility of the stimulus increases but the VEMP response decreases – in line with the progressively poorer high frequency response of primary saccular afferents [7]. Secondly, human subjects who have no hearing but still have vestibular function (e.g. calorics) still show VEMP responses similar to those in normal subjects, presumably because their inferior vestibular nerve is still functional. Recently BC stimuli have been used to generate VEMPs instead of AC stimuli. BC stimuli have the great advan-

tage that they are not affected by conductive hearing loss which is common in older subjects and prevents the use of AC stimuli for the VEMP test.

Conclusion

The data in this review show that neurons in the vestibular division of the labyrinth can respond to sound and in some cases at quite low intensities. The conclusion is that the distinction between the ‘auditory’ and ‘nonauditory’ sensory regions of the labyrinth is a convenient fiction which tends to obscure the fact that the receptor element in all labyrinthine sensory regions is an almost ideal mechanoreceptor, which can respond to any stimulus which can deflect its cilia, be they sounds or accelerations. Usually structural factors determine the stimuli to which the receptor will respond. What differs between sensory regions is the means by which the receptors are stimulated – the ‘machinery’ which has evolved to allow each of these forces to stimulate this ‘ideal’ receptor optimally. It seems that the sealed character of the labyrinth structure acts to prevent the receptor from responding to inappropriate stimuli. These sensitive receptors are stimulated once these factors are compromised (e.g. by opening the labyrinth).

Acknowledgments This work was supported by the National Health and Medical Research Council of Australia.

References 1 Koyama H, Lewis ER, Leverenz EL, Baird RA: Acute seismic sensitivity in the bullfrog ear. Brain Res 1982;250:168–172. 2 Narins PM, Lewis ER: The vertebrate ear as an exquisite seismic sensor. J Acoust Soc Am 1984;76:1384–1387. 3 Russolo M: Attualità delle ricerche di Pietro Tullio. Acta Otorhinolaryngol Ital 1998; 18: 342–347. 4 Goldberg JM: Afferent diversity and the organization of central vestibular pathways. Exp Brain Res 2000;130:277–297.

Peripheral Vestibular Responses to Sound

5 Goldberg JM, Desmadryl G, Baird RA, Fernandez C: The vestibular nerve of the chinchilla. V. Relation between afferent discharge properties and peripheral innervation patterns in the utricular macula. J Neurophysiol 1990;63:791–804. 6 Mikaelian D: Vestibular response to sound: single unit recording from the vestibular nerve in fenestrated deaf mice (Df/Df). Acta Otolaryngol 1964;58:409–422. 7 Young ED, Fernandez C, Goldberg JM: Responses of squirrel monkey vestibular neurons to audio-frequency sound and head vibration. Acta Otolaryngol 1977;84:352–360. 8 McCue MP, Guinan JJ: Acoustically responsive fibers in the vestibular nerve of the cat. J Neurosci 1994;14:6058–6070.

9 McCue MP, Guinan JJ: Influence of efferent stimulation on acoustically responsive vestibular afferents in the cat. J Neurosci 1994; 14:6071–6083. 10 McCue MP, Guinan JJ: Spontaneous activity and frequency selectivity of acoustically responsive vestibular afferents in the cat. J Neurophysiol 1995; 74:1563–1572. 11 McCue MP, Guinan JJ: Sound-evoked activity in primary afferent neurons of a mammalian vestibular system. Am J Otol 1997;18: 355–360. 12 Murofushi T, Curthoys IS, Topple AN, Colebatch JG, Halmagyi GM: Responses of guinea pig primary vestibular neurons to clicks. Exp Brain Res 1995;103:174–178.

Neuroembryol Aging 2004–05;3:207–214

213

13 Carey JP, Hirvonen TP, Hullar TE, Minor LB: Acoustic responses of vestibular afferents in a model of superior canal dehiscence. Otol Neurotol 2004; 25:345–352. 14 Curthoys IS, Kim J, McPhedran SK, Camp AJ: Bone conducted vibration selectively activates irregular primary otolithic vestibular neurons in the guinea pig. Exp Brain Res DOI: 10.1007/s00221–006–0544–1. 15 Cazals Y, Aran JM, Erre JP: Frequency sensitivity and selectivity of acoustically evoked potentials after complete cochlear hair cell destruction. Brain Res 1982;231:197–203. 16 Murofushi T, Curthoys IS: Physiological and anatomical study of click-sensitive primary vestibular afferents in the guinea pig. Acta Otolaryngol 1997;117:66–72.

214

17 Hardy M: Observations on the innervation of the macula sacculi in man. Anat Rec 1934; 59:403–418. 18 Fernandez C, Goldberg JM, Baird RA: The vestibular nerve of the chinchilla. III. Peripheral innervation patterns in the utricular macula. J Neurophysiol 1990; 63: 767– 780. 19 Curthoys IS, Betts GA, Burgess AM, MacDougall HG, Cartwright AD, Halmagyi GM: The planes of the utricular and saccular maculae of the guinea pig. Ann NY Acad Sci 1999;871:27–34. 20 Halmagyi GM, Curthoys IS, Colebatch JG, Aw ST: Vestibular responses to sound. Ann NY Acad Sci 2005;1039:1–14. 21 Minor LB, Solomon D, Zinreich JS, Zee DS: Sound and/or pressure-induced vertigo due to bone dehiscence of the superior semicircular canal. Arch Otolaryngol Head Neck Surg 1998;124:249–258.

Neuroembryol Aging 2004–05;3:207–214

22 Bickford RG, Jacobson JL, Cody DTR: Nature of averaged evoked potentials to sound and other stimuli in man. Ann NY Acad Sci 1964;112:204–218. 23 Colebatch JG, Halmagyi GM: Vestibular evoked potentials in human neck muscles before and after unilateral vestibular deafferentation. Neurology 1992;42:1635–1636. 24 Colebatch JG, Halmagyi GM, Skuse NF: Myogenic potentials generated by a clickevoked vestibulocollic reflex. J Neurol Neurosurg Psychiatry 1994;57:190–197. 25 Uchino Y, Sasaki M, Sato H, Bai R, Kawamoto E: Otolith and canal integration on single vestibular neurons in cats. Exp Brain Res 2005;164:271–258.

Curthoys

Neuroembryol Aging 2004–05;3:215–229 DOI: 10.1159/000096799

Published online: November 3, 2006

The Inferior Colliculus: A Center for Convergence of Ascending and Descending Auditory Information Manuel S. Malmierca Laboratory for the Neurobiology of Hearing, Department of Cellular Biology and Pathology, Faculty of Medicine, University of Salamanca, and Institute of Neuroscience of Castilla y Léon, Campus Miguel de Unamuno, Salamanca, Spain

Key Words Auditory midbrain
Hearing
Inferior colliculus
Laminar organization
Tonotopic organization

Abstract A major goal in auditory research is to understand completely how we hear, the physiology of the human auditory system and to identify the causes and treatments for hearing impairment. By understanding all the elements of the ‘auditory scaffold’ we will begin to achieve these important goals. The inferior colliculus (IC) occupies a strategic position in the central auditory system and may be considered a central hub or an interface between the lower auditory pathway, the auditory cortex and motor systems. The IC is the site for termination of the ascending fibers of the lateral lemniscus and also receives a heavy innervation from the auditory cortex. Furthermore, the IC receives crossed projections from its counterpart and possesses a dense network of local connections. Thus, the IC is the main site of auditory integration at the midbrain level. Anatomical and physiological experiments demonstrate that the IC is involved in a great diversity of functional roles in the auditory system, and that most of the interesting auditory features might already be extracted from incoming sounds by this midbrain nucleus. Copyright © 2005 S. Karger AG, Basel

© 2005 S. Karger AG, Basel 1661–3406/05/0034–0215$22.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

Introduction

A basic role of the auditory system in all mammals is to identify sounds, selectively activate neural systems that focus attention to sounds, and generate suitable motor responses. In the first auditory relay center in the brain, the cochlear nuclear complex, information carried by the fibers of the cochlear nerve is conveyed to a number of different neuronal populations that, in turn, give rise to a number of parallel ascending pathways that project to a variety of brainstem targets. Projections from these targets as well as direct projections from the cochlear nuclei ultimately converge on the auditory midbrain, the inferior colliculus (IC) [1–6]. In contrast to the role of the superior colliculus within the visual system, the IC is the principal source of input to the auditory thalamus [7]. The IC probably also represents a major output to premotor pathways that initiate or regulate soundevoked motor behavior [2]. Whereas most sensory systems have only two relay stations between the periphery and cerebral cortex, there is a minimum of three relays in the auditory system with several stages of convergence and divergence and at least seven levels of crossings from one side to the other [7]. Thus, the auditory system is unique among sensory systems with its highly complex network of pathways in the

Manuel S. Malmierca, Laboratory for the Neurobiology of Hearing Department of Cellular Biology and Pathology Institute of Neuroscience of Castilla y Léon, Faculty of Medicine University of Salamanca, Campus Miguel de Unamuno, ES–37007 Salamanca (Spain) Tel. +34 923 29 45 00 ext. 1861, Fax +34 923 29 45 49, E-Mail [email protected]

lower brainstem and a significant amount of processing accomplished in the IC, prior to the level of the thalamus. An important question is why an obligatory relay exists in the auditory midbrain. One intriguing explanation is that divergence at the early stages of auditory processing results in several distinct acoustic maps in the brainstem. The midbrain relay is required to enable their fusion into a single, integrated map at the level of the IC prior to relay to higher centers and for descending modulation from the neocortex. The IC is not only the main site of termination of the ascending fibers of the lateral lemniscus but also receives a heavy innervation from the auditory cortex (AC) [8]. Furthermore, the IC receives crossed projections from the contralateral IC [9] and possesses a dense network of intrinsic connections [9–12]. Thus, the IC occupies a strategic position in the central auditory system and may be considered an interface between the lower auditory pathway and the AC and motor systems [2]. In this review, our current understanding of the structure and function of the mammalian IC will be discussed.

A General View of the Anatomy of the IC

The IC is visible on the dorsal surface of the midbrain immediately caudal to the superior colliculus. In the cat, it is nearly spherical while it is ellipsoid in the rat. The IC was originally subdivided using classical neuroanatomical methods. Ramón y Cajal [13, 14] identified three main subdivisions in Golgi-impregnated material in a variety of mammals: the nucleus, the internuclear cortex and the lateral cortex. This simple parcellation (fig. 1) has been preserved in studies for the past 25 years with some minor modifications [9, 10, 15–18]. Thus, the IC is made of a central nucleus (CNIC), a laterally and rostrally placed external cortex (ECIC) and a dorsal cortex (DCIC) that covers the CNIC dorsally and caudally. The lateral and rostral parts of the ECIC contain several distinct cell types and, for this reason, Malmierca [10] defined them as two separate cortices, the lateral cortex of the IC (LCIC, cf. [3, 19]) and the rostral cortex of the IC (RCIC, also referred to as the intercollicular zone, cf. [20]). Molecular mapping techniques based on calcium-binding proteins also distinguish the CNIC from the cortices. Thus, parvalbumin has a higher concentration in the CNIC while calbindin and calretinin show a higher concentration in the DCIC [reviewed in ref. 21]. The metabolic marker cytochrome oxidase also delineates the CNIC [reviewed in ref. 18, 21]. 216

Neuroembryol Aging 2004–05;3:215–229

In the rat, the ECIC appears to occupy a relatively larger proportion of the IC than in the cat [16, 17, 19]. In the mustache bat, the ‘dorsoposterior division’ of the CNIC has an expanded representation for the high tonal frequencies used for echolocation [22]. The CNIC is defined by the presence of laminae (fig. 2), distinguishable in Golgi material as a parallel orientation of afferent lemniscal fibers and neurons with flattened dendritic arbors, usually referred to as ‘fibrodendritic laminae’ [23]. The characteristic laminar organization of the CNIC has been observed in all species studied so far [21]. The laminar organization of the CNIC constitutes the structural basis for its tonotopic map [24–26].

Neuron Types of the IC Subdivisions

The Central Nucleus The CNIC possesses two main types of neurons that were first defined in the cat as disk-shaped and stellate neurons, which have flattened dendritic arbors and dendritic arbors that often transverse the laminae, respectively [12, 27]. In the rat, corresponding types have been referred to as flat (F) and less flat (LF) neurons using computer-assisted 3D reconstructions of Golgi-impregnated material [17] (fig. 1, 2). The F neurons clearly conform to the definition of disk-shaped neurons described in the cat [27], but the correspondence between the LF and stellate is less clear. F and LF neurons (fig. 2) differ in dendritic arbor thickness, dendritic branching pattern, location and orientation with regard to the laminae. The thickness of the dendritic arbor of the F neurons is 50
m while that of LF neurons is 100
m, with the latter being less dense and branched than the former. The dendritic arbors of most F and LF neurons are elongated and located in parallel with the ventrolaterally to dorsomedially oriented long axis of the laminae. The F neurons are strictly parallel and form laminae mostly one cell thick (
40–70
m) [16]. The LF neurons lie roughly parallel to the F neurons [17] and populate interlaminar compartments that separate the laminae defined by the F neurons. The orientation of the F neurons (i.e. laminae) is almost horizontal in the dorsolateral part of the nucleus, but a gradual shift takes place so that they become more vertical in the medial part [17]. Similarly, a gradient in cell size and packing density of cell bodies as seen in Nissl-stained sections also prevails with the smallest cell bodies and highest packing density in the dorsolateral (low frequency) area (fig. 3G from Faye-Lund and Osen [16]). Malmierca

Fig. 1. a Computer-assisted 3D reconstruction of 35 neurons from the low- and high-frequency regions of the CNIC maintaining their mutual relationship. Camera lucida drawings of two F (b) and LF neurons (c). Redrawn from Malmierca et al. [65].

Auditory Processing in the Midbrain

Neuroembryol Aging 2004–05;3:215–229

217

Fig. 2. Diagram of the laminar and interlaminar compartments of the CNIC seen en face (a) and on edge (b). Redrawn from Malmierca et al. [65].

Thus far, I have described the neuronal types in the IC based primarily on their dendritic morphology, as F and LF. However, this dual classification does not correlate entirely with some of the functional properties observed [21]. IC neurons also vary in their neurochemistry, projections and their electrophysiological properties. The neurochemistry of the F and LF neurons has been studied in the rat [28] and cat [29]. In the cat, about 20% of the cells are -aminobutyric acid (GABA)ergic while in rat the proportion may be slightly larger (up to 25%) accord218

Neuroembryol Aging 2004–05;3:215–229

ing to Merchán et al. [28]. It seems that GABAergic cells may have either F or LF morphology [28]. In a recent study using whole-cell patch-clamp techniques in CNIC neurons in brain slices of rats, Sivaramakrishnan and Oliver [30] have characterized the potassium currents present and correlated them with the firing patterns observed by Peruzzi et al. [31]. Their study demonstrated the presence of six distinct physiological cell types: sustained regular, onset, pause build, rebound regular, rebound adapting and rebound transient. Each of these six Malmierca

types possesses a firing pattern caused by unique potassium currents and a set of other parameters (fig. 3). Because of differences in ionic currents, some neurons in the CNIC are likely to integrate their temporal and intensity inputs while others may act as simple relays. There is apparently no simple correlation between the anatomy and physiology of the F and LF cells [31]. Thus, while the F and LF morphology is clearly related to the maintenance of tonotopic organization, there may be several types of F and LF cells with complex functional roles. The Lateral Cortex The definition of the LCIC varies among species and the homology in rat and cat is just beginning to be clear [19]. Three layers are defined in the lateral part [10, 16]. Layer 1 is a continuation of the fibrodendritic capsule of the DCIC. Layer 2 is composed of small and mediumsized neurons, partly aggregated in dense clusters in myelin-rich neuropil. The aggregates are also rich in parvalbumin, cytochrome oxidase, nicotinamide adenine dinucleotide phosphate-diaphorase, and acetylcholinesterase, but they are immunonegative for glycine, calbindin, serotonin and choline acetyltransferase [32, 33]. Ascending auditory input to layer 2 is sparse, but the dorsal column nuclei and spinal trigeminal nucleus provide it with primary ascending sensory input. Therefore, the external cortex could participate in spatial orientation and somatic motor control through its intrinsic and extrinsic projections [33]. Layer 3 constitutes the largest part of the ECIC and appears to continue into the non-stratified rostral part (rostral cortex), which is topographically related to the fascicles of the commissural fibers. In addition to small and medium-sized cells, layer 3 contains large multipolar cells, especially ventromedially and rostrally. The border of the ECIC with the CNIC is indicated by a distinct shift in dendritic orientation, particularly conspicuous dorsolaterally as seen in caudal transverse sections [10, 16] where three morphologically distinct neuron types (bitufted, pyramidal-like and chandelier neurons) have been described [10]. Similar neurons have been described in the mouse. Willard and Ryugo [34] described large stellate cells with elongated dendritic arbors whose main axis is aligned perpendicular to the pial surface. Furthermore, computer-assisted 3D reconstructions of neurons in this region demonstrated that their dendritic arbors are different from those of the F neurons in the CNIC in several respects, including their thickness and orientation [10]. Because of this, some authors have hypothesized that the LCIC is compatible with a corticallike architecture (i.e. a laminar architecture with distinct Auditory Processing in the Midbrain

Fig. 3. The six firing patterns found in the IC after depolarizing and hyperpolarizing current pulses. a–f The top two traces are

the voltage response to each current pulse (bottom traces). Redrawn from Sivaramakrishnan and Oliver [30] .

input-output strata) in conformity with the original description made by Ramón y Cajal [13, 14]. This notion is supported by McCown and Breese [35] in neonatal rats, who proposed that the modulation of sensorimotor function may be modulated by the IC cortices prior to the maturation of the cerebral cortex. The Dorsal Cortex The DCIC covers the dorsomedial and caudal aspects of the CNIC. In the cat, the dorsomedial part consists of four layers and the thinner caudal part is unlayered [15]. The rat possesses only three layers [16]. The superficialmost layer (layer 1) is a thin fibrocellular capsule that continues with that over the LCIC. It contains scattered, small, flattened neurons. The deeper, slightly thicker layNeuroembryol Aging 2004–05;3:215–229

219

er 2 consists of small and medium-sized, mostly multipolar neurons. These two layers together constitute about one third the maximum thickness of the DCIC. Layer 3 contains small and medium-sized cells. There are large multipolar neurons at the border with the CNIC that differ in several respects from the F and LF neurons of the CNIC as demonstrated with computer-assisted 3D reconstructions of Golgi-impregnated neurons [10]. Layer 3 also contains elongated neurons located at the border between the CNIC and the DCIC whose dendritic arbors parallel the orientation of the laminae [10]; these may represent modified disk-shaped cells and have been referred to as transitional neurons [10]. The Rostal Cortex The neurons in the RCIC also differ from those in the CNIC and LCIC as they are very large multipolar cells [10, 16, 17]. In addition, small and medium-sized multipolar neurons are present in the rostral cortex.

Connections of the IC Subdivisions

Ascending afferent inputs to the IC arise from lower auditory centers and tend to terminate more densely in the ventral portions of the IC, whereas the afferent input from the AC and commissural input from the contralateral IC terminate more densely in the dorsal portions [2]. Therefore, the ventral portion of the CNIC appears to be functionally connected with lower auditory centers, while neurons in the DCIC may be more influenced by descending pathways from the AC. However, there is an area of overlap between regions receiving the ascending and descending inputs at the border of the CNIC and DCIC [36, 37], so neurons located in this region may receive a combination of ascending, descending, intrinsic and commissural inputs. Altogether, the neuroanatomical studies suggest the presence of complementary gradients of innervation by the ascending and descending pathways to the IC. These gradients may serve to produce functional gradients within a given lamina and such overlapping gradients of ascending and descending inputs may differ from species to species [2]. In the following, first a review of the afferent ascending connections to the different IC subdivisions is given, subsequently focusing on its descending connections. Finally, the functional organization of the efferent connections (both ascending and descending) as well as that of the local and commissural connections within and between the two colliculi is shown in detail. 220

Neuroembryol Aging 2004–05;3:215–229

Afferent (Ascending) Projections (fig. 4) The CNIC receives ascending input from more than 10 brainstem auditory centers [reviewed in ref. 2, 18], each of which is unique in structure and function. The main pathways arise from the cochlear nuclear complex, superior olivary complex (SOC) and nuclei of the lateral lemniscus, as demonstrated by retrograde axonal transport of tracers injected into the IC. The projections originate in the ventral cochlear nucleus (VCN) and dorsal cochlear nucleus (DCN) contralaterally, ventral nucleus of the lateral lemniscus and medial superior olive (MSO) ipsilaterally and the dorsal nucleus of the lateral lemniscus (DNLL) and lateral superior olive (LSO) bilaterally [1–5, 38–41]. In addition to these main projections, some species also have a projection from the ipsilateral cochlear nucleus to the low frequency part of the CNIC and a bilateral projection from the highest frequency parts of the MSO [38, 39]. Experiments using small injections of retrograde tracers injected into the IC demonstrate that the afferents are topographically (tonotopically) organized. Complementary experiments using anterograde tracers such as wheat germ agglutinin-conjugated horseradish peroxidase or 3H-leucine injected into the lower centers have shown that besides being tonotopically organized, many of the ascending systems show a non-uniform distribution in the CNIC, exhibiting a ‘banded’ pattern, with dense axonal bands about 200
m thick separated by bands of less dense labeling [21]. Bands formed by the projections from the ipsilateral and contralateral LSOs are intercalated, rather than overlapping. The banded pattern of bilateral projections to the CNIC from nuclei like the LSO and the DNLL may be more distinct on one side than on the other side, and the terminal fields of the various ascending projections may also vary in extent along the main axis of the IC. The terminal fibers from the SOC are confined to the ventral part of the laminae in the CNIC, whereas those from the VCN, DCN and the DNLL extend more dorsally in the laminae, extending into the deep region of the DCIC. Comparisons of the distribution of afferent axons from the DCN and the LSO to the contralateral IC in the same animal [39] show that layered axons from the DCN and LSO are superimposed only in the ventral part of the contralateral central nucleus. In the dorsal part of the central nucleus, the layer of axons from the DCN does not terminate with afferents from the LSO. Similar experiments in the rat, combining the injection of two different tracers in the VCN and DCN [40], suggest that some parts of these projections remain segregated within the CNIC laminae. Furthermore, two main components in the IC laminae have been reported: Malmierca

a major lamina that included the largest fibers and largest boutons, and a broader lamina, composed of thin fibers and only small boutons, which flank the major lamina. The major laminae originate from larger cell types in the cochlear nucleus, while the paralaminar zone may represent an input from small cells. Thus, two types of IC laminar structures may originate from the cochlear nucleus, and the small boutons in the paralaminar zone may provide an important modulatory input to the neurons of the IC. Very recently, Cant and Benson [18] demonstrated that the CNIC of the gerbil is made of two parts based on the inputs they receive from the brainstem. Lateral and rostral zones of the CNIC receive input from both the cochlear nuclei and SOC, whereas the medial and caudal zones of the CNIC receive inputs from the cochlear nucleus but not from the SOC. These and previous data have led to the hypothesis that specific functional zones may be created within the laminae of the CNIC [21]. The DCIC receives ascending input from the sagulum [41]. The LCIC and RCIC also receive fibers from many non-auditory structures, including the cuneate and trigeminal nuclei, the lateral nucleus of the substantia nigra, the parabrachial region, the midbrain central gray, the periventricular nucleus and the globus pallidus [42]. Afferent (Descending) Projections (fig. 4) Neurons in the collicular cortices may be more influenced by the AC than by ascending connections because of their bias toward descending projections [6, 20, 37, 42]. The descending input to the rat DCIC may originate largely from the primary AC bilaterally, with a small component to layer 1 from the area ventrocaudal to the primary cortex. Like the ascending projections to the CNIC, the neocortical terminals terminate in a topographic, banded pattern in parallel with the isofrequency contours of the CNIC [37]. The external cortex receives descending input from the cerebral cortex originating immediately rostral to the primary AC [20, 36, 43]. This projection is ipsilateral and terminates in layer 3. Secondary regions of the AC also project to the IC [20]. Cortical area Te2 projects primarily to the superficial layers of the DCIC and LCIC, while Te3 primarily innervates the RCIC. These projections originate in the pyramidal cells of layer V [44]. In addition, the AC has been shown to project not only to the cortical regions of the IC but also to the CNIC in the rat. Ultrastructural studies have demonstrated that the corticocollicular fibers terminate on thin dendritic shafts and spines, forming small boutons with round synaptic vesicles and asymmetric glutamatergic synapses [36, 45]. However, electrical stimAuditory Processing in the Midbrain

AC

AC

MGB

MGB

Inferior colliculus Inferior colliculus

Inferior colliculus Inferior colliculus

DNLL

DNLL

VNLL

VNLL

LSO MSO

MSO LSO

VNTB DCN

SPN

VNTB DCN SPN

VCN

VCN

Fig. 4. Schematic wiring diagram of the afferent connections to

the left IC and efferent connections of the right IC. Thicker lines show heavier projections than thinner lines. Solid lines indicate excitatory projections and discontinuous lines inhibitory connections. VNLL = Ventral nucleus of the lateral lemniscus; VNTB = ventral nucleus of the trapezoid body; SPN = superior paraolivary nucleus.

ulation of the cat AC elicits not only excitatory effects but also inhibitory and complex interactions in IC neurons [46]. Thus, the AC may modulate the processing of sounds in the IC both directly and also through the activation of local inhibitory connections within the IC. A direct descending pathway from the medial geniculate to the ECIC was recently demonstrated in the rat [47], a finding also reported in other species. In addition, Marsh et al. [48] recently established a direct, widespread projection from the basal amygdala to the IC in the mustache bat. They suggest the presence of a rapid thalamoamygdalo-collicular feedback circuit that may impose emotional content onto processing of sensory stimuli at a relatively low level of an ascending sensory pathway. Efferent (Ascending) Projections (fig. 4) Thus far, the general pattern of the inputs to the IC has been detailed. In turn, the IC projects to the medial geNeuroembryol Aging 2004–05;3:215–229

221

niculate body (MGB) [49–51] and to lower auditory centers such as the SOC and the cochlear nuclear complex [52–54]. Furthermore, the IC also has projections to nonauditory nuclei such as the pontine nuclei, a route by which auditory information can reach the cerebellum for coordination of motor responses to sound [55]. The CNIC projects to the laminated ventral division of the MGB in a strictly tonotopic manner. This projection is largely to the ipsilateral side but there is also a small crossed component. The CNIC also has a weak projection to the medial and dorsal divisions. The ECIC projects mainly to the dorsal and medial divisions of the MGB. The DCIC projects to the dorsal division of the MGB. The projections from the three subdivisions of the IC overlap, especially in the medial division [1–6]. The CNIC projections originate from both the F and LF neurons [50, 56]. Although the majority of neurons that project from CNIC to the MGB are glutamatergic, recent studies have shown that a significant proportion of the projection is GABAergic [56, 57]. The GABAergic projection originates from CNIC, DCIC and ECIC neurons, although the proportion from the cortical regions is lower. This projection has been confirmed in in vitro studies in which short-latency, monosynaptic inhibitory postsynaptic potentials from thalamocortical inputs have been demonstrated in the MGB [revised in ref. 4, 5]. Efferent (Descending) Projections (fig. 4) IC neurons also contribute to the descending auditory pathways, targeting the SOC (colliculo-olivary projections) and the cochlear nuclei (colliculo-cochlear projection). The rat IC also projects to the non-auditory pontine and mesencephalic reticular nuclei [53]. The colliculo-olivary projections form a band of terminals in the ventral nucleus of the trapezoid body [53, 54]. This projection is topographic and originates from the CNIC and ECIC. The terminals in the ventral nucleus of the trapezoid body overlap the site of origin of the medial olivocochlear system [54], although it remains to be demonstrated by electron microscopy whether these fibers from the IC make synaptic contact on the medial olivocochlear neurons. Physiological studies have shown that electrical stimulation of the IC produces an increase in the latency and a reduction in the amplitude of the auditory whole-nerve response and also reduces the temporal threshold shift that appears after the exposure to a loud noise [58]. These effects are similar to those elicited by electrical stimulation of the medial olivocochlear system. More recent studies have shown that selective electrical stimulation within the CNIC produces frequency222

Neuroembryol Aging 2004–05;3:215–229

specific reductions in neural activity in the cochlea [59] that are spatially restricted and bilateral. These effects are greater in the contralateral ear [60]. The colliculo-cochlear projection originates in the CNIC and ECIC (fig. 4) and targets the DCN and granule cell domain of the VCN, but its functional role is currently unknown [53, 61]. Intrinsic and Commissural Connections (fig. 4) In addition to the connections discussed above, the IC possesses well-developed fiber systems made up of intrinsic and commissural inputs. Fibers that interconnect the three subdivisions of the IC on one side are referred to as local or intrinsic while fibers that interconnect the two sides are referred to as commissural. [9, 11, 42, 62]. Both types of fibers may represent collaterals of axons with projections to the thalamus or lower brainstem or, alternatively, they may represent the sole projection of a neuron that is truly an interneuron restricted to the IC [11]. The terminal territories of the intrinsic fibers form ‘sheets’ that are parallel to the isofrequency contours of the CNIC [11, compare fig. 22 and 25]. The sheets extend into the DCIC, and, via a sharp bend, into the ECIC [9, 62]. Retrograde transport of horseradish peroxidase has also shown that the CNIC receives input from the DCIC bilaterally and from the ECIC ipsilaterally. The DCIC and ECIC on the same side are also mutually interconnected. Recent studies are beginning to uncover the functional role for both local and commissural connections. Miller et al. [11] have demonstrated that within a given isofrequency contour, intrinsic connections ascend from the ventrolateral portion to more dorsomedial points along the contour, forming a cascaded system of intrinsic feedforward connections that seem ideally suited to provide the delay lines necessary to produce several forms of selectivity for temporal patterns in IC neurons (see below). Injections of Phaseolus vulgaris leukoagglutinin or biotinylated dextran amine in one IC show that labeled fibers extend over the midline forming a mirror-like sheet on the contralateral side, thus indicating connections between the lamina devoted to the same frequency on the two sides. The majority of cells that project to the ipsilateral MGB also send collaterals to the contralateral IC. In guinea pigs, the commissural projection may be glutamatergic [63]. Consistent with the presence of both excitatory and inhibitory transmitters, physiological studies in vitro have shown that the commissural inputs may have either an excitatory or inhibitory influence on the contralateral IC [64]. In further in vivo studies, MalMalmierca

mierca et al. [65, 66] recorded sound-evoked responses of single neurons in one IC while injecting kynurenic acid into a corresponding region of the opposite IC. This procedure allowed the reversible blockage of excitation by commissural projections to the recorded IC. The changes observed in the neural responses when inputs from the opposite IC are blocked again confirmed that the commissural projection exerts both excitatory and inhibitory influences. The inhibition could be accounted for by monosynaptic or disynaptic connections, and the responses to both monaural and binaural stimulation are affected. Furthermore, the effects are proportionately greater at near-threshold sound levels. The results suggest that one function of the commissure of the IC may be to modulate the response gain of IC neurons to acoustic stimulation.

Neurochemistry

shown in in vitro studies demonstrating that some IC neurons exhibit long-term potentiation [70]. In addition to glutamatergic receptors, IC neurons also possess GABA A, GABAB and glycine receptors [67]. In vivo studies using microiontophoresis have demonstrated that both GABA and glycine inhibit IC neurons in several species [67, 71, 72]. Other neurotransmitters are also present in the IC and appear to be distributed differently by region. Serotonin terminals and receptors as well as noradrenergic fibers have been reported in the rat [73]. They originate from the locus ceruleus and dorsal raphe nucleus and seem to be more abundant in the cortical regions. Their functional role is unclear but recent studies in the bat have shown that serotonin modulates responses to species-specific vocalizations in the IC [74].

Basic Functional Properties of IC Neurons

As already mentioned, the neurochemistry of the F and LF neurons has been studied in detail. Both F and LF may be GABAergic [28, 29], but the majority of neurons (75% or more) are not inhibitory. It is tempting to suggest that the excitatory neurotransmitter of the IC neurons is glutamatergic. The IC lacks glycinergic neurons [28]. IC neurons possess both N-methyl-D-aspartic acid (NMDA) and 2-amino-3-(3-hydroxy-5-methylisoxazol4-yl)propionic acid (AMPA) receptors [67]. The different physiological roles of the NMDA and AMPA receptors have been studied using microiontophoretic application of NMDA and AMPA antagonists in vivo [67]. Both AMPA and NMDA receptors contribute to excitatory responses at all levels of acoustic stimulation that elicit action potentials, although there are more GluR2 and GluR3 receptors in the IC than GluR1 and GluR4 [68]. The NMDA and AMPA receptors have a selective influence on early and late components of tone-evoked responses [67]. Thus, the AMPA receptors are important at the onset of the neuronal responses in the IC while both AMPA and NMDA are involved in the maintenance of the response for the duration of the stimulus. The NMDA receptors are more abundant in the cortices than in the CNIC [67]. Thus, their distribution pattern matches that of the denser projection of the descending projections from the AC (v.s.). The cortico-collicular projection has been shown to cause long-lasting changes (12 h) in the neuronal responses of the IC [69], suggesting that the NMDA receptors play a significant role in neuronal plasticity. Further evidence of this role has been

Space limitation precludes a detailed account of the functional properties of the IC neurons; therefore the major spectral, binaural and temporal functional properties of the IC neurons are highlighted. Multi- and single-unit recording to pure tone stimulation and functional mapping studies with c-fos revealed that a fundamental physiological feature of the CNIC is its tonotopic organization (fig. 5). A narrow range of best frequencies is represented within each isofrequency lamina [25, 26]. In addition, the laminae have highly organized representations of acoustic signals based on both spectral and temporal properties [for review see ref. 25, 75]. Neurons in the IC exhibit several different types of peri-stimulus time histograms, including onset, on-sustained, pauser and sustained and regular responses [71]. The majority of neurons in the IC have V-shaped tuning curves similar to those seen in the auditory nerve, but frequency response areas in the IC may also include nonV-shaped maps, as described in many species [72, 76–79]. The non-V-shaped maps form a heterogeneous group that includes closed, narrow, low- and high-tilt, and multipeaked types (fig. 6). Binaural processing is initiated at the level of the SOC where interaural time and intensity differences are first encoded, but there appears to be further binaural processing in the IC. Kelly et al. [80] classified the responses of neurons to interaural intensity differences in the rat as either suppression, summation or mixed. Binaural suppression responses were more numerous at high frequen-

Auditory Processing in the Midbrain

Neuroembryol Aging 2004–05;3:215–229

223

Attenuation (dB)

cies and summation responses were more numerous at low frequencies. Studies based on the iontophoretic application of GABA and glycine antagonists have shown that neural inhibition contributes to the binaural response of neurons in the IC [81]. An important issue is the functional significance of binaural interaction at the level of the IC, given that the basic binaural comparisons occur at the level of the SOC. Kuwada et al. [81] have shown the emergence of interaural time difference (ITD) functions with an asymmetrical shape (sawtooth ITD functions) in the IC through the convergence of excitatory input from MSO and inhibitory input from DNLL. These sawtooth ITD functions are different from the classical peak or trough ITD functions seen in neurons in the SOC. Spitzer and Semple [82] have also suggested that the emergence of motion sensitivity in the IC might reflect the same pattern of convergence. Sound sources are not stationary in nature, and it may be that the binaural processing that takes place in the IC is related to the analysis of dynamic properties of sound source location. It may also be that the convergence of binaural pathways and spectral information at 224

Neuroembryol Aging 2004–05;3:215–229

Attenuation (dB)

Fig. 5. Cartoon of the tonotopic and laminar organization of the IC. The cartoon shows a 3D schematic view of two ICs, and the anatomical laminae in the the right CNIC in different colors. D = Dorsal; A = anterior; L = left; R = right. The figure was kindly provided by Dr. G. Langner.

Fig. 6. Frequency response areas (FRA) from the IC. a V shaped. b Narrow (non-V shaped). Data from Hernández et al. [78].

the IC could contribute to a map of auditory space, as it is the case in the barn owl’s midbrain [83]. In the first part of this review, an idea of the complexity of the ascending auditory pathways was given. As detailed very elegantly by Casseday et al. [2], the ascending input to the IC may produce an array of delay lines and temporally redesigned response patterns that alter the ways in which the neuronal representation of any given stimulus is distributed in time. Temporally modified inputs from multiple pathways, combined with excitation Malmierca

Fig. 7. Example of two (a, b) duration-tuned neurons from the IC and one non-duration-tuned neuron (c).

Redrawn from Pérez-González et al. [88].

and inhibition, may be the basis for creating selectivity to temporal features of sounds in the IC. The first account suggesting that neural delay lines might be used to analyze auditory temporal patterns was presented by Jeffress in [84] in his now classical model for encoding ITDs.

Licklider [85] also proposed a model for frequency discrimination based on coincidence detection and synaptic delays. Ascending inputs provide the IC with different temporal response patterns that comprise onset, sustained

Auditory Processing in the Midbrain

Neuroembryol Aging 2004–05;3:215–229

225

and offset responses with a wide range of latencies. They may be either excitatory or inhibitory (fig. 4). The results of the convergence onto single IC cells must result in a multifaceted temporal response of excitation and inhibition as demonstrated by intracellular recordings [81]. Temporal features that are generated de novo in the IC include tuning to rate and direction of frequency-modulated sweeps [86], tuning to sound duration [87, 88] (fig. 7) or delay of two sounds [89], and tuning to temporally sensitive facilitation in frequency combination neurons [90]. Thus far, the physiological features of the neurons of the central nucleus have been described. There are no detailed studies of the other subdivisions apart from the early studies from Aitkin et al. [91, 92], in which they demonstrated that ECIC neurons are multimodal, in agreement with their multisensory inputs. Only very recently, Pérez-Gonzalez et al. [93] have revealed that some neurons in the rat DCIC and ECIC show rapid and pronounced habituation to repeated presentations of identical stimuli but briefly recovered their responsiveness when some stimulus parameter was changed. Hence these neurons have been referred to as ‘novelty detectors’. An important function of the auditory system is to differentiate behaviorally uninteresting patterns of sound, which are often repetitive, from sounds that may require attention or action, and thus novelty neurons may be important for this vital function. Furthermore, the properties of novelty neurons in the IC are consistent with stimulation paradigms that produce mismatch negativity in humans and animals, so novelty neurons might be the neuronal correlate of mismatch negativity. Similar neurons have been shown also to be present in the AC [94]. Finally, a few additional comments about plasticity in the IC are worth mentioning. As described above, the IC is equipped with neuronal machinery (NMDA, AMPA and GABA receptors as well as neuromodulators such as acetylcholine, serotonin and adrenaline) that has been shown to be the basis for plasticity in other brain areas. Nevertheless, it is somewhat puzzling that there is little evidence for plasticity in the adult IC after cochlear damage or spiral ganglion lesions. Despite the wide variety in techniques employed to produce cochlear trauma in order to alter the response properties of the IC neurons [95, 96], most of the changes seen can be explained as a postlesional expression of preexisting inputs [96, 97]. This is in contrast to the well-established plastic changes demonstrated at thalamic and cortical levels in the auditory system [97–99]. 226

Neuroembryol Aging 2004–05;3:215–229

Concluding Remarks

The IC is not only the main site of termination for the ascending fibers of the lateral lemniscus, it is also heavily innervated by the AC. Furthermore, the IC receives crossed projections from its counterpart and possesses a dense network of local connections [1–5]. Thus, the IC occupies a strategic position in the central auditory system and may be considered as a central hub or an interface between the lower auditory pathway, the AC and motor systems [2]. Anatomical and physiological experiments demonstrate that the IC is involved in a great diversity of functional roles in the auditory system, and that most of the interesting auditory features might already be extracted from incoming sounds by this midbrain nucleus. It has even been suggested that the IC might be considered as the auditory analog of the primary visual cortex [100], leaving the AC to organize these features into auditory objects. The ultimate goal of auditory science is to understand completely how we hear. The aim of this review was to outline the interaction between structure and function in the auditory midbrain. Elucidation of the elements of the ‘auditory scaffold’ will help to fit the pieces of the puzzle together and solve the mechanisms by which the auditory system processes acoustic stimuli, i.e. how we hear.

Acknowledgments I thank Nell Cant and Fernan Jarmillo for their critical comments and correction of the English in a previous version. Gerald Langner kindly provided figure 5 first publised in Acta Otolaryngol Suppl. 1997;532:68–76. This study was supported by the Spanish JCyL-UE (SA040/04), JCYL SA007C05 and DGES (BFI-200309147-02-01).

Malmierca

References 1 Irvine DRF: Physiology of the auditory brainstem; in Popper AN, Fay RR (eds): Springer Handbook of Auditory Pathway Neurophysiology. New York, Springer, 1992, pp 153–231. 2 Casseday JH, Fremouw T, Covey E: The inferior colliculus: a hub for the central auditory system; in Oertel D, Popper AN, Fay RR (eds): Springer Handbook of Auditory Research. New York, Springer, 2002, pp 238– 318. 3 Waitzman D, Oliver DL: The midbrain; in Ramachandram S (ed): Encyclopedia of the Human Brain. New York, Academic Press, 2002, pp 43–68. 4 Malmierca MS: The structure and physiology of the rat auditory system: an overview. Int Rev Neurobiol 2003;56:147–211. 5 Malmierca MS, Merchán M: The auditory system; in Paxinos G (ed): The Rat Nervous System. San Diego, Academic Press, 2004, pp 997–1082. 6 Winer JA, Schreiner C: The central auditory system: a functional analysis; in Winer JA, Schreiner CE (eds): The Inferior Colliculus. New York, Springer, 2005, pp 1–68. 7 Malmierca MS, Merchán M, Henkel CK, Oliver DL: Direct projections from cochlear nucleus complex to auditory thalamus in the rat. J Neurosci 2002;22:10891–10897. 8 Winer JA: Decoding the auditory corticofugal systems. Hear Res 2006;212:1–8. 9 Malmierca MS, Rees A, LeBeau FEN, Bjaalie JG: Laminar organization of frequency-defined local axons within and between the inferior colliculi of the guinea pig. J Comp Neurol 1995;357:124–144. 10 Malmierca MS: Computer-assisted 3-D reconstructions of Golgi-impregnated cells in the rat inferior colliculus; thesis, University of Oslo & Salamanca, 1991. 11 Miller KE, Casseday JH, Covey E: Relation between intrinsic connections and isofrequency contours in the inferior colliculus of the big brown bat Eptesicus fuscus. Neuroscience 2005;136:895–905. 12 Oliver DL, Kuwada S, Yin TC, Haberly LB, Henkel CK: Dendritic and axonal morphology of HRP-injected neurons in the inferior colliculus of the cat. J Comp Neurol 1991; 303:75–100. 13 Ramón y Cajal S: Histología del Sistema Nervioso del Hombre y los Vertebrados. Madrid, Moya, 1904. 14 Ramón y Cajal S: Histologie du système nerveux de l’homme et des vertébrés (translated by Azoulay L). Paris, Maloine, 1909. 15 Morest DK, Oliver DL: The neuronal architecture of the inferior colliculus in the cat: defining the functional anatomy of the auditory midbrain. J Comp Neurol 1984; 222: 237–264.

Auditory Processing in the Midbrain

16 Faye-Lund H, Osen KK: Anatomy of the inferior colliculus in rat. Anat Embryol 1985; 175:35–52. 17 Malmierca MS, Blackstad TW, Osen KK, Karagülle T, Molowny RL: The central nucleus of the inferior colliculus in rat. A Golgi and computer reconstruction study of neuronal and laminar structure. J Comp Neurol 1993;333:1–27. 18 Cant NB, Benson C: Organization of the inferior colliculus of the gerbil (Meriones unguiculatus): differences in distribution of projections from the cochlear nuclei and the superior olivary complex. J Comp Neurol 2006;495:511–528. 19 Loftus WC, Malmierca MS, Bishop DC, Oliver DL: Ascending inputs to the lateral cortex of the inferior colliculus (IC) in rat and cat. ARO Abstr 2004;27:1115. 20 Herbert H, Aschoff A, Ostwald J: Topography of projections from the auditory cortex to the inferior colliculus in the rat. J Comp Neurol 1991;304:103–122. 21 Oliver DL: Neuronal organization in the inferior colliculus; in Winer JA, Schreiner CE (eds): The Inferior Colliculus. New York, Springer, 2005, pp 69–114. 22 Zook JM, Winer JA, Pollak GD, Bodenhamer RD: Topology of the central nucleus of the mustache bat’s inferior colliculus: correlation of single unit properties and neuronal architecture. J Comp Neurol 1985; 231: 530– 546. 23 Morest DK: The laminar structure of the inferior colliculus of the cat. Anat Rec 1964; 148:314. 24 Merzenich MM, Reid MD: Representation of the cochlea within the inferior colliculus of the cat. Brain Res 1974;77:397–415. 25 Schreiner CE, Langner G: Laminar fine structure of frequency organization in auditory midbrain. Nature 1997;388:383–386. 26 Malmierca MS, Izquierdo MA, Cristaudo S, Pérez-González D, Hernández O, Covey E, Oliver DL: Stepwise organization of tonotopy in the inferior colliculus of the rat: electrophysiological evidence. ARO Abstr 2006; 29:751. 27 Oliver DL, Morest DK: The central nucleus of the inferior colliculus in the cat. J Comp Neurol 1984;222:237–264. 28 Merchan M, Aguilar LA, Lopez-Poveda EA, Malmierca MS: The inferior colliculus of the rat: quantitative immunocytochemical study of GABA and glycine. Neuroscience 2005; 136:907–925. 29 Oliver DL, Winer JA, Beckius GE, Saint Marie RL: Morphology of GABAergic neurons in the inferior colliculus of the cat. J Comp Neurol 1994;340:27–42.

30 Sivaramakrishnan S, Oliver DL: Distinct K+ currents result in physiologically distinct cell types in the inferior colliculus of the rat. J Neurosci 2001;21:2861–2877. 31 Peruzzi D, Sivaramakrishnan S, Oliver DL: Identification of cell types in brain slices of the inferior colliculus. Neuroscience 2000; 101:403–416. 32 Ottersen OP, Størm-Mathisen J: Glutamateand GABA-containing neurons in the mouse and rat brain as demonstrated with a new immunocytochemical technique. J Comp Neurol 1984;229:374–392. 33 Chernock ML, Larue DT, Winer JA: A periodic network of neurochemical modules in the inferior colliculus. Hear Res 2004; 188: 12–20. 34 Willard FH, Ryugo D: Anatomy of the central auditory system; in Willot JF (ed): The Auditory Psychobiology of the Mouse. Springfield, Thomas, 1983, pp 201–304. 35 McCown TJ, Breese GR: The role of the inferior collicular cortex in the neonatal rat sensorimotor modulation. Dev Brain Res 1985; 59:1–5. 36 Saldaña E, Feliciano M, Mugnaini E: Distribution of descending projections from primary auditory neocortex to inferior colliculus mimics the topography of intracollicular projections. J Comp Neurol 1996; 371: 15– 40. 37 Bajo VM, Nodal FR, Bizley JK, Moore DR, King AJ: The ferret auditory cortex: descending projections to the inferior colliculus. Cereb Cortex 2006 Mar 31 [Epub ahead of print]. 38 Oliver DL: Projections to the inferior colliculus from the anteroventral cochlear nucleus in the cat: possible substrates for binaural interaction. J Comp Neurol 1987;264:24–46. 39 Oliver DL, Beckius GE, Bishop DC, Kuwada S: Simultaneous anterograde labeling of axonal layers from lateral superior olive and dorsal cochlear nucleus in the inferior colliculus of cat. J Comp Neurol 1997;382:215–229. 40 Malmierca MS, Saint Marie RL, Merchan MA, Oliver DL: Laminar inputs from dorsal cochlear nucleus and ventral cochlear nucleus to the central nucleus of the inferior colliculus: two patterns of convergence. Neuroscience 2005;136:883–894. 41 Shneiderman A, Oliver DL, Henkel CK: Connections of the dorsal nucleus of the lateral lemniscus: an inhibitory parallel pathway in the ascending auditory system? J Comp Neurol 1988;276:188–208. 42 Coleman JR, Clerici WJ: Sources of projections to subdivisions of the inferior colliculus in the rat. J Comp Neurol 1987;262:215– 226. 43 Faye-Lund H: The neocortical projection to the inferior colliculus in the albino rat. Anat Embryol 1985;173:53–70.

Neuroembryol Aging 2004–05;3:215–229

227

44 Games KD, Winer JA: Layer V in rat auditory cortex: projections to the inferior colliculus and contralateral cortex. Hear Res 1988;34:1–25. 45 Feliciano M, Potashner SJ: Evidence for a glutamatergic pathway from the guinea pig auditory cortex to the inferior colliculus. J Neurochem 1995;65:1348–1357. 46 Mitani A, Shimokouchi M, Nomura S: Effects of stimulation of the primary auditory cortex upon colliculogeniculate neurons in the inferior colliculus of the cat. Neurosci Lett 1983;42:185–189. 47 Senatorov VV, Hu B: Extracortical descending projections to the rat inferior colliculus. Neuroscience 2002;115:243–250. 48 Marsh RA, Fuzessery ZM, Grose CD, Wenstrup JJ: Projection to the inferior colliculus from the basal nucleus of the amygdala. J Neurosci 2002;22:10449–10460. 49 Oliver DL: Neurons types in the central nucleus of the inferior colliculus that project to the medial geniculate body. Neuroscience 1984;11:409–424. 50 Oliver DL, Ostapoff EM, Beckius GE: Direct innervation of identified tectothalamic neurons in the inferior colliculus by axons from the cochlear nucleus. Neuroscience 1999;93: 643–658. 51 Wenstrup JJ: The tectothalamic system; in: Winer JA, Schreiner CE (eds): The Inferior Colliculus. New York, Springer, 2005, pp 200–230. 52 Warr WB: Organization of olivocochlear efferent systems in mammals; in Webster DB, Popper AN, Fay RR (eds): The Mammalian Auditory Pathway Neuroanatomy. Berlin, Springer, 1992, pp 410–448. 53 Caicedo A, Herbert H: Topography of descending projections from the inferior colliculus to auditory brainstem nuclei in the rat. J Comp Neurol 1993;328:377–392. 54 Vetter DE, Saldaña E, Mugnaini E: Input from the inferior colliculus to medial olivocochlear neurons in the rat: a double label study with PHA-L and cholera toxin. Hear Res 1993;70:173–186. 55 Schuller G, Covey E, Casseday JH: Auditory pontine grey: connections and response properties in the horseshoe bat. Eur J Neurosci 1991;3:648–662. 56 Peruzzi D, Bartlett E, Smith PH, Oliver DL: A monosynaptic GABAergic input from the inferior colliculus to the medial geniculate body in rat. J Neurosci 1997;17:3766–3777. 57 Winer JA, Saint-Marie RL, Laure DT, Oliver DL: GABAergic feedforward projectiones from the inferior colliculus to the medial geniculate body. Proc Natl Acad Sci USA 1996; 93:8005–8010. 58 Rajan R: Centrifugal pathways protect hearing sensitivity at the cochlea in noisy environments that exacerbate the damage induced by loud sound. J Neurosci 2000; 20: 6684–6693.

228

59 Ota Y, Oliver DL, Dolan DF: Frequency-specific effects on cochlear responses during activation of the inferior colliculus in the guinea pig. J Neurophysiol 2004; 91:2185–2193. 60 Zhang W, Dolan DF: Inferior colliculus stimulation causes similar efferent effects on ipsilateral and contralateral cochlear potentials in the guinea pig. Brain Res 2006;1081: 138–149. 61 Malmierca MS, Le Beau FE, Rees A: The topographical organization of descending projections from the central nucleus of the inferior colliculus in guinea pig. Hear Res 1996; 93:167–180. 62 Saldaña E, Merchán MA: Intrinsic and commissural connections of the rat inferior colliculus. J Comp Neurol 1992;319:417–437. 63 Saint-Marie RL: Glutamatergic connections of the auditory midbrain selective uptake and axonal transport of D-[3H]-aspartate. J Comp Neurol 1996;373:255–270. 64 Moore DR, Kotak VC, Sanes DH: Commissural and lemniscal synaptic input to the gerbil inferior colliculus. J Neurophysiol 1998; 80:2229–2236. 65 Malmierca MS, Hernández O, Falconi A, Lopez-Poveda EA, Merchán MA, Rees A: The commissure of the inferior colliculus shapes frequency response areas in rat: an in vivo study using reversible blockade with microinjection of kynurenic acid. Exp Brain Res 2003;153:522–529. 66 Malmierca MS, Hernandez O, Rees A: Intercollicular commissural projections modulate neuronal responses in the inferior colliculus. Eur J Neurosci 2005;21:2701–2710. 67 Kelly JB, Caspary DM: Pharmacology of the inferior colliculus; in Winer JA, Schreiner CE (eds): The Inferior Colliculus. New York, Springer, 2005, pp 248–310. 68 Caicedo A, Eybalin M: Glutamate receptor phenotypes in the auditory brainstem and mid-brain of the developing rat. Eur J Neurosci 1999;11:51–74. 69 Suga N, Gao E, Zhang Y, Ma X, Olsen JF: The corticofugal system for hearing: recent progress. Proc Natl Acad Sci USA 2000;97:11807– 11814. 70 Zhang Y, Wu SH: Long-term potentiation in the inferior colliculus studied in rat brain slice. Hear Res 2000;147:92–103. 71 LeBeau FEN, Rees A, Malmierca MS: Contribution of GABA- and glycine-mediated inhibition to the monaural temporal response properties of neurons in the inferior colliculus. J Neurophysiol 1996;75:902–919. 72 LeBeau FEN, Malmierca MS, Rees A: Iontophoresis in vivo demonstrates a key role for GABA A and glycinergic inhibition in shaping frequency response areas in the inferior colliculus of guinea pig. J Neurosci 2001; 21: 7303–7312.

Neuroembryol Aging 2004–05;3:215–229

73 Klepper A, Herbert H: Distribution and origin of noradrenergic and serotonergic fibers in the cochlear nucleus and inferior colliculus of the rat. Brain Res 1991;557:190–201. 74 Hurley LM, Pollak GD: Serotonin shifts first-spike latencies of inferior colliculus neurons. J Neurosci 2005;25:7876–7886. 75 Schreiner CE, Langner G: Coding of temporal patterns in the central auditory nervous system; in Edelman GM, Gall WE, Cowan WM (eds): Auditory Function. New York, Wiley, 1988, pp 337–340. 76 Ramachandran R, Davis KA, May BJ: Singleunit responses in the inferior colliculus of decerebrate cats. I. Classification based on frequency response maps. J Neurophysiol 1999;82:152–163. 77 Egorova M, Ehret G, Vartanian I, Esser KH: Frequency response areas of neurons in the mouse inferior colliculus. I. Threshold and tuning characteristics. Exp Brain Res 2001; 140:145–161. 78 Hernandez O, Espinosa N, Perez-Gonzalez D, Malmierca MS: The inferior colliculus of the rat: a quantitative analysis of monaural frequency response areas. Neuroscience 2005;132:203–217. 79 Alkhatib A, Biebel UW, Smolders JW: Inhibitory and excitatory response areas of neurons in the central nucleus of the inferior colliculus in unanesthetized chinchillas. Exp Brain Res 2006;174:124–143. 80 Kelly JB, Glenn SL, Beaver CJ: Sound frequency and binaural response properties of single neurons in rat inferior colliculus. Hear Res 1991;56:273–280. 81 Kuwada S, Batra R, Yin TC, Oliver DL, Haberly LB, Stanford TR: Intracellular recordings in response to monaural and binaural stimulation of neurons in the inferior colliculus of the cat. J Neurosci 1997; 17: 7565–7581. 82 Spitzer MW, Semple MN: Interaural phase coding in auditory midbrain: influence of dynamic stimulus features. Science 1991; 254:721–724. 83 Konishi M: Listening with two ears. Sci Am 1993;268:66–73. 84 Jeffress LA: A place theory of sound localization. J Comp Physiol Psychol 1948; 41: 35– 39. 85 Licklider JCR: A duplex theory of pitch perception. Experientia 1951;7:128–134. 86 Rees A, Malmierca MS: Processing of dynamic spectral properties of sounds. Int Rev Neurobiol 2005;70:299–330. 87 Casseday JH, Ehrlich D, Covey E: Neural tuning for sound duration: role of inhibitory mechanisms in the inferior colliculus. Science 1994;264:847–850. 88 Perez-Gonzalez D, Malmierca MS, Moore JM, Hernandez O, Covey E: Duration selective neurons in the inferior colliculus of the rat: topographic distribution and relation of duration sensitivity to other response properties. J Neurophysiol 2006; 95:823–836.

Malmierca

89 Portfors CV, Wenstrup JJ: Delay-tuned neurons in the inferior colliculus of the mustached bat: implications for analyses of target distance. J Neurophysiol 1999; 82: 1326–1338. 90 Nataraj K, Wenstrup JJ: Roles of inhibition in creating complex auditory responses in the inferior colliculus: facilitated combination-sensitive neurons. J Neurophysiol 2005; 93:3294–3312. 91 Aitkin LM, Dickhaus H, Schult W, Zimmermann M: External nucleus of inferior colliculus: auditory and spinal somatosensory afferents and their interactions. J Neurophysiol 1978;41:837–847. 92 Aitkin LM, Kenyon CE, Philpott P: The representation of the auditory and somatosensory systems in the external nucleus of the cat inferior colliculus. J Comp Neurol 1981; 196:25–40.

Auditory Processing in the Midbrain

93 Perez-Gonzalez D, Malmierca MS, Covey E. Novelty detector neurons in the mammalian auditory midbrain. Eur J Neurosci 2005; 22: 2879–2885. 94 Ulanovsky N, Las L, Nelken I: Processing of low-probability sounds by cortical neurons. Nat Neurosci 2003;4:391–398. 95 Snyder RL, Sinex DG: Immediate changes in tuning of inferior colliculus neurons following acute lesions of cat spiral ganglion. J Neurophysiol 2002; 87:434–452. 96 Irvine DR, Rajan R, Smith S: Effects of restricted cochlear lesions in adult cats on the frequency organization of the inferior colliculus. J Comp Neurol 2003;467:354–374.

97 Kamke MR, Brown M, Irvine DR: Plasticity in the tonotopic organization of the medial geniculate body in adult cats following restricted unilateral cochlear lesions. J Comp Neurol 2003;459:355–367. 98 Rajan R, Irvine DR, Wise LZ, Heil P: Effect of unilateral partial cochlear lesions in adult cats on the representation of lesioned and unlesioned cochleas in primary auditory cortex. J Comp Neurol 1993; 338: 17– 49. 99 Kamke MR, Brown M, Irvine DR: Basal forebrain cholinergic input is not essential for lesion-induced plasticity in mature auditory cortex. Neuron 2005;48:675–686. 100 Nelken I: Processing of complex stimuli and natural scenes in the auditory cortex. Curr Opin Neurobiol 2004; 14:474–480.

Neuroembryol Aging 2004–05;3:215–229

229

Neuroembryol Aging 2004–05;3:230–238 DOI: 10.1159/000096800

Published online: November 3, 2006

From Receptive Field Dynamics to the Rate of Transmitted Information: Some Facets of the Thalamocortical Auditory System Chloé Huetz Jean-Marc Edeline Laboratoire de Neurobiologie de l’Apprentissage, de la Mémoire et de la Communication, Unité Mixte de Recherche, Centre National de la Recherche Scientifique, et Université Paris-Sud, Orsay, France

Key Words Information theory
Learning-induced plasticity
Sleep
Spike timing precision
Thalamocortical auditory system
Vocalization

Abstract In this article, we first evaluate the literature describing reorganizations of auditory cortex topography after behavioral training. We then review the studies showing that receptive fields of auditory thalamocortical neurons express large dynamics in unanesthetized animals. During the time course of different behavioral training protocols, the frequency tuning curves of thalamocortical neurons can be selectively modified to code for the learned importance of acoustic stimuli. In other circumstances, when the vigilance state shifts from waking to sleep, the functional properties of thalamic and cortical neurons exhibit drastic modifications. Finally, we point out new lines of research. First, investigations describing the responses of neurons to communication signals (e.g. species-specific vocalizations) are important because they reveal how the thalamocortical auditory system processes biologically relevant sounds. Second, we suggest that the spike timing precision can largely increase the amount of information transmitted in the thalamocortical auditory system. This urges for more systematic studies in which the temporal organization of spike trains will be considered at presentation of natural stimuli. Copyright © 2005 S. Karger AG, Basel

© 2005 S. Karger AG, Basel 1661–3406/05/0034–0230$22.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

Introduction

In the auditory modality, as well as in others, the foundations of sensory physiology have been established in anesthetized preparations that have been judged more stable than the awake ones. It is out of the scope of this article to review the myriad of studies that have described the functional properties of auditory neurons under general anesthesia. This step is necessary and essential to unravel the fundamental principles that allow auditory neurons to extract information from the external word and to code this information for the subsequent stages of processing. From the initial description of response selectivity to pure tones [1] to the most recent descriptions of the functional properties of neurons [reviewed in ref. 2– 4], impressive progress has been made in our understanding of the thalamocortical auditory system. However, we should keep in mind that anesthetics are drugs. Therefore, it is absolutely necessary to examine to what extent the functional properties discovered under general anesthesia indeed operate in the waking brain. In the auditory modality, initial observations suggested that the responses of thalamocortical neurons can exhibit a large dynamic range in unanesthetized animals. To cite a few examples, Hubel et al. [5] described ‘attention units’ in the auditory cortex; Evans and Whitfield [6] noted that the response patterns can be profoundly modified from time to time. The effects produced by shifts in vigilance states were also reported [7] in the auditory cortex of un-

Jean-Marc Edeline NAMC, UMR 8620, Bat 446 Université Paris-Sud FR–91405 Orsay (France) Tel. +33 1 6915 4972, Fax +33 1 6915 7726, E-Mail [email protected]

anesthetized macaque monkeys ‘the most dramatic shifts in excitability occur when the animal drifts through periods of wakefulness and sleep’. It should be noted that being able to express large dynamics in the functional properties of auditory neurons does not mean that they exhibit unstable properties. As we will see below, reliable responses can be collected in all states of vigilance, and evoked responses are as reliable before than after a learning situation. Nonetheless, because each neuron probably receives a large diversity of inputs in the awake animal, the response dynamics observed in awake animals is incomparable with the one observed in anesthetized ones.

Topographic maps are often viewed as fundamental to sensory processing [8, 9]. For this reason, the modifications in topographic maps reported after behavioral training had a large impact on the field of sensory physiology. However, the very first studies describing map reorganization after behavioral training did not attract much attention, probably because maps were quantified with metabolic activity (2DG), an indirect method to evaluate neuronal activity (note: functional magnetic resonance imaging, which has been used extensively in humans, is also quite an indirect method). In a set of remarkable studies, Gonzalez-Lima and Scheich [10–12] described the consequences of repeated pairing between a frequency-modulated tone (used as conditioned stimulus, CS) and an aversive stimulation of the reticular formation (used as unconditoned stimulus, US). Importantly increased labeling confined to the frequency band of the CS was noted after pairing, whereas five control groups developed less pronounced labeling. These studies also pointed out that increased labeling occurred at the thalamic and inferior colliculus levels, suggesting an involvement of subthalamic nuclei to cortical plasticity. Undoubtedly, the findings which have the most profoundly modified our way to consider adult sensory systems are those reported by Jenkins et al. [13] and Recanzone et al. [14, 15]. In the mid 80s, these authors initiated a set of outstanding experiments which demonstrated that adult sensory cortices exhibit large-scale plasticity in varied situations, ranging from denervation, deafferentation, intracortical microstimulation, nursing behavior to behavioral training. In the studies describing the consequences of behavioral training, adult monkeys were submitted to an extensive training period (2–3 months, with several hundreds of trials/day), and the cortical map of

primary sensory cortices was tested thereafter under general anesthesia. Both in the auditory and in the somatosensory cortex, enlarged map representations were reported after behavioral training. In the somatosensory system, two experiments using slightly different behavioral training revealed the same effect: when the training situation required that the animal actively and intensively used a small cutaneous portion of its hand, enlarged cortical representations were observed in favor of the trained skin surface [13, 14]. In the auditory cortex, an enlargement of the tonotopic map organization was found after a 3-month training in a perceptive discrimination task [15]. In this task, adult monkeys had to distinguish between pairs of stimuli made of two identical tone frequencies (the S1 stimulus) and pairs made of two slightly different frequencies (the S2 stimulus). Correct detection of the S2 stimulus allowed the animal to obtain a food reward, but this S2 stimulus was modified from session to session, being harder to detect each time the animal reached a detection level of 70%. The number of cortical sites responding to the frequency of the S1 stimulus was greater in the trained animals, leading to an increase in the cortical area corresponding to the trained frequency. Also, it was noted that the Q10dB (an index of the sharpness of tuning) was greater in trained than in control animals. The minimum latency of the evoked responses was increased in the trained animals, which might indicate that a recruitment of connections underlies the effects observed on the cortical map organization. More recently, two attempts to extend these findings led to opposite results. Rutkowski and Weinberger [16] used a particularly elegant strategy to obtain different levels of performance in rats trained to barpress for water delivery at presentation of a 6-kHz tone used as CS. During associative learning, the tone level of behavioral importance was controlled by the amount of supplemental water that the animals received in their home cage, so the asymptotic level of performance across subjects was correct in 60 to 190%. Maps of AI showed a large expanded representation for the frequency band centered on 6 kHz (the 4- to 8-kHz band), and there was a significant correlation between the final level of performance and selective expansion of the frequency band corresponding to the CS. In addition, the authors showed that other sounds associated with reward delivery also produced map reorganizations which, in that case, were not correlated with behavioral performance. In contrast, the study by Brown et al. [17] failed to detect map reorganization in AI after extensive perceptual learning, using a task that, in principle, was similar to that employed by Recanzone et al.

Dynamics of Auditory Receptive Fields

Neuroembryol Aging 2004–05;3:230–238

Plasticity of Cortical Maps after Behavioral Training

231

[15]. Cats trained on a 8-kHz frequency discrimination task showed improvements in performance that reflected changes in discriminative capacity. However, quantification of the AI tonotopic map indicated that the frequency organization in trained cats did not differ from that in controls. Nonetheless, quantitative measures of the response characteristics indicated that neurons with a characteristic frequency (CF) immediately above 8 kHz had slightly broader tuning in the trained cats and had significantly shorter response latency. The authors concluded that substantial changes in perceptual discriminative capacity can occur without change in primary cortical topography and with only modest changes in neuronal response characteristics. It is obviously difficult to reconcile these results, mostly because many factors can influence the changes in map organization: (i) the final differential threshold achieved during the last training sessions, (ii) the perceptual abilities acquired when procedural learning is at the asymptotic level, and (iii) the degree of motivation involved when the animal performed the behavioral task can potentially influence the map reorganizations. At the present time, data are too sparse to evaluate the impact of each of these factors. Comparisons with results on humans obtained with functional magnetic resonance imaging are inappropriate because maps in animals are collected under general anesthesia, whereas human data are obtained in awake subjects offering the possibility for multiple cascades of top-down processing.

Receptive Field Dynamics in Awake Animals

During Behavioral Training The determination of topographic maps is a very powerful technique to reveal large-scale plasticity in sensory cortices. However, with rare exceptions [18], maps derived in animals are usually determined under general anesthesia, partially because they require presentation of large sets of stimuli that need to be precisely controlled. Therefore, they provide somewhat static pictures of the thalamocortical system at a particular time point (for example after behavioral training). In contrast, single-unit recordings of behavior in animals allow to track the dynamics of sensory processing while the animal is submitted to a behavioral training [for a recent review see ref. 19]. The first evidence that functional properties of auditory cortical cells can be selectively affected by a brief learning experience came from the works of Diamond and Weinberger [20, 21] who demonstrated that the fre232

Neuroembryol Aging 2004–05;3:230–238

quency tuning curves of cortical neurons display selective modifications at the frequency of the tone associated with an aversive US. In these pioneering studies, tuning curves collected in secondary auditory areas (AII/VE) after 40–70 conditioning trials displayed either selective increases or selective decreases at the CS frequency. In contrast, several subsequent studies performed in primary auditory cortex (AI) revealed only selective increases at the CS frequency [22–24]. This ‘retuning’ of auditory cortex neurons is also prominent after a two-tone discrimination [25] and after instrumental training [26]. A striking property is the rapidity of occurrence of this plasticity: selective changes in the neuron tuning curves were detected after only 5–10 trials (10–20 min of training). Although cortical plasticity is usually emphasized, we have to keep in mind that the frequency tuning curves of thalamic cells can also display selective modifications after a brief learning experience [27–29]. The pitfall of all these initial studies was that the dynamics of sensory processing was tested ‘off line’: the animals were trained in a brief behavioral session and the tuning curves were tested before and after training while the animals were not actively using the tones for their current behavior. This problem was circumvented in the studies performed by two research groups. In their experimental design, Ohl and Scheich [30, 31] engaged gerbils in a multiple-frequency discrimination task involving a single CS+ (followed by an aversive US) and 11 CS– (not followed by the US). In that study, selective tuning curve alterations were found but they involved decreases, or no changes, at the CS+ frequency and increases at frequencies adjacent to the CS+, which, according to the authors, enhanced locally the gradient of responses versus sound frequency. This effect was interpreted as an enhancement of sensitivity for frequency changes in the spectral neighborhood of the training tone. Recently, an enhancement of contrast spectral sensitivity has also been reported in an appetitive task involving discrimination between a repeated standard tone (S–) and a deviant tone of higher frequency (S+). Over weeks, when cats decreased their frequency discrimination threshold, the response strength to the training frequencies gradually located in a local minimum compared to adjacent frequencies [32]. A quite different strategy was employed by Fritz et al. [33]: they trained ferrets in instrumental tasks during which complex acoustic stimuli allowed quantitative measurement of the spectrotemporal receptive fields (STRFs). More precisely, at each trial a set of stimuli (named TORC) were presented, first when the animal Huetz /Edeline

was passively listening, second when it was actively looking for the presence of a pure tone within the set of TORC, and third when it was again passively listening. Thus, the animal was trained to detect a pure tone in a background of TORCs, and the STRFs obtained during the detection task were compared with those obtained before and after the task. For a large majority of cells (79%, 31/39), an enhancement in the excitatory field (or a reduction of the inhibitory sideband) appeared during the detection task in favor of the frequency to be detected [34]. Importantly, only a third of the facilitated effects reverted back to their original shape after completion of the task; in the other cases the STRF changes tended to persist after the behavior (in some cases, they were maintained during more than 5 h). Recently, similar effects were also obtained using a discrimination task: during the behavioral session, TORCs were accompanied by a reference tone, and the animal had to respond to the TORC accompanied by the target tone (differed in frequency from the reference tone). In this situation, most of the STRFs exhibiting significant changes expressed decreased response to the reference tone and facilitated responses to the target tone [35]. The advantage of this strategy is clear: the functional properties of the neurons can be assessed while the animal is actively using the stimuli to perform the behavioral task. As the target tones (or the reference tones and the target ones) are embedded in TORC stimuli, this technique provides snapshot moment-to-moment images of functional changes related to behavior. In both the detection task and the discrimination task, significant STRF changes can be detected very rapidly (in 2–3 min, the maximum temporal resolution of this technique). Although it is difficult to reconcile all these results, some hypotheses have recently been proposed. Reviewing the literature on learning-induced plasticity in the auditory cortex, Ohl and Scheich [36] put forward that the type of plasticity that emerged in the auditory cortex is task dependent. In other words, depending on the constraint of the task in which subjects are trained, different types of plasticity are expressed. However, this appealing possibility should be called only when it is clear that other factors, such as the selectivity of behavioral performance, can be ruled out. In fact, studies using similar tasks gave quite different results: in a discrimination task between a fixed target frequency and a deviant frequency, Fritz et al. [35] reported increased responses at the deviant frequency whereas Witte and Kipke [32] reported decreased responses. Conversely, studies using quite different behavioral tasks led to similar results [30, 32].

In the Absence of Behavioral Training It is essential to take into account that multiple factors can influence the functional properties of thalamocortical neurons in unanesthetized animals. The most trivial one is the animal’s state of vigilance. Both at the cortical and at the thalamic level, neurons display profound receptive field modifications when the state of vigilance shifts from waking to slow-wave sleep and to paradoxical sleep. At the thalamic level, the majority (69/102) of the cells recorded during slow-wave sleep displayed smaller receptive fields at suprathreshold intensities, higher thresholds and as a consequence smaller frequency response areas. During paradoxical sleep, two populations of thalamic cells appeared: one which exhibited a more pronounced version of the changes detected in slow-wave sleep, and the other which tended to show receptive field properties similar to those in waking [37]. At the cortical level, many cells displayed alterations in their receptive field from waking to slow-wave and paradoxical sleep, but the heterogeneity of changes from one cell to the next finally led to a lack of global effect on the whole population [38]. It is important to note that when determined in a given stable state, the tuning curves are relatively stable. However, even during a well-defined state of vigilance, some changes can be detected when the neurons modify their mode of discharge. It is well documented that thalamic neurons can display two distinct modes of discharge: a ‘tonic’ mode where neurons emit single action potentials and a ‘burst’ mode where neurons emit clusters of actions potentials at high (1200 Hz) frequency [for review see ref. 39]. This dual mode of discharge has recently been described in auditory thalamus [40, 41], and some of its consequences have been reported; in a given behavioral state (e.g. waking), when a neuron switches from a tonic to a burst mode, its response latency and the variability in its response latency are decreased [42]. These changes can have important functional implications because it was shown that the timing of neuronal discharge carries more information than the spike count [43].

Dynamics of Auditory Receptive Fields

Neuroembryol Aging 2004–05;3:230–238

Responses to Natural Vocalizations

Obviously, neurons of the thalamocortical auditory system are not specialized to process pure tones or complex artificial stimuli such as dynamic ripples or TORCs. Pure tones are useful because they provide the easiest way to test the most accessible property which can be determined from the cochlea to the cortex: the frequency selectivity. The introduction of dynamic ripples by several 233

research groups [44–46] allowed considerable progress because the transfer function of neurons can be determined both in the temporal and in the spectral domain. However, it is not unreasonable to envision that the exquisite properties expressed by auditory neurons were progressively shaped by evolution to process natural sounds such as conspecific vocalizations. In many species, particularly in the primate, species-specific communication sounds are important for social interactions, reproductive success and survival. Initial studies performed several decades ago have evaluated how thalamocortical neurons process communication sounds and have discarded the concepts of call detectors because most of the neurons respond to multiple calls [47–49]. However, over the last years, an increasing number of studies evaluated the neuronal representation of species-specific vocalizations. For example, it was found that in the primary auditory cortex of marmosets, neurons responded much stronger to natural vocalizations than to synthetic variations that had the same spectral but different temporal characteristics [50]. Several recent studies aimed at determining whether the responses to vocalizations of auditory cortex neurons can be predicted from responses to artificial sounds. In the inferior colliculus, a good match was reported between the neuron CF and the response to the spectral content of vocalizations [51]. In other words, one can predict the responses of inferior colliculus neurons to vocalizations based on the frequency tuning of neurons. Results obtained in the auditory cortex are at variance with this simple view. On the basis of tuning curves, frequency response areas and responses obtained in two-tone paradigm, it was not always possible to predict the responses to natural sounds [52]. For example, some neurons respond to natural sounds with sparse responses to pure tones and no clearly definable frequency response areas [52, fig. 4]; other neurons do not respond to a natural sound even when the peak of its power spectrum coincides with their best frequency. In fact, recordings in the rat auditory cortex reveal that only 11% of the responses to natural stimuli can be inferred by the linear transformation from the sound spectrogram to responses of neurons (known as the STRF of neurons [53]). Therefore, whereas the STRF successfully predicts the responses to some of the natural stimuli, it fails completely to predict the responses to others. A similar picture seems to emerge at the thalamic level. In a recent study during which responses to guinea pig vocalizations were tested, no relationship was found between the CF of thalamic neurons and the responsiveness of neurons to vocalizations having different spectral con234

Neuroembryol Aging 2004–05;3:230–238

tent [54]. Preliminary results from our laboratory indicate that the STRFs of thalamic neurons can also fail in predicting the responses to natural stimuli (fig. 1). In contrast, in the inferior colliculus, the STRF model usually provides a good match between the predicted responses and the real responses to natural stimuli [55].

Temporal Coding and Transmitted Information

In the past, the functional properties of thalamocortical cells have been evaluated based solely on the firing rate, i.e. on the number of action potentials emitted at presentation of natural or artificial auditory stimuli. This is quite surprising given that it has long been pointed out that the temporal organization of neuronal discharges can code for the sound characteristics (such as sound frequency or intensity) more reliably than rate code does [56]. It is out of the scope of this review to describe the various aspects of temporal organizations which can participate in the neural code [reviewed in ref. 57, 58]. Regarding large cell populations, induced neuronal oscillations (particularly in the
range) are often viewed as a main actor in the construction of object representation both in the visual and in the auditory modality [reviewed in ref. 59]. Regarding small cell populations, temporal code can be expressed by the short time-scale coordination of neuronal discharges assessed by cross-correlograms. Auditory cortex neurons, for example, can display selective neuronal coordination for a stimulus location or a stimulus movement [60]. Furthermore, short time-scale interactions between neuronal discharges were found to be related with sound frequency independently of the firing rate: neurons can synchronize their firing rate without firing more action potentials [61]. At the single-cell level, temporal coding can involve the exact time of occurrence of the action potentials and/or the succession of the interspike intervals. In the auditory cortex, the critical parameter for the first-spike latency is the acceleration of peak pressure at tone onset rather than the rise time or the sound pressure level per se [62, 63]. Over a neuronal population, this property can be used to track properties of transients, and thus might contribute to the instantaneous coding of transients thought to underlie the categorical perception of speech and some nonlinguistic sounds [see discussion in ref. 63]. In addition, several studies provided evidence that auditory cortex neurons code stimulus location using the temporal structure of spike trains more than the rate of discharge of individual neurons [64, 65]. When an artificial neural netHuetz /Edeline

150 Response Prediction Spike rate (spikes/s)

100

50

0

–50 0

200

400

600

800

1,000

1,200

1,400

1,600

Time (ms)

Fig. 1. Predicted neuron poststimulus time histogram (PSTH) with original PSTH: lack of predictability of auditory thalamus responses based on STRF. Auditory thalamus neurons were tested with four guinea pig vocalizations presented 20 times in their natural and time-reversed version (total: 8 stimuli). Seven sets of stimulus responses were used to construct the STRF and the eighth stimulus was used to test the adequacy between the predic-

tion of STRF and the actual PSTH of the neuron. Construction of the STRF and prediction stage were done using the software STRFPAK [71]. The actual PSTH is represented by the black line and the prediction by the grey line. For this particular neuron (CF at 4 kHz, threshold 20 dB), there is no match between the actual responses and the responses predicted based on the STRF. Other neurons can provide good or very good predictions [72].

work is trained to recognize a stimulus location among 18 possible sources using neuron spike trains as inputs, it appears that, for most of the cells, azimuth coding by complete spike pattern is more accurate than by spike count, probably because of additional stimulus-related information contained in the timing of spikes. Based on single-unit recordings obtained in the auditory cortex of ferrets and cats, Nelken et al. [66] have demonstrated that for responses to natural and artificial sounds, spike count alone falls very short of computing the full information between stimuli and spikes trains. Essentially all the information can be grasped by adding another measure related to the timing of spikes in the calculation: the mean response time [66]. Recently, we started to investigate whether spike timing precision can reliably encode more information than firing rate at presentation of natural communication sounds. To address this question, we used the technique developed by Victor and Purpura [67, 68] to assess the precision of temporal coding from spike trains of visual cortex neurons. This metric-space method allows the estimation of information contained in spike trains and avoids the classical binning problem of the ‘direct’ method [69] when estimating this information [70]. Similar to

some other methods, the metric-space analysis computes distances between spike trains (as a measure of their dissimilarity), but it does not require binning of spike trains or embedding them in vector spaces, with dimensions dramatically increasing when one wants to keep a good time resolution (e.g. millisecond range) over several seconds of stimulus presentation. This technique also has the advantage to estimate the temporal resolution which maximizes the transmitted information. Analyzing the responses of auditory thalamus neurons to guinea pig vocalizations, we found that taking into account the temporal organization of neuronal discharges largely increases transmitted information, including cells that transmit little or no information based on spike count (fig. 2).

Dynamics of Auditory Receptive Fields

Neuroembryol Aging 2004–05;3:230–238

Conclusions

Over the last 2 decades, electrophysiological recordings performed in awake animals have revealed the dynamics of sensory processing. We have made considerable progress in our understanding of the thalamocortical auditory system: this system can no longer be considered as a passive analyzer of the external word, but should be 235

a

0

100

200

300

400

500

600

700

0

100

200

300

400

500

600

700

b

Time (ms)

Fig. 2. The rate of transmitted information

236

Transmitted information Chance level

1.0

Transmitted information

0.8

0.6

0.4

0.2

c

Neuroembryol Aging 2004–05;3:230–238

1

10 5

100 50

0 1,000 500

increases when spike timing is considered. a, b Raster plots from responses of an auditory thalamus neuron to 100 presentations of a guinea pig vocalization (a ‘purr’) in its natural version (a) and in its time-reversed version (b). Stimulus begins at time 0 and lasts for 700 ms. c Transmitted information (black line) and chance level (grey line) as a function of the spike timing precision computed with the metric-space analysis of Victor and Purpura [67, 68]. The curve indicates that there is almost no transmitted information based on the spike count (value of information at the origin, when spike timing precision tends to infinite). The maximum of transmitted information is reached for values of spike timing in the 10-ms range.

Spike timing precision (ms)

Huetz /Edeline

viewed as part of the network contributing to cognitive representation. In the future, the studies in awake animals should probably benefit from the development of new recording techniques such as calcium imaging or cortical intrinsic signaling. However, the electrophysiological techniques will remain essential because, when associated with new analytical tools, they allow descriptions of the neural code focused on the temporal organization of neural discharges. As the neural code probably evolved over evolution to provide precise and reliable processing of sounds necessary for survival, a straightforward approach to understand this code is certainly to use natural communication sounds. Dissecting the neural code used by awake animals to extract information

from natural stimuli is an enormous and very exciting challenge.

Acknowledgments We thank Elizabeth Hennevin for helpful comments on an earlier draft of this paper. The preliminary analyses presented here were made with the help of Israel Nelken at the Advanced Course in Computational Neuroscience (Arcachon, August 2005). C.H. was supported by a fellowship from the French Ministère de la Recherche et de l’Enseignement Supérieur. The research described in this review was partially supported by a grant from the Action Concertée Incitative ‘Neurosciences Intégratives et Computationnelles’.

References 1 Galambos R: Microelectrode studies on medial geniculate body of cat. III. Response to pure tones. J Neurophysiol 1952; 15: 381– 400. 2 Griffiths TD, Warren JD, Scott SK, Nelken I, King AJ: Cortical processing of complex sound: a way forward? Trends Neurosci 2004;27:181–185. 3 Shamma S: On the role of space and time in auditory processing. Trends Cogn Sci 2001; 5:340–348. 4 Winer JA, Miller LM, Lee CC, Schreiner CE: Auditory thalamocortical transformation: structure and function. Trends Neurosci 2005;28:255–263. 5 Hubel DH, Henson CO, Rupert A, Galambos R: ‘Attention’ units in the auditory cortex. Science 1959;129:1279–1280. 6 Evans EF, Whitfield IC: Classification of unit responses in the auditory cortex of the unanesthetized and unrestrained cat. J Physiol 1964;171:476–493. 7 Brugge JF, Merzenich MM: Responses of neurons in auditory cortex of macaque monkey to monaural and binaural stimulation. J Neurophysiol 1973; 36:1130–1158. 8 Kaas JH: Topographic maps are fundamental to sensory processing. Brain Res Bull 1997;44:107–112. 9 Shamma SA: Topographic organization is essential for pitch perception. Proc Natl Acad Sci USA 2004;101:1114–1115. 10 Gonzalez-Lima F, Scheich H: Neural substrates for tone-conditioned bradycardia demonstrated with 2-deoxyglucose. I. Activation of auditory nuclei. Behav Brain Res 1984;14:213–233. 11 Gonzalez-Lima F, Scheich H: Classical conditioning enhances auditory 2-deoxyglucose patterns in the inferior colliculus. Neurosci Lett 1984;51:79–85.

Dynamics of Auditory Receptive Fields

12 Gonzalez-Lima F, Scheich H: Neural substrates for tone-conditioned bradycardia demonstrated with 2-deoxyglucose. II. Auditory cortex plasticity. Behav Brain Res 1986;20:281–293. 13 Jenkins WM, Merzenich M, Ochs MT, Allard T, Guic-Robles E: Functional reorganization of primary somatosensory cortex in adult owl monkeys after behaviorally controlled tactile stimulation. J Neurophysiol 1990;63:82–104. 14 Recanzone GH, Merzenich MM, Jenkins WM, Grajski KA, Dinse HA: Topographic reorganization of the hand representation in cortical area 3b of owl monkeys trained in a frequency discrimination task. J Neurophysiol 1992;67:1031–1056. 15 Recanzone GH, Schreiner CE, Merzenich MM: Plasticity in the frequency representation of primary auditory cortex following discrimination training in adult owl monkeys. J Neurosci 1993;13:87–103. 16 Rutkowski RG, Weinberger NM: Encoding of learned importance of sound by magnitude of representational area in primary auditory cortex. Proc Natl Acad Sci USA 2005; 102:13664–13669. 17 Brown M, Irvine DR, Park VN: Perceptual learning on an auditory frequency discrimination task by cats: association with changes in primary auditory cortex. Cereb Cortex 2004;14:952–965. 18 Chen LM, Friedman RM, Roe AW: Optical imaging of SI topography in anesthetized and awake squirrel monkeys. J Neurosci 2005;25:7648–7659. 19 Weinberger NM: Specific long-term memory traces in primary auditory cortex. Nat Rev Neurosci 2004;5:279–290.

20 Diamond DM, Weinberger NM: Classical conditioning rapidly induces specific changes in frequency receptive fields of single neurons in secondary and ventral ectosylvian auditory cortical fields. Brain Res 1986; 372: 357–360. 21 Diamond DM, Weinberger NM: Role of context in the expression of learning-induced plasticity of single neurons in auditory cortex. Behav Neurosci 1989;103:471–494. 22 Bakin JS, Weinberger NM: Classical conditioning induces CS-specific receptive field plasticity in the auditory cortex of the guinea pig. Brain Res 1990;536:271–286. 23 Weinberger NM, Javid R, Lepan B: Longterm retention of learning-induced receptive-field plasticity in the auditory cortex. Proc Natl Acad Sci USA 1993; 90: 2394– 2398. 24 Edeline JM, Pham P, Weinberger NM: Rapid development of learning-induced receptive field plasticity in the auditory cortex. Behav Neurosci 1993;107:539–551. 25 Edeline JM, Weinberger NM: Receptive field plasticity in the auditory cortex during frequency discrimination training: selective retuning independent of task difficulty. Behav Neurosci 1993;107:82–103. 26 Bakin JS, South DA, Weinberger NM: Induction of receptive field plasticity in the auditory cortex of the guinea pig during instrumental avoidance conditioning. Behav Neurosci 1996;110:905–913. 27 Edeline JM, Weinberger NM: Subcortical adaptive filtering in the auditory system: associative receptive field plasticity in the dorsal medial geniculate body. Behav Neurosci 1991;105:154–175. 28 Edeline JM, Weinberger NM: Thalamic short-term plasticity in the auditory system: associative retuning of receptive fields in the ventral medial geniculate body. Behav Neurosci 1991;105:618–639.

Neuroembryol Aging 2004–05;3:230–238

237

29 Edeline JM, Weinberger NM: Associative retuning in the thalamic source of input to the amygdala and auditory cortex: receptive field plasticity in the medial division of the medial geniculate body. Behav Neurosci 1992;106:81–105. 30 Ohl FW, Scheich H: Differential frequency conditioning enhances spectral contrast sensitivity of units in auditory cortex (field Al) of the alert Mongolian gerbil. Eur J Neurosci 1996;8:1001–1017. 31 Ohl FW, Scheich H: Learning-induced dynamic receptive field changes in primary auditory cortex of the unanaesthetized Mongolian gerbil. J Comp Physiol [A] 1997; 181: 685–696. 32 Witte RS, Kipke DR: Enhanced contrast sensitivity in auditory cortex as cats learn to discriminate sound frequencies. Cogn Brain Res 2005;23:171–184. 33 Fritz J, Elhilali M, Shamma S: Active listening: task-dependent plasticity of spectrotemporal receptive fields in primary auditory cortex. Hear Res 2005;206:159–176. 34 Fritz J, Shamma S, Elhilali M, Klein D: Rapid task-related plasticity of spectrotemporal receptive fields in primary auditory cortex. Nat Neurosci 2003;6:1216–1223. 35 Fritz JB, Elhilali M, Shamma SA: Differential dynamic plasticity of A1 receptive fields during multiple spectral tasks. J Neurosci 2005; 25:7623–7635. 36 Ohl FW, Scheich H: Learning-induced plasticity in animal and human auditory cortex. Curr Opin Neurobiol 2005;15:470–477. 37 Edeline JM, Manunta Y, Hennevin E: Auditory thalamus neurons during sleep: changes in frequency selectivity, threshold and receptive field size. J Neurophysiol 2000; 84: 934–953. 38 Edeline JM, Dutrieux G, Manunta Y, Hennevin E: Diversity of receptive field changes in auditory cortex during natural sleep. Eur J Neurosci 2001;14:1865–1880. 39 Sherman SM: Tonic and burst firing: dual modes of thalamocortical relay. Trends Neurosci 2001;24:122–126. 40 He J, Hu B: Differential distribution of burst and single-spike responses in auditory thalamus. J Neurophysiol 2002; 88:2152–2156. 41 Massaux AE, Edeline JM: Bursts in the medial geniculate body: a comparison between anesthetized and unanesthetized states in guinea-pig. Exp Brain Res 2003; 153: 573– 578. 42 Massaux A, Dutrieux G, Cotillon-Williams N, Manunta Y, Edeline JM: Auditory thalamus bursts in anesthetized and non-anesthetized states: contribution to functional properties. J Neurophysiol 2004; 91: 2117– 2134.

238

43 Mickey BJ, Middlebrooks JC: Representation of auditory space by cortical neurons in awake cats. J Neurosci 2003;23:8649–8663. 44 Eggermont JJ: Wiener and Volterra analyses applied to the auditory system. Hear Res 1993;66:177–201. 45 Kowalski N, Depireux DA, Shamma SA: Analysis of dynamic spectra in ferret primary auditory cortex. I. Characteristics of single-unit responses to moving ripple spectra. J Neurophysiol 1996; 76:3503–3523. 46 Escabi MA, Schreiner CE: Nonlinear spectrotemporal sound analysis by neurons in the auditory midbrain. J Neurosci 2002; 22: 4114–4131. 47 Newman JD, Wollberg Z: Multiple coding of species-specific vocalizations in the auditory cortex of squirrel monkeys. Brain Res 1973;54:287–304. 48 Glass I, Wollberg Z: Auditory cortex responses to sequences of normal and reversed squirrel monkey vocalizations. Brain Behav Evol 1983;22:13–21. 49 Glass I, Wollberg Z: Responses of cells in the auditory cortex of awake squirrel monkeys to normal and reversed species-specific vocalizations. Hear Res 1983;9:27–33. 50 Wang X, Merzenich MM, Beitel R, Schreiner CE: Representation of a species-specific vocalization in the primary auditory cortex of the common marmoset: temporal and spectral characteristics. J Neurophysiol 1995; 74: 2685–2706. 51 Suta D, Kvasnak E, Popelar J, Syka J: Representation of species-specific vocalizations in the inferior colliculus of the guinea pig. J Neurophysiol 2003; 90:3794–3808. 52 Rotman Y, Bar-Yosef O, Nelken I: Relating cluster and population responses to natural sounds and tonal stimuli in cat primary auditory cortex. Hear Res 2001;152:110–127. 53 Machens CK, Wehr MS, Zador AM: Linearity of cortical receptive fields measured with natural sounds. J Neurosci 2004; 204: 1089– 1100. 54 Philibert B, Laudanski J, Edeline JM: Auditory thalamus responses to species-specific guinea-pig vocalizations: a comparison between rat and guinea-pig. Hear Res 2005; 209:97–103 55 Woolley SM, Gill PR, Theunissen FE: Stimulus-dependent auditory tuning results in synchronous population coding of vocalizations in the songbird midbrain. J Neurosci 2006;26:2499–2512. 56 Hind JE, Goldberg JM, Greenwood DD, Rose JE: Some discharge characteristics of single neurons in the inferior colliculus of the cat. II. Timing of discharges and observations on binaural stimulation. J Neurophysiol 1963; 26:321–341. 57 Eggermont JJ: Is there a neural code? Neurosci Biobehav Rev 1998;22:355–370.

Neuroembryol Aging 2004–05;3:230–238

58 Eggermont JJ: Between sound and perception: reviewing the search for a neural code. Hear Res 2001;157:1–42. 59 Tallon-Baudry C, Bertrand O: Oscillatory gamma activity in humans and its role in object representation. Trends Cogn Sci 1999; 3: 151–162. 60 Ahissar M, Ahissar E, Bergman H, Vaadia E: Encoding of sound-source location and movement: activity of single neurons and interactions between adjacent neurons in the monkey auditory cortex. J Neurophysiol 1992;67:203–215. 61 deCharms RC, Merzenich MM: Primary cortical representation of sounds by the coordination of action-potential timing. Nature 1996;381:610–613. 62 Heil P, Irvine DR: On determinants of firstspike latency in auditory cortex. Neuroreport 1996;7:3073–3076. 63 Heil P: Auditory cortical onset responses revisited. I. First-spike timing. J Neurophysiol 1997;77:2616–2641. 64 Middlebrooks JC, Clock AE, Xu L, Green DM: A panoramic code for sound location by cortical neurons. Science 1994; 264: 842– 844. 65 Middlebrooks JC, Xu L, Clock AE, Green DM: Codes for sound-source location in nontonotopic auditory cortex. J Neurophysiol 1998;80:863–868. 66 Nelken I, Chechik G, Mrsic-Flogel TD, King AJ, Schnupp JW: Encoding stimulus information by spike numbers and mean response time in primary auditory cortex. J Comput Neurosci 2005;19:199–221. 67 Victor JD, Purpura KP: Nature and precision of temporal coding in visual cortex: a metricspace analysis. J Neurophysiol 1996; 76: 1310–1326. 68 Victor JD, Purpura KP: Metric-space analysis of spike trains: theory algorithms and application. Network 1997;8:127–164. 69 Strong SP, Koberle R, Ruyter van Steveninck R, Bialek W: Entropy and information in neural spike trains. Phys Rev Lett 1998; 80: 197–200. 70 Victor JD: How the brain uses time to represent and process visual information. Brain Res 2000;886:33–46. 71 Theunissen FE, David SV, Singh NC, Hsu A, Vinje WE, Gallant JL: Estimating spatiotemporal receptive fields of auditory and visual neurons from their responses to natural stimuli. Network 2001;12:289–316. 72 Laudanski J, Huetz C, Edeline JM: Spectrotemporal receptive field of guinea-pig auditory cortex neurons: differences between dynamic noise and natural communication sounds. Nottingham, Auditory Cortex: The Listening Brain, Sept 17–21, 2006.

Huetz /Edeline

Neuroembryol Aging 2004–05;3:239–248 DOI: 10.1159/000096801

Published online: November 6, 2006

Thalamocortical and Corticothalamic Interaction in the Auditory System Zhuo Zhang a Ying-Shing Chan b Jufang He a a

Department of Rehabilitation Sciences, Hong Kong Polytechnic University, Kowloon, and Department of Physiology and Research Center of Heart, Brain, Hormone and Healthy Aging, LKS Faculty of Medicine, The University of Hong Kong, Hong Kong, SAR, China

b

Key Words Auditory cortex
Corticothalamic projections
Dynamic filter
Medial geniculate body
Thalamic reticular nucleus
Thalamocortical projections

Abstract The neuronal interconnections between thalamus and cortex modulate information transmission in the central auditory system. To facilitate the understanding of its modulation effects and mechanism, first, the recent progress on investigating the anatomical connections and the physiological properties of the medial geniculate body and auditory cortex is summarized. Second, the connectional pattern and functional organization in thalamocortical and corticothalamic network are described. Third, the strategic position and role of the thalamic reticular nucleus in the auditory system are demonstrated. Lastly, the functional implication of this integrated system and possible interactions between auditory thalamus and cortex are discussed. The thalamocortical and corticothalamic projections may work as a dynamic filter array in integrating sensory information and optimizing signal processing in the auditory system. Copyright © 2005 S. Karger AG, Basel

© 2005 S. Karger AG, Basel 1661–3406/05/0034–0239$22.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

Introduction

In all sensory systems except olfaction, the thalamus is the last relay along the ascending pathway before afferent information reaches the cortex. The auditory thalamus includes the medial geniculate body (MGB), the lateral part of the posterior nucleus group and the auditory sector of the thalamic reticular nucleus (TRN) [1, 2]. The MGB is the principal auditory thalamic nucleus and the obligatory synaptic station for auditory information directed towards the telencephalon. The auditory cortex (AC) in mammals consists of multiple fields and represents the central processing stage of auditory information [3, 4]. Compared with the ascending projection to the cortex, the auditory thalamus receives a much stronger reciprocal projection from the cortex [5–7]. The reciprocity of the thalamocortical (TC) and corticothalamic (CT) projections is not quantitatively rigid [6, 8]. One of the functions of the CT projections, as well as other corticofugal projections to the inferior colliculus (IC) and brainstem, is to provide a gating or gain control mechanism in the transmission of information from the periphery to the cortex [9–15]. Interconnections and functional interactions between the thalamus and the cortex are of great interest because they are not only the physical basis of mutual modula-

Zhuo Zhang Department of Rehabilitation Sciences Hong Kong Polytechnic University, Hung Hom Kowloon, Hong Kong, SAR (China) Tel. +852 2766 6723, Fax +852 2330 8656, E-Mail [email protected]

tions in sensory information transmission, but also the foundation upon which the state-dependent network activities of the forebrain are built. The present article reviews (1) the anatomy and physiology of the MGB and AC, (2) the TC and CT projections in the auditory system, (3) the strategic position and role of the TRN, and (4) the possible function of the interacting connections.

Anatomy and Physiology of the MGB

In all mammals and most vertebrates, the MGB relays information from the IC to the cortex. It consists of three divisions: the ventral division (MGv), the dorsal division (MGd) and the medial division (MGm). MGv, the largest division with predominantly a densely packed aggregate of medium-sized neurons, contains the fewest and but best-studied neuron types in the MGB. In the MGd, the neurons have slightly smaller somata and are more dispersed, while the MGm has a broad range of somatic sizes, including the largest neurons and lowest packing density in the auditory thalamus [16–19]. Physiologically, the ascending auditory pathways from the thalamus may be subdivided into two parallel systems known as primary (lemniscal) and secondary (nonlemniscal) projections [20]. The lemniscal core of the MGB is the tonotopically organized MGv. In the cat, the fibrodendritic orientation confers a laminar structure upon the MGv and the isofrequency contours parallel to this laminae arrangement with the low-frequency region located laterally, the middle-frequency region caudomedially, and the high-frequency region rostromedially [20–23]. In the guinea pig, the tonotopic map runs rostrocaudally with high frequencies located rostrally and low frequencies caudally [24, 25]. Neurons in the MGv have extremely uniform properties, which include high levels of spontaneous activity, sharp frequency tuning, and the shortest latency (8–15 ms) responses to acoustic stimuli. Neurons in non-lemniscal MGB, MGd and MGm show broader or less well-tuned frequency response properties, more variable firing patterns, longer latencies, non-tonotopic organization, and integrative features to multisensory afferent inputs [26–31].

Anatomy and Physiology of the AC

Anatomically, the primary auditory field (AI) is defined by its distinct cytoarchitecture and connections with the MGv. Recently, tonotopically comparable non240

Neuroembryol Aging 2004–05;3:239–248

primary subfields in the AC as well as AI are found to provide corticofugal excitatory projections to the same part of MGv in the rat [32]. The number of auditory fields increases from three or four in insectivores, to four to seven in rodents, and six to more than eight in carnivores and primates [19, 33–35]. Physiologically, the AI is defined by its tonotopic organization mapped in a variety of mammals. In both anesthetized and unanesthetized animal preparations, a large proportion of AI neurons appear tuned to narrow ranges of frequency while others are more broadly tuned or exhibit multipeaked tuning curves [19, 36, 37]. Neurons with similar characteristic frequencies occupy cortical bands that are oriented orthogonal to the tonotopic gradient. In many species, there is evidence that at least one other auditory field with a regular tonotopy is found besides AI. Adjacent tonotopic fields form mirror images of one another, and thus adjacent fields share a common range of frequencies at their border sites. However, these adjoining reversed tonotopic fields do not always represent the full audible frequency range as AI. In a number of mammals, one of these fields lying ventrally to the AI is named secondary auditory cortical field (AII). At the outer borders of the fields of the AII, multisensory association areas are located in which neurons receive visual or somatosensory in addition to auditory inputs [38–45]. The AC is involved in the analysis of natural stimuli [19]. Pitch-selective neurons are found in the primate AI, and the sustained firing can be evoked by their preferred acoustic stimuli in a particular neuronal population [46, 47].

TC and CT Projections

The auditory TC-CT projection loops are illustrated in figure 1 [48]. The intricate regional and laminar organization of the system is the foremost problem to be solved. The TC pathways arise from different populations of thalamic neurons terminating in different areas and layers in the cortex. The organization of the MGB projections to the AC is proved to be quite similar in different mammals, including the cat, monkey and bat [49, 50]. In the cat, the MGv neurons project primarily to AI and its mirror region, the anterior auditory field (AAF) [22]. Equivalent auditory fields in the guinea pig are the AI and the dorsocaudal auditory fields [51, 52]. The MGd projects to areas of the AC surrounding the primary cortex while the MGm projects to all auditory cortices including its association cortex [1, 5, 53–55]. The MGm also proZhang /Chan /He

AI/ AAF/PAF

AII

IV V VI

MGd

MGm

MGv Brainstem

Fig. 1. A schematic representation of CT-TC loops for the tono-

topically organized auditory cortical areas AI, AII, AAF and posterior auditory cortical fields (PAF) in the cat. O = CT neurons; X = TC neurons; I = axon terminals. The cortical layers IV, V, and VI are indicated (modified based on Rouiller and Welker [48] and Winer and Prieto [84]).

jects to the somatosensory cortex, prefrontal cortex, amygdala and basal ganglia [56–61]. The pathway that projects from the MGv to the middle layers of AI is also called the primary TC pathway, while the projections from the MGd, MGm, peripeduncular nucleus, and other non-primary thalamic nuclei are called non-primary TC [18, 62–67]. In many species, each MGB subdivision projects to more than one cortical field (rat [18], monkey [55], cat [68], and guinea pig [69]). Although the TC projections to AI and AAF originate mainly from tonotopically corresponding parts of the MGv, significant non-topographic thalamic projections to frequency-specific domains in the cat AC have been found [70]. The heterotopic projections may play a role in the cortical plasticity and signal presentation [71]. Neurons of the MGm provide parallel thalamic inputs to Interaction of Auditory Thalamus and Cortex

all ACs. Different from the MGv, the projections from the MGm are multisensory and capable of long-term potentiation [72]. A large targeting area of MGm projections may minimize its selectivity over frequency and space. TC projections also differ in laminar organization. In the cat, (1) axons from the MGv and parts of the MGd mainly project to layer III–IV and slightly to layer I; (2) axons from the MGd project mainly to layer I and less to layers III–IV, and (3) axons from the MGm have the largest axonal trunks, run laterally for a long distance in layer I, show the lowest density of labeling and a trimodal concentration in layers I, III–IV, and VI. Notably, layer V, which is regarded widely as independent of thalamic inputs, receives more than 10% of the total boutons [68, 73, 74]. In the monkey, (1) fibers from the MGv terminate mainly in layer IV and deep portion of layer III (IIIB); (2) fibers from the MGd terminate largely in layer IIIB, and (3) fibers from the MGm have relatively few terminals; some terminate preferentially in middle layers and others exclusively in layer I [55]. In the rat, projections from the MGv to AI terminate in layers III and IV of discrete cortical areas, while diffuse projections from the MGd and MGm terminate throughout layer I or VI of the temporal cortex but selectively in lower layer III and layer IV of specific parts of the AII and AIII [61]. In the rabbit, some TC axons that send input to layer I also terminate in layers II–V, implying concomitant activation across 1,500-
m-wide tangential zones and perhaps more than one mode of lemniscal TC activation [75]. The descending cortical projections feeding back to the thalamus are quantitatively important and present in all sensory systems. CT projections originate from heterogeneous pyramidal neurons in layers V and VI and are as divergent as the TC projection [7, 76–81]. CT projections exert both excitatory and inhibitory influence on thalamic neurons. The excitatory corticofugal influence is achieved by monosynaptic connections that are markedly robust in number and located on the distal dendrites of thalamic principal neurons [82], while the inhibitory corticofugal influence is by polysynaptic connections either with intrinsic -aminobutyric acid (GABA)ergic interneurons within the thalamic relay nuclei in the MGB or with GABAergic neurons located in the TRN. The corticogeniculate axons also influence their target neurons (both relay neurons and interneurons) in a topographic manner [35]. In the cat, about half of all layer VI pyramidal neurons contribute to the corticogeniculate pathway and each auNeuroembryol Aging 2004–05;3:239–248

241

Cortex

TRN

A Thalamus

In a variety of mammalian species, two functionally distinct CT projection systems have been recognized: (1) small axon terminals (!1
m) arising from layer VI, presynaptic to small peripheral dendritic profiles [82, 86, 87] and innervating only the dorsal thalamus and the TRN, and (2) large terminals (2–10
m) from layer V pyramidal neurons giving no branches to the TRN [85, 88– 90]. The former type plays a role in transforming the firing properties of thalamic neurons and the feedback control of the cerebral cortex on the thalamic nucleus from which it receives its main projection. The latter provides a primary drive for thalamic neurons and affects feedforward projections by which activity from a cortical area is distributed to other parts of the cortex via the thalamus and may participate in corticocortical communication [91, 92].

B

The TRN Fig. 2. Schematic view of two thalamic nuclei that are function-

ally related to the same modality. Some of their connections with the TRN and layer VI pyramidal neurons in the cerebral cortex are shown. The thick black line connections show how a reticulothalamic axon that branches to innervate more than one thalamic nucleus can provide a disynaptic inhibitory link between one of the nuclei and the other. The thinner black lines show possible comparable links that may be formed by branches of CT axons. The gray lines show possible links established by CT axons of layer VI neurons, which can send branches into a sector of the TRN and the two thalamic relay nuclei (from Guillery and Harting [100]).

ditory cortical area projects back reciprocally to the MGB: AI and AAF project to the MGv, AII to the MGd, and all fields to the MGm [5, 83]. Corticogeniculate projections also arise from layer V, where neurons of origin are mainly classical pyramidal neurons with long, well-filled apical dendrites [84]. One type of pyramidal neurons, with intrinsic bursting characteristics, provides the major inputs from layer V to subcortical targets (MGB and IC), and those to the secondary thalamic areas may involve the CT-TC pathway [85]. Injection of a bidirectional tracer into the cortex labeled a large number of MGv and MGd somas and CT terminals. In contrast, the same injection retrogradely labeled many neurons in the MGm but very few CT terminals [8], suggesting that the MGm is likely to receive relatively weaker projections from the cortex than other subdivisions. 242

Neuroembryol Aging 2004–05;3:239–248

The TRN, strategically situated along the axonal paths that link the thalamus and the cortex, is a thin sheet of GABAergic neurons wrapping around the rostral, lateral, and dorsal aspects of the thalamus. Both TC axons and CT axons of layer VI pyramidal neurons give off collaterals to the TRN, making excitatory synaptic contacts (glutamatergic) with TRN neurons [93–96]. The TRN neurons, in turn, provide inhibitory GABAergic innervation of the thalamic nuclei from which they receive afferents [93, 97, 98]. However, as the main part of the ventral thalamus, the TRN sends no axons to the cortex. Through the TRN, a disynaptic inhibitory link between two functionally related thalamic nuclei has also been proposed (fig. 2) [99, 100]. Such possibilities create further modulatory interaction mechanisms in CT-TC pathways. Neurons in different TRN sectors are devoted to different sensory modalities. The studies by Crabree et al. [101–103] revealed that there are cross-nucleus and crossmodality interconnections between the TRN and the dorsal thalamus. The auditory sector of the TRN is located in the caudoventral region of the nucleus [104, 105]. These neurons are topographically organized according to the MGB region and/or the nuclear subdivision they innervate [99]. Recordings from auditory TRN neurons have shown that the majority (160%) are responsive to white noise [106, 107]. Stimulation of the TRN auditory region suppresses spontaneous activity in MGB neurons as well as their response to auditory stimuli [105, 108]. TRN axon terminals act on MGd, MGv and MGm neuZhang /Chan /He

300 Electric stimulation

Fig. 3. Responses of a deep layer neuron of

Electrical MGm acoustic 100 Acoustic only 0 Acoustic stimulus 0

rons via GABA-A and GABA-B receptors [109]. In the guinea pig, activation of the AC results in a strong and long-lasting inhibition of non-lemniscal MGB neurons [25, 110, 111] in contrast to a strong facilitation and a small inhibitory effect on lemniscal MGB neurons [112]. The strong inhibition is proposed to be caused via the AC]TRN]MGB pathway and not via GABAergic neurons in the IC [110, 113]. Considering the multiple cortical and thalamic maps with interconnecting projections necessitated by their mirror reversals, as well as the significant amount of convergence and divergence in the collateral pathways, the TRN can serve as a crucial nexus in modulating both global and local information transmission. Until now we know rather little about this complex axonal latticework. Understanding the intrinsic circuits and functional connection patterns of the TRN should offer one of the keys to reveal the interaction between the thalamus and the cortex.

Functional Implications

In the auditory thalamus, the physiological and anatomical differences between the MGm and MGv have long been recognized, but the difference in their cortical functions is yet to be proven. The MGv and primary CT pathway are believed to carry fast, sensory specific inputs and peripheral tonotopic maps to the AI [114]. The MGm is found to be involved in neuronal plasticity during audiInteraction of Auditory Thalamus and Cortex

Acoustic only

200 Trial

area AI (600
m) to a combination of electrical stimulation of MGm and an acoustic stumulus. The first 100 and the last 80 trials of the raster display show the neuronal response to an acoustic stimulus, and the middle 80 trials show the response to the combination of electrical stimulation of MGm and an acoustic stimulus with 100ms interval between them. The neuron is totally inhibited for about 200 ms after the MGm stimulation. The latency of auditory response is lengthened for about 100 ms with MGm stimulation [Xu and He, unpubl. results].

400

800 Time (ms)

tory fear conditioning [115–118]. Learning in a fear condition is possibly associated with an elevated alertness of the AC and amygdala, where the MGm also sends projections. In the auditory TC slice, stimulation of the MGB with average stimulus intensity can antidromically activate cortical neurons in layers V/VI [119]. Since each axon of the MGm is widespread in the cortex, the direct modulatory effect of the MGm is likely to have a broad and general effect on the AC and/or beyond. In our preliminary extracellular recording results from over 15 cortical neurons in anesthetized guinea pigs, mainly inhibitory effects on the middle layer neurons to noise stimulus were recorded when electrical stimulation was applied to the MGm (fig. 3). Although the effect appears more complicated than a simple inhibition or excitation, it is obviously more modulatory, rather than a spike-evoked effect. The CT input affects many facets of physiology and signal selection in the central auditory system: first, sensory responses and receptive field properties, and second, firing mode and/or activity state [120–122]. The massive corticofugal system is able to extensively adjust and improve subcortical auditory signal processing in the frequency, amplitude, time, and spatial domains [12, 123–129]. The enhanced suppression supplied by CT feedback is a means to sharpen receptive fields or increase the filtering properties of thalamic neurons. A related view of CT feedback suggests that the increased filtering supplied by feedback may serve to improve the saliency of specific sensory stimuli, allowing these stimNeuroembryol Aging 2004–05;3:239–248

243

Fig. 4. Functional relations between CT, TRN, and TC neurons in the cat. a Three neurons (cortical, TRN, and TC) are intracellularly recorded and stained. Signs of excitation and inhibition are indicated by + and –. For the sake of simplicity, local-circuit inhibitory neurons in the cortex and thalamus are not illustrated. The response of TRN and TC neurons to cortical stimulation (arrowheads point to stimulus artifacts): the GABAergic TRN neuron responded to cortical stimulation with a high-frequency spike burst, followed by a sequence of spindle waves on a depolarizing envelope (membrane potential, –68 mV). The TC neuron responded to cortical stimulation (arrowhead) with biphasic inhibitory postsynaptic potentials, leading to a low-threshold spike (LTS) and a sequence of hyperpolarizing spindle waves (membrane potential, –70 mV). b Relationships between cortical (area 4), TRN and TC neurons during spontaneously occurring, cortically generated seizure with polyspike-wave complexes at 2 Hz. Note inhibitory postsynaptic potentials in a TC neuron (filled circles) in close time relationship with spike bursts fired by a TRN neuron, driven from the cortex (from Steriade [135]).

uli to ‘pop out’ of noisy or inhomogeneous surroundings [122, 129]. The synaptic network in the CT systems modulates intrinsic neuronal properties, often with a decisive impact. During various states of vigilance, brain oscilla244

Neuroembryol Aging 2004–05;3:239–248

tions are grouped together through reciprocal connections between the neocortex and thalamus. It is found that the neocortex governs the synchronization of network or intrinsically generated oscillations in the thalamus. These spontaneous oscillations may be involved in the consolidation of memory traces and neuronal plasticity [130, 131]. The CT pathway not only selects the input acoustic properties of neurons in the frequency, amplitude, time and spatial domains [132], but also modifies auditory responses via the TRN. Since the corticofugal positive feedback has a high gain, long-lasting discharges (perhaps responsible for tinnitus) would be produced if the TRN does not operate properly [133]. In addition, the corticofugal system probably mediates attentional modulation of auditory signal processing [134]. Synchronous cortical volleys (which occur naturally during slow-wave sleep when neurons exhibit highly coherent activity) or electrical stimulation of the cortex produce rhythmic spike bursts in TRN neurons but inhibitory postsynaptic potentials, occasionally followed by rebound excitation, in TC neurons (fig. 4) [135]. The TRN, the main source of inhibition in the TC-CT loop, probably plays the key role in the gain-control modulation, especially in different modes of auditory oscillatory activities. Dynamic sensory processing in the TC-CT loop includes both the precise representation of peripheral inputs and highly transformed intracortical responses [136]. Supposing a dynamic filter array to selectively gate or attend the signals in the auditory system, this filter array (1) has to be placed as early as possible along the auditory pathway according to the information processing theory and (2) must have its gain controlled via the judgment of the central system. To find out what signal is wanted and what should be filtered, the cortex must be involved. The effects of corticofugal modulation on the auditory system have been confirmed in conscious bats, anesthetized cats and guinea pigs [125, 14, 110]. In the visual system, the lateral geniculate nucleus has been reported to behave as a dynamic filter to transform the spatiotemporal properties of visual stimuli with different responses in distinct ways [137]. Recently, Sillito and Jones [138] and Sillito et al. [139] found that the perception of visual motion involves a dynamic interplay between the cortical motion area, the primary visual cortex and the thalamus. They proposed that thalamic mechanisms are selectively focused by visually driven feedback mechanisms to optimize the thalamic contribution to segmentation and global integration. In the auditory system, we proposed that the filter array should be located in the Zhang /Chan /He

Fig. 5. A hypothetic model of dynamic filter for auditory attention. In our hearing process, we can selectively attend to a particular speaker in the noise party room (i.e. the well-known ‘cocktail party phenomenon’). In the proposed model, the filter locates between the thalamus and the cortex (Cx). Spatial and spectral cues are used to identify the specific speaker, store the information in the cortex, and refresh this information all the time. As recognition and memory are required here, the AC and/or other cortical areas must be involved in this process. Cues are used to form the dynamic filter array in the auditory system. Besides auditory cues, one also uses visual cues to supplement the control switching of the filter, though more experimental evidence is needed to consolidate such a notion.

thalamus and controlled by the CT projections (fig. 5) [140]. Since this dynamic filtering process is likely to be a universal one for the auditory and other sensory systems under both the conscious and behaving preparations, the TC-TC system is the ideal candidate for the filter array.

Concluding Remarks

The thalamus is no longer considered as a simple, passive relay of sensory information to the cortex. Instead, recent findings corroborate its role as an elaborate circuit that is designed to perform multiple computations for processing of sensory information relayed to the cortex. The cortex, on the other hand, constitutes a unique entity that

engages in analyzing and adjusting the input information arising from changes in both the external environment and variations in behavioral states. The complexity of the TC-CT system is therefore poised to deal with the rich dimensions of signals in the auditory system. Despite recent efforts in delineating the anatomical circuitry and the physiological properties of these neuronal networks, continuing investigation is needed to increase our understanding of the reciprocal modulation and functional integration between the thalamus and the cortex.

Acknowledgments The study was partially supported by the Hong Kong Research Grants Council (CERG PolyU5407/03M).

References 1 Imig TJ, Morel A: Organization of the thalamocortical auditory system in the cat. Annu Rev Neurosci 1983;6:95–120. 2 Smith PH, Spirou GA: From the cochlea to the cortex and back; in Oertel D, Fay RR, Popper AN (eds): Integrative Functions in the Mammalian Auditory Pathway. New York, Springer, 2002, pp 6–71. 3 Ehret G: The auditory cortex. J Comp Physiol 1997;181:547–557.

Interaction of Auditory Thalamus and Cortex

4 Budinger E: Introduction: auditory cortical fields and their functions; in Konig R, Heil P, Budinger E, Scheich H (eds): Auditory Cortex: A Synthesis of Human and Animal Research. Mahwah, Erlbaum, 2005, pp 3–6. 5 Andersen RA, Knight PL, Merzenich MM: The thalamocortical and corticothalamic connections of AI, AII, and the anterior auditory field (AAF) in the cat: evidence for two largely segregated systems of connections. J Comp Neurol 1980;194:663–701.

6 Winer JA, Larue DT: Patterns of reciprocity in auditory thalamocortical and corticothalamic connections: study with horseradish peroxidase and autoradiographic methods in the rat medial geniculate body. J Comp Neurol 1987;257:282–315. 7 Winer JA, Diehl JJ, Larue DT: Projections of auditory cortex to the medial geniculate body of the cat. J Comp Neurol 2001;430:27– 55.

Neuroembryol Aging 2004–05;3:239–248

245

8 He J, Hashikawa T: Connections of the dorsal zone of cat auditory cortex. J Comp Neurol 1998;400:334–348. 9 Ryugo DK, Weinberger NM: Corticofugal modulation of the medial geniculate body. Exp Neurol 1976;51:377–391. 10 Crick F: Function of the thalamic reticular complex: the searchlight hypothesis. Proc Natl Acad Sci USA 1984;81:4586–4590. 11 Deschênes M, Hu B: Electrophysiology and pharmacology of the corticothalamic input to lateral thalamic nuclei: an intracellular study in the cat. Eur J Neurosci 1990;2:140– 152. 12 Villa AE, Rouiller EM, Simm GM, Zurita P, de Ribaupierre Y, de Ribaupierre F: Corticofugal modulation of the information processing in the auditory thalamus of the cat. Exp Brain Res 1991;86:506–517. 13 Suga N, Zhang Y, Yan J: Sharpening of frequency tuning by inhibition in the thalamic auditory nucleus of the mustached bat. J Neurophysiol 1997; 77:2098–2114. 14 He J: Modulatory effects of regional cortical activation on the onset responses of the cat medial geniculate neurons. J Neurophysiol 1997;77:896–908. 15 Zhou X, Jen PH: Brief and short-term corticofugal modulation of subcortical auditory responses in the big brown bat, Eptesicus fuscus. J Neurophysiol 2000; 84:3083–3087. 16 Morest DK: The laminar structure of the medial geniculate body of the cat. J Anat 1965; 99:143–160. 17 Winer JA: The medial geniculate body of the cat. Adv Anat Embryol Cell Biol 1985; 86: 1– 97. 18 Clerici WJ, Coleman JR: Anatomy of the rat medial geniculate body: I. Cytoarchitecture, myeloarchitecture, and neocortical connectivity. J Comp Neurol 1990;297:14–31. 19 Winer JA: The functional architecture of the medial geniculate body and the primary auditory cortex; in Webster DB, Popper AN, Fay RR (eds): The Mammaliam Auditory Pathway: Neuroanatomy. New York, Springer, 1992, pp 222–409. 20 Clarey JC, Barone P, Imig TJ: Physiology of thalamus and cortex; in Popper AN, Fay RR (eds): Mammalian Auditory Pathway: Neurophysiology. New York, Springer, 1992, pp 232–334. 21 Aitkin LM, Webster WR: Medial geniculate body of the cat: organization and responses to tonal stimuli of neurons in ventral division. J Neurophysiol 1972;35:365–380. 22 Imig TJ, Morel A: Topographic and cytoarchitectonic organization of thalamic neurons related to their target in low-, middle-, and high-frequency representations in cat auditory cortex. J Comp Neurol 1984; 227: 511–539. 23 Imig TJ, Morel A: Topographic organization in the ventral nucleus of the medial geniculate body in the cat. J Neurophysiol 1985;53: 309–340.

246

24 Redies H, Brandner S: Functional organization of the auditory thalamus in the guinea pig. Exp Brain Res 1991;86:384–392. 25 He J: OFF responses in the auditory thalamus of the guinea pig. J Neurophysiol 2002; 88:2377–2386. 26 Calford MB: The parcellation of the medial geniculate body of the cat defined by the auditory response properties of single units. J Neurosci 1983;3:2350–2364. 27 Calford MB, Webster WR: Auditory representation within principal division of cat medial geniculate body: an electrophysiology study. J Neurophysiol 1981; 45: 1013– 1028. 28 Imig TJ, Morel A: Organization of the cat’s auditory thalamus; in Edelman GM, Gall WE, Cowan WM (eds): Auditory Function: Neurobiological Bases of Hearing. New York, Wiley, 1988, pp 457–485. 29 Hu B: Cellular basis of temporal synaptic signalling: an in vitro electrophysiological study in rat auditory thalamus. J Physiol 1995; 483(Pt 1):167–182. 30 Winer JA, Morest DK: The medial division of the medial geniculate body of the cat: implications for thalamic organization. J Neurosci 1983;3:2629–2651. 31 Yu YQ, Xiong Y, Chan YS, He J: In vivo intracellular responses of the medial geniculate neurones to acoustic stimuli in anaesthetized guinea pigs. J Physiol 2004; 560(Pt 1):191–205. 32 Kimura A, Donishi T, Okamoto K, Tamai Y: Topography of projections from the primary and non-primary auditory cortical areas to the medial geniculate body and thalamic reticular nucleus in the rat. Neuroscience 2005;135:1325–1342. 33 Morel A, Kaas JH: Subdivisions and connections of auditory cortex in owl monkeys. J Comp Neurol 1992;318:27–63. 34 Hackett TA, Stepniewska I, Kaas JH: Thalamocortical connections of the parabelt auditory cortex in macaque monkeys. J Comp Neurol 1998;400:271–286. 35 Stiebler I, Neulist R, Fichtel I, Ehret G: The auditory cortex of the house mouse: leftright differences, tonotopic organization and quantitative analysis of frequency representation. J Comp Physiol [A] 1997; 8: 559– 571. 36 Imig TJ, Reale RA: Patterns of cortico-cortical connections related to tonotopic maps in cat auditory cortex. J Comp Neurol 1980; 192:293–332. 37 Imig TJ, Reale RA: Ipsilateral corticocortical projections related to binaural columns in cat primary auditory cortex. J Comp Neurol 1981;203:1–14. 38 Berman AL: Overlap of somatic and auditory cortical response fields in anterior ectosylvian gyrus of cat. J Neurophysiol 1961; 24: 595–607.

Neuroembryol Aging 2004–05;3:239–248

39 Berman AL: Interaction of cortical responses to somatic and auditory stimuli in anterior ectosylvian gyrus of cat. J Neurophysiol 1961;24:608–620. 40 Irvine DR, Huebner H: Acoustic response characteristics of neurons in nonspecific areas of cat cerebral cortex. J Neurophysiol 1979;42(1 Pt 1):107–122. 41 Toldi J, Feher O, Wolff JR: Sensory interactive zones in the rat cerebral cortex. Neuroscience 1986;18:461–465. 42 Clarey JC, Irvine DR: The anterior ectosylvian sulcal auditory field in the cat: I. An electrophysiological study of its relationship to surrounding auditory cortical fields. J Comp Neurol 1990;301:289–303. 43 Clarey JC, Irvine DR: The anterior ectosylvian sulcal auditory field in the cat: II. A horseradish peroxidase study of its thalamic and cortical connections. J Comp Neurol 1990;301:304–324. 44 Hofstetter KM, Ehret G: The auditory cortex of the mouse: connections of the ultrasonic field. J Comp Neurol 1992;323:370–386. 45 Barth DS, Kithas J, Di S: Anatomic organization of evoked potentials in rat parietotemporal cortex: somatosensory and auditory responses. J Neurophysiol 1993; 69: 1837– 1849. 46 Bendor D, Wang X: The neuronal representation of pitch in primate auditory cortex. Nature 2005;436:1161–1165. 47 Wang X, Lu T, Snider RK, Liang L: Sustained firing in auditory cortex evoked by preferred stimuli. Nature 2005;435:341–346. 48 Rouiller EM, Welker E: A comparative analysis of the morphology of corticothalamic projections in mammals. Brain Res Bull 2000;53:727–741. 49 Radtke-Schuller S: Cytoarchitecture of the medial geniculate body and thalamic projections to the auditory cortex in the rufous horseshoe bat (Rhinolophus rouxi). I. Temporal fields. Anat Embryol (Berl) 2004; 209: 59–76. 50 Radtke-Schuller S, Schuller G, O’Neill WE: Thalamic projections to the auditory cortex in the rufous horseshoe bat (Rhinolophus rouxi). II. Dorsal fields. Anat Embryol (Berl) 2004;209:77–91. 51 Redies H, Sieben U, Creutzfeldt OD: Functional subdivisions in the auditory cortex of the guinea pig. J Comp Neurol 1989; 282: 473–488. 52 Wallace MN, Rutkowski RG, Palmer AR: Identification and localization of auditory areas in guinea pig cortex. Exp Brain Res 2000;132:445–456. 53 Winer JA, Diamond IT, Raczkowski D: Subdivisions of the auditory cortex of the cat: the retrograde transport of horseradish peroxidase to the medial geniculate and posterior thalamic nuclei. J Comp Neurol 1977; 176: 387–417.

Zhang /Chan /He

54 Niimi K, Matsuoka H: Thalamocortical organization of the auditory system in the cat studied by retrograde axonal transport of horseradish peroxidase. Adv Anat Embryol Cell Biol 1979; 57:1–56. 55 Hashikawa T, Molinari M, Rausell E, Jones EG: Patchy and laminar terminations of medial geniculate axons in monkey auditory cortex. J Comp Neurol 1995;362:195–208. 56 Wepsic JG, Sutin J: Posterior thalamic and septal influence upon pallidal and amygdaloid slow wave and unitary activity. Exp Neurol 1964;10:67–80. 57 LeDoux JE, Farb C, Ruggiero DA: Topographic organization of neurons in the acoustic thalamus that project to the amygdala. J Neurosci 1990;10:1043–1054. 58 Cruikshank SJ, Edeline JM, Weinberger NM: Stimulation at a site of auditory-somatosensory convergence in the medial geniculate nucleus is an effective unconditioned stimulus for fear conditioning. Behav Neurosci 1992;106:471–483. 59 Shinonaga Y, Takada M, Mizuno N: Direct projections from the non-laminated divisions of the medial geniculate nucleus to the temporal polar cortex and amygdala in the cat. J Comp Neurol 1994;340:405–426. 60 Russchen FT: Amygdalopetal projections in the cat. I. Cortical afferent connections. A study with retrograde and anterograde tracing techniques. J Comp Neurol 1982; 206: 159–179. 61 Kimura A, Donishi T, Sakoda T, Hazama M, Tamai Y: Auditory thalamic nuclei projections to the temporal cortex in the rat. Neuroscience 2003;117:1003–1016. 62 Caviness VS Jr, Frost DO: Tangential organization of thalamic projections to the neocortex in the mouse. J Comp Neurol 1980; 194: 335–367. 63 Romanski LM, LeDoux JE: Information cascade from primary auditory cortex to the amygdala: corticocortical and corticoamygdaloid projections of temporal cortex in the rat. Cereb Cortex 1993;3:515–532. 64 Arnault P, Roger M: Ventral temporal cortex in the rat: connections of secondary auditory areas Te2 and Te3. J Comp Neurol 1990;302: 110–123. 65 Herkenham M: Laminar organization of thalamic projections to the rat neocortex. Science 1980;207:532–535. 66 Linke R, Schwegler H: Convergent and complementary projections of the caudal paralaminar thalamic nuclei to rat temporal and insular cortex. Cereb Cortex 2000; 10: 753– 771. 67 Ryugo DK, Killackey HP: Differential telencephalic projections of the medial and ventral divisions of the medial geniculate body of the rat. Brain Res 1974;82:173–177. 68 Huang CL, Winer JA: Auditory thalamocortical projections in the cat: laminar and areal patterns of input. J Comp Neurol 2000; 427: 302–331.

Interaction of Auditory Thalamus and Cortex

69 Redies H, Brandner S, Creutzfeldt OD: Anatomy of the auditory thalamocortical system of the guinea pig. J Comp Neurol 1989; 282: 489–511. 70 Lee CC, Schreiner CE, Imaizumi K, Winer JA: Tonotopic and heterotopic projection systems in physiologically defined auditory cortex. Neuroscience 2004;128:871–887. 71 Winer JA, Lee CC, Imaizumi K, Schreiner CE: Challenges to a neuroanatomical theory of forebrain auditory plasticity; in Merzenich MM (ed): Plasticity and Signal Representation in the Auditory System. New York, Kluver Academic/Plenum, 2004, pp 91–106. 72 Gerren RA, Weinberger NM: Long term potentiation in the magnocellular medial geniculate nucleus of the anesthetized cat. Brain Res 1983;265:138–142. 73 Mitani A, Shimokouchi M, Itoh K, Nomura S, Kudo M, Mizuno N: Morphology and laminar organization of electrophysiologically identified neurons in the primary auditory cortex in the cat. J Comp Neurol 1985; 235: 430–447. 74 Niimi K, Ono K, Kusunose M: Projections of the medial geniculate nucleus to layer 1 of the auditory cortex in the cat traced with horseradish peroxidase. Neurosci Lett 1984; 45: 223–228. 75 Cetas JS, de Venecia RK, McMullen NT: Thalamocortical afferents of Lorente de No: medial geniculate axons that project to primary auditory cortex have collateral branches to layer I. Brain Res 1999;830:203–208. 76 Kelly JP, Wong D: Laminar connections of the cat’s auditory cortex. Brain Res 1981;212: 1–15. 77 Rouiller EM, Colomb E, Capt M, De Ribaupierre F: Projections of the reticular complex of the thalamus onto physiologically characterized regions of the medial geniculate body. Neurosci Lett 1985;53:227–232. 78 Mitani A, Shimokouchi M: Neuronal connections in the primary auditory cortex: an electrophysiological study in the cat. J Comp Neurol 1985;235:417–429. 79 Hoogland PV, Wouterlood FG, Welker E, Van der Loos H: Ultrastructure of giant and small thalamic terminals of cortical origin: a study of the projections from the barrel cortex in mice using Phaseolus vulgaris leucoagglutinin (PHA-L). Exp Brain Res 1991; 87: 159–172. 80 Schwartz ML, Dekker JJ, Goldman-Rakic PS: Dual mode of corticothalamic synaptic termination in the mediodorsal nucleus of the rhesus monkey. J Comp Neurol 1991;309: 289–304. 81 Guillery RW: Anatomical evidence concerning the role of the thalamus in corticocortical communication: a brief review. J Anat 1995;187(Pt 3):583–592. 82 Liu XB, Honda CN, Jones EG: Distribution of four types of synapse on physiologically identified relay neurons in the ventral posterior thalamic nucleus of the cat. J Comp Neurol 1995;352:69–91.

83 Merzenich MM, Colwell SA, Anderson RA: Auditory forebrain organization: thalamocortical and corticothalamic connections in the cat; in Woolsey CN (ed): Cortical Sensory Organization: Multiple Auditory Areas. Clifton, Humana, 1982, pp 43–57. 84 Winer JA, Prieto JJ: Layer V in cat primary auditory cortex (AI): cellular architecture and identification of projection neurons. J Comp Neurol 2001;434:379–412. 85 Hefti BJ, Smith PH: Anatomy, physiology, and synaptic responses of rat layer V auditory cortical cells and effects of intracellular GABA(A) blockade. J Neurophysiol 2000;83: 2626–2638. 86 Jones EG, Powell TP: Electron microscopy of synaptic glomeruli in the thalamic relay nuclei of the cat. Proc R Soc Lond B Biol Sci 1969;172:153–171. 87 Jones EG, Powell TP: An electron microscopic study of the mode of termination of cortico-thalamic fibres within the sensory relay nuclei of the thalamus. Proc R Soc Lond B Biol Sci 1969; 172:173–185. 88 Rouiller EM, Welker E: Morphology of corticothalamic terminals arising from the auditory cortex of the rat: a Phaseolus vulgarisleucoagglutinin (PHA-L) tracing study. Hear Res 1991;56:179–190. 89 Ojima H: Terminal morphology and distribution of corticothalamic fibers originating from layers 5 and 6 of cat primary auditory cortex. Cereb Cortex 1994;4:646–663. 90 Bartlett EL, Stark JM, Guillery RW, Smith PH: Comparison of the fine structure of cortical and collicular terminals in the rat medial geniculate body. Neuroscience 2000; 100:811–828. 91 Sherman SM, Guillery RW: Functional organization of thalamocortical relays. J Neurophysiol 1996;76:1367–1395. 92 Sherman SM, Guillery RW: Exploring the Thalamus. San Diego, Academic Press, 2001. 93 Jones EG: The Thalamus. New York, Plenum, 1985. 94 Murphy PC, Sillito AM: Functional morphology of the feedback pathway from area 17 of the cat visual cortex to the lateral geniculate nucleus. J Neurosci 1996; 16: 1180– 1192. 95 Cox CL, Sherman SM: Glutamate inhibits thalamic reticular neurons. J Neurosci 1999; 19:6694–6699. 96 Pinault D, Smith Y, Deschenes M: Dendrodendritic and axoaxonic synapses in the thalamic reticular nucleus of the adult rat. J Neurosci 1997;17:3215–3233. 97 Ahlsen G, Lindstrom S, Lo FS: Interaction between inhibitory pathways to principal cells in the lateral geniculate nucleus of the cat. Exp Brain Res 1985;58:134–143. 98 Arcelli P, Frassoni C, Regondi MC, De Biasi S, Spreafico R: GABAergic neurons in mammalian thalamus: a marker of thalamic complexity? Brain Res Bull 1997;42:27–37.

Neuroembryol Aging 2004–05;3:239–248

247

99 Crabtree JW: Organization in the auditory sector of the cat’s thalamic reticular nucleus. J Comp Neurol 1998;390:167–182. 100 Guillery RW, Harting JK: Structure and connections of the thalamic reticular nucleus: advancing views over half a century. J Comp Neurol 2003;463:360–371. 101 Crabtree JW, Collingridge GL, Isaac JT: A new intrathalamic pathway linking modality-related nuclei in the dorsal thalamus. Nat Neurosci 1998;1:389–394. 102 Crabtree JW: Intrathalamic sensory connections mediated by the thalamic reticular nucleus. Cell Mol Life Sci 1999; 56: 683– 700. 103 Crabtree JW, Isaac JT: New intrathalamic pathways allowing modality-related and cross-modality switching in the dorsal thalamus. J Neurosci 2002;22:8754–8761. 104 Jones EG: Some aspects of the organization of the thalamic reticular complex. J Comp Neurol 1975;162:285–308. 105 Shosaku A, Sumitomo I: Auditory neurons in the rat thalamic reticular nucleus. Exp Brain Res 1983;49:432–442. 106 Simm GM, de Ribaupierre F, de Ribaupierre Y, Rouiller EM: Discharge properties of single units in auditory part of reticular nucleus of thalamus in cat. J Neurophysiol 1990; 63:1010–1021. 107 Villa AE: Physiological differentiation within the auditory part of the thalamic reticular nucleus of the cat. Brain Res Rev 1990;15:25–40. 108 Yingling CD, Skinner JE: Selective regulation of thalamic sensory relay nuclei by nucleus reticularis thalami. Electroencephalogr Clin Neurophysiol 1976; 41:476–482. 109 Bartlett EL, Smith PH: Anatomic, intrinsic, and synaptic properties of dorsal and ventral division neurons in rat medial geniculate body. J Neurophysiol 1999; 81: 1999– 2016. 110 Xiong Y, Yu YQ, Chan YS, He J: Effects of cortical stimulation on auditory-responsive thalamic neurons in anaesthetized guinea pigs. J Physiol 2004; 560(Pt 1):207– 217. 111 Yu YQ, Xiong Y, Chan YS, He J: Corticofugal gating of auditory information in the thalamus: an in vivo intracellular recording study. J Neurosci 2004;24:3060–3069. 112 He J, Yu YQ, Xiong Y, Hashikawa T, Chan YS: Modulatory effect of cortical activation on the lemniscal auditory thalamus of the guinea pig. J Neurophysiol 2002; 88: 1040– 1050.

248

113 He J: Corticofugal modulation on both ON and OFF responses in the nonlemniscal auditory thalamus of the guinea pig. J Neurophysiol 2003;89:367–381. 114 Hu B: Functional organization of lemniscal and nonlemniscal auditory thalamus. Exp Brain Res 2003;153:543–549. 115 Quirk GJ, Armony JL, LeDoux JE: Fear conditioning enhances different temporal components of tone-evoked spike trains in auditory cortex and lateral amygdala. Neuron 1997;19:481–484. 116 Duvel AD, Smith DM, Talk A, Gabriel M: Medial geniculate, amygdalar and cingulate cortical training-induced neuronal activity during discriminative avoidance learning in rabbits with auditory cortical lesions. J Neurosci 2001;21:3271–3281. 117 Maren S, Yap SA, Goosens KA: The amygdala is essential for the development of neuronal plasticity in the medial geniculate nucleus during auditory fear conditioning in rats. J Neurosci 2001;21:RC135. 118 Doyere V, Schafe GE, Sigurdsson T, LeDoux JE: Long-term potentiation in freely moving rats reveals asymmetries in thalamic and cortical inputs to the lateral amygdala. Eur J Neurosci 2003;17:2703–2715. 119 Rose HJ, Metherate R: Thalamic stimulation largely elicits orthodromic rather than antidromic cortical activation in an auditory thalamocortical slice. Neuroscience 2001;106:331–340. 120 Villa AE, Tetko IV, Dutoit P, De Ribaupierre Y, De Ribaupierre F: Corticofugal modulation of functional connectivity within the auditory thalamus of rat, guinea pig and cat revealed by cooling deactivation. J Neurosci Methods 1999;86:161–178. 121 King AJ: Signal selection by cortical feedback. Curr Biol 1997;7:R85–R88. 122 Alitto HJ, Usrey WM: Corticothalamic feedback and sensory processing. Curr Opin Neurobiol 2003;13:440–445. 123 Yan J, Suga N: Corticofugal modulation of time-domain processing of biosonar information in bats. Science 1996; 273: 1100– 1103. 124 Zhang Y, Suga N: Corticofugal amplification of subcortical responses to single tone stimuli in the mustached bat. J Neurophysiol 1997;78:3489–3492. 125 Zhang Y, Suga N, Yan J: Corticofugal modulation of frequency processing in bat auditory system. Nature 1997;387:900–903.

Neuroembryol Aging 2004–05;3:239–248

126 Jen PH, Chen QC, Sun XD: Corticofugal regulation of auditory sensitivity in the bat inferior colliculus. J Comp Physiol [A] 1998;183:683–697. 127 Yan W, Suga N: Corticofugal modulation of the midbrain frequency map in the bat auditory system. Nat Neurosci 1998;1:54–58. 128 Jen PH, Zhang JP: Corticofugal regulation of excitatory and inhibitory frequency tuning curves. Brain Res 1999;841:184–188. 129 He J: Slow oscillation in non-lemniscal auditory thalamus. J Neurosci 2003;23:8281– 8290. 130 Steriade M: Impact of network activities on neuronal properties in corticothalamic systems. J Neurophysiol 2001; 86:1–39. 131 Steriade M, Timofeev I: Neuronal plasticity in thalamocortical networks during sleep and waking oscillations. Neuron 2003; 37: 563–576. 132 Suga N, Ma X: Multiparametric corticofugal modulation and plasticity in the auditory system. Nat Rev Neurosci 2003;4:783– 794. 133 Suga N, Gao E, Zhang Y, Ma X, Olsen JF: The corticofugal system for hearing: recent progress. Proc Natl Acad Sci USA 2000;97: 11807–11814. 134 Suga N, Gao E, Ma X, Sakai M, Chowdhury S: Corticofugal system and processing of behaviorally relevant sounds: perspective. Acoust Sci Tech 2001;22:85–91. 135 Steriade M: The GABAergic reticular nucleus: a preferential target of corticothalamic projections. Proc Natl Acad Sci USA 2001;98:3625–3627. 136 Miller LM, Escabi MA, Read HL, Schreiner CE: Functional convergence of response properties in the auditory thalamocortical system. Neuron 2001;32:151–160. 137 Alitto HJ, Usrey WM: Dynamic properties of thalamic neurons for vision. Prog Brain Res 2005;149:83–90. 138 Sillito AM, Jones HE: Corticothalamic interactions in the transfer of visual information. Philos Trans R Soc Lond B Biol Sci 2002;357:1739–1752. 139 Sillito AM, Cudeiro J, Jones HE: Always returning: feedback and sensory processing in visual cortex and thalamus. Trends Neurosci 2006;29:307–316. 140 He J: Auditory cognition and filtering; in Luo Y, Jiang Y and Cheng K (eds): Cognitive Neuroscience Tutorial. Beijing, Beijing University Press, 2006, pp 263–295.

Zhang /Chan /He

Author Index Vol. 3, No. 4, 2004–05

Angelaki, D.E. 171 Bergquist, F. 183 Chan, Y.S. 161, 162, 239 Curthoys, I.S. 194 de Waele, C. 194 Dutia, M.B. 183 Edeline, J.M. 230

He, J. 161, 239 Huetz, C. 230 Lai, C.H. 162 Malmierca, M.S. 215 Menzies, J.R.W. 183 Moore, L.E. 194 Newlands, S.D. 171

Paterson, J.M. 183 Shum, D.K.Y. 162 Straka, H. 194 Tse, Y.C. 162 Vibert, N. 194 Vidal, P.P. 194 Zhang, Z. 239

Subject Index Vol. 3, No. 4, 2004–05

Auditory cortex 239 – midbrain 215 Bone conduction 207 Brainstem 183 Cerebellum 183 Compensation 194 Corticothalamic projections 239 Deafferentation 183 Dynamic filter 239 Electrical stimulation 171 Eye movement 171 Glutamatergic neurotransmission 162 Hearing 215 Inferior colliculus 215 Information theory 230

© 2005 S. Karger AG, Basel Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

Intrinsic membrane properties 194 Labyrinth 183, 207 Labyrinthectomy 171 Laminar organization 215 Learning-induced plasticity 230 Lesion 171 Medial geniculate body 239 Myogenic potential 207 Oculomotor control 194 Otolith 207 Plasticity 183 Postural control 194 Pre- and postnatal development 162 Primate 171 Proprioception 194

Saccule 207 Sleep 230 Spatiotemporal information 162 Spike timing precision 230 Stress 183 Thalamic reticular nucleus 239 Thalamocortical auditory system 230 – projections 239 Tonotopic organization 215 VEMP 207 Vestibulo-ocular reflex 171 Vibration 207 Vocalization 230

249

Author Index Vol. 3, 2004–05

Kozlov, V.A. 99 Krsnik, Ž. 19

Abramov, V.V. 99 Angelaki, D.E. 171 Anstrom, J.A. 4 Bergquist, F. 183 Block, S.M. 4 Brown, W.R. 4 Challa, V.R. 4 Chan, A.O.-K. 75 102 Chan, W.Y. 27, 69, 102, 115 Chan, Y.-S. 36, 161, 162, 239 Cheon, M.S. 1 Chuah, M.I. 152 Curthoys, I.S. 194

Lai, C.-H. 162 Lai, H.W.L. 47 Lam, C.-W. 75 Lam, R.W.C. 36 Lam, S.T.-S. 75 Leung, T.C.H. 123 Li, Q. 111 Liu, F. 70, 136 Liu, Z. 115 Lo, I.F.-M. 75 Lubec, G. 1, 60 Mak, Y.T. 92 Malmierca, M.S. 215 Menzies, J.R.W. 183 Moody, D.M. 4 Moore, L.E. 194

de Waele, C. 194 Demirant, A. 149 Dong, D. 115 Dong, M. 102 Dutia, M.B. 183

Newlands, S.D. 171 Ng, T.B. 70, 128, 136

Edeline, J.-M. 230 Engidawork, E. 60

Taşdemir, S. 149 Thore, C.R. 4 Timmermans, J.-P. 142 Tiu, S.C. 27 Tiu, S.-C. 75 Tong, S.-F. 75 Tse, Y.-C. 162 Tuncyurek, O. 65 Ulfig, N. 60 Vibert, N. 194 Vidal, P.P. 194 Vincent, A.J. 152 Wang, C.C. 78 Wang, H.X. 70, 136 Wang, L. 102 Wang, M. 111 West, A.K. 152 Wolff, L.T. 111 Wong, H.W. 92 Wu, M. 142 Wu, Y. 47

Orguc, S. 65 Xia, L.X. 70, 136

Ferrando-Miguel, R. 1 Paterson, J.M. 183 Petanjek, Z. 19 Pollak, A. 60 Poveshchenko, A.F. 99

Goktan, C. 65 Golalipour, M.J. 146 He, J. 161, 239 Hepner, F. 60 Heydari, K. 146 Hu, Y. 115 Huetz, C. 230

Rebenko, N.M. 99 Serter, S. 65 Shek, C.-C. 75 Shi, Y. 123 Shum, D.K.-Y. 162 Straka, H. 194

Ichinohe, A. 13, 42 Itoh, M. 42 Jiang, Y. 115 Jovanov-Milošević, N. 19 Judaš, M. 19 Korotkova, N.A. 99 Kostović, I. 19

© 2005 S. Karger AG, Basel Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

Tabira, T. 13 Tacar, O. 149 Takahashi, K. 13 Takashima, S. 13, 42 Tarhan, S. 65 Taşdemir, N. 149

Yang, W. 123 Ye, X.Y. 70, 128 Yew, D.T. 27, 47, 69, 111 Yew, D.T.W. 92 Yu, M.C. 92 Yung, K.K.L. 36, 123 Zhang, A. 47 Zhang, C. 142 Zhang, J.-S. 123 Zhang, L. 111 Zhang, Y. 115 Zhang, Z. 239

Subject Index Vol. 3, 2004–05

Aging 70, 128, 136, 142 Alkaline phosphatase 4 Anthropometry 146 Antioxidant 70 Apoptosis 47, 92 Arnold-Chiari syndrome 149 Atrophy 111 Auditory cortex 239 – midbrain 215 Basal ganglia 36, 111 Bcl-2 47 Bone conduction 207 Brain 4, 47, 70, 128, 136 – development 13 – hypoxia 42 – malformation 149 – weight 146 Brainstem 183 CADASIL 13 Cajal Retzius cells 19 Carnitine 128 Caspase 47 Caudate nucleus 115 Cell fate 13 – proliferation 27 Cephalometry 146 Cerebellum 183 Chiari 2 malformation 65 Chinese medicine 123 Choroid plexus 115 Cognitive ability 136 Collagen 4 Compensation 194 Cortex 111 Cortical afferents 19 Corticothalamic projections 239 Cranial capacity 146 Cresyl violet staining 102 Deafferentation 183 Development 27, 60 DJ-1 60 Dynamic filter 239 Electrical stimulation 171 Enteric nervous system 142 Esophagus 142 Eye movement 171 Fetal brain 60 Fos-immunoreactive neurons 123

© 2005 S. Karger AG, Basel Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/nba

Genetic regulation 78 Glial fibrillary acidic protein 27 Glutamatergic neurotransmission 162 G-protein-coupled receptor 36 Hearing 215 Hippocampus 123 Human fetal neocortex 27 – neonate 4 Hypothalamus 115 Immunofluorescence 36 Inferior colliculus 215 Information theory 230 Inhibitor of apoptosis proteins 47 Innate immune system 152 Intrinsic membrane properties 194 Labyrinth 183, 207 Labyrinthectomy 171 Laminar organization 215 Learning-induced plasticity 230 Lesion 171 Limb buds 102 Lipoic acid 128 Long-term potentiation 92 Magnetic resonance imaging 65, 149 Mass spectrometry 60 Medial geniculate body 239 Medicinal herbs 136 Melatonin 70 Microtubule-associated protein 2 19 Migration disorder 13 Molecular evolution 78 Muscle fatigue 123 Myogenic potential 207 Neonatal screening 75 Neural tube 102 Neurodegenerative disease 47 Neuron necrosis 42 Neuroprotection 152 Nitrergic neurons 142 Nitric oxide 92 – – synthase isoform 92 Notch3 13 Nuclear factor-kappa B 115 Oculomotor control 194 Olfactory ensheathing cells 152 Otolith 207

251

Parietal cortex 115 Parkinson’s disease 111 Phagocytosis 152 Plasticity 183 Postural control 194 Pre- and postnatal development 162 Prenatal brain damage 42 – diagnosis 75 Primate 171 Proliferation cell nuclear antigen 27 Proprioception 194 Proteomics 60 Rhesus monkey 115 Rhombencephalon 78 Saccule 207 Sleep 230 SNAP-25 19 Somites 102 Spatiotemporal information 162 Spike timing precision 230 Stress 183

252

Neuroembryol Aging Vol. 3, 2004–05

Thalamic reticular nucleus 239 Thalamocortical auditory system 230 – projections 239 Thalamus 111, 115 Thyroglossal duct cyst 65 Thyroid hormone resistance syndrome 75 Thyrotropin 75 Tonotopic organization 215 Transient sublayers 19 Two-dimensional electrophoresis 60 Vascular development 4 VEMP 207 Vestibulo-ocular reflex 171 Vibration 207 Vimentin 27 Vitamins 128 Vocalization 230 Wheat germ agglutinin-gold conjugates 102 Whole embryo culture 102

Subject Index

Contents Vol. 3, 2004–05

No. 1

No. 2

Editorial 1 Aberrant Chromosome 21 Gene Products: Explaining the

Down Syndrome Phenotype? Ferrando-Miguel, R.; Cheon, M.S.; Lubec, G. (Vienna)

Original Papers 4 Histological Analysis of Vascular Patterns and Connections in

the Ganglionic Eminence of Premature Neonates Anstrom, J.A.; Thore, C.R.; Moody, D.M.; Challa, V.R.; Block, S.M.; Brown, W.R. (Winston-Salem, N.C.) 13 Early and Late Development of Notch3 in Human Brains Ichinohe, A. (Sendai); Takahashi, K.; Tabira, T. (Aichi); Takashima, S. (Fukuoka)

69 Editorial Yew, D.T. (Hong Kong); Chan, W.Y. (Hong Kong)

Mini Review 70 Melatonin and the Aging Brain Ng, T.B. (Hong Kong); Liu, F. (Tianjin); Xia, L.X. (Shenzhen); Wang, H.X. (Beijing); Ye, X.Y. (Fuzhou)

Case Report 75 Normal Cord Blood Thyroid-Stimulating Hormone in a Child

with Resistance to Thyroid Hormone Chan, A.O.-K.; Lam, C.-W.; Lo, I.F.-M.; Lam, S.T.-S.; Shek, C.-C.; Tiu, S.-C.; Tong, S.-F. (Hong Kong)

19 Laminar Organization of the Marginal Zone in the Human

Fetal Cortex Kostović, I.; Jovanov-Milošević, N.; Krsnik, Ž.; Petanjek, Z.; Judaš, M. (Zagreb) 27 Cell Proliferation in the Developing Human Cerebral Cortex Tiu, S.C.; Chan, W.Y.; Yew, D.T. (Hong Kong) 36 Differential Localization of Metabotropic Glutamate

Receptors 1
and 2/3 in the Rat Striatum during Early Postnatal Development Lam, R.W.C. (Kowloon Tong); Chan, Y.S. (Hong Kong); Yung, K.K.L. (Kowloon Tong)

42 Pathogenesis and Prevention of Pontosubicular Necrosis Takashima, S. (Yanagawa); Ichinohe, A. (Sendai); Itoh, M. (Tokyo)

Review 78 Development of the Rhombencephalon: Molecular Evolution

and Genetic Regulation Wang, C.C. (Hong Kong)

Original Papers 92 Different Aging Patterns of the Neuronal and Inducible

Isoforms of Nitric Oxide Synthase in Mouse Hippocampus by Immunohistochemistry Wong, H.W.; Mak, Y.T. (Hong Kong); Yu, M.C. (Newark, N.J.); Yew, D.T.W. (Hong Kong) 102 Somite as a Morphological Reference for Staging and Axial

Levels of Developing Structures in Mouse Embryos

Review

Chan, A.O.K.; Dong, M.; Wang, L.; Chan, W.Y. (Hong Kong) 47 Apoptosis – A Brief Review Zhang, A. (Beijing); Wu, Y. (Munich); Lai, H.W.L.; Yew, D.T. (Hong Kong)

Short Communication 60 Chemical Identification and Characterization of DJ-1 Protein

in Human Fetal Amygdala Hepner, F. (Vienna); Engidawork, E. (Vienna/Addis Ababa); Ulfig, N. (Rostock); Pollak, A.; Lubec, G. (Vienna)

Case Report 65 Chiari Type 2 Anomaly Associated with a Thyroglossal Cyst Orguc, S.; Tuncyurek, O.; Serter, S.; Tarhan, S.; Goktan, C. (Manisa)

© 2005 S. Karger AG, Basel Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Access to full text and tables of contents, including tentative ones for forthcoming issues: www.karger.com/nba_issues

Letter to the Editor 99 Asymmetrical Expression of Interleukin-1, Interleukin-1

Receptor and Erythropoietin Receptor in Mouse Brain Hemispheres Poveshchenko, A.F.; Korotkova, N.A.; Rebenko, N.M.; Kozlov, V.A.; Abramov, V.V. (Novosibirsk)

No. 3

No. 4

Original Papers 111 Degeneration in the Brains of Parkinson’s Disease Patients Zhang, L.; Wang, M. (Shijizhuang); Li, Q.; Yew, D.T.; Wolff, L.T. (Shatin) 115 Expression of Nuclear Factor-Kappa B in Early Developing

Rhesus Monkey Brains Dong, D.; Zhang, Y.; Jiang, Y.; Hu, Y.; Liu, Z. (Huazhong); Chan, W.Y. (Hong Kong) 123 Effects of Traditional Chinese Medicine Radix Astragali,

161 Editorial Overview Chan, Y.S.; He, J. (Hong Kong) 162 Toward Maturation of the Vestibular System: Neural Circuits

and Neuronal Properties Lai, C.H.; Tse, Y.C.; Shum, D.K.Y.; Chan, Y.S. (Hong Kong) 171 What Peripheral Vestibular Manipulations Reveal about the

Function and Plasticity in the Primate Oculomotor System Newlands, S.D. (Galveston, Tex.); Angelaki, D.E. (St. Louis, Mo.)

Fructus Aurantii and Rhizoma Cypei Extracts on c-Fos Expression in the Aging Rat Hippocampus Induced by Exercise

183 Cellular Mechanisms of Vestibular Compensation Paterson, J.M.; Menzies, J.R.W.; Bergquist, F.; Dutia, M.B. (Edinburgh)

Shi, Y.; Yang, W. (Hong Kong); Zhang, J.-S. (Xian); Leung, T.C.H.; Yung, K.K.L. (Hong Kong)

194 Recovery after Vestibular Lesions: From Animal Models to

Reviews

207 Peripheral Vestibular Responses to Sound Curthoys, I.S. (Sydney)

Patients Vidal, P.P.; Straka, H.; Vibert, N.; Moore, L.E.; de Waele, C. (Paris)

128 Beneficial Effects of Dietary Vitamins, Lipoic Acid and

Carnitine on the Aging Brain Ng, T.B. (Hong Kong); Ye, X.Y. (Fuzhou) 136 Plants Beneficial to the Aging Brain Ng, T.B. (Hong Kong); Wang, H.X. (Beijing); Liu, F. (Tianjin); Xia, L.X. (Shenzhen)

215 The Inferior Colliculus: A Center for Convergence of Ascending

and Descending Auditory Information Malmierca, M.S. (Salamanca) 230 From Receptive Field Dynamics to the Rate of Transmitted

Information: Some Facets of the Thalamocortical Auditory System Huetz, C.; Edeline, J.M. (Orsay)

Mini Review 142 Morphological Features of the Esophageal Enteric Nervous

System and Its Stereological Changes during Aging Wu, M.; Zhang, C. (Qingdao); J.-P.Timmermans, J.-P. (Antwerp)

Short Communication 146 Effect of the Ethnic Factor on Cranial Capacity and Brain

239 Thalamocortical and Corticothalamic Interaction in the

Auditory System Zhang, Z.; Chan, Y.S.; He, J. (Hong Kong) 249 Author Index/Subject Index Vol. 3, No.4, 2004–05 250 Author Index Vol. 3, 2004–05 251 Subject Index Vol. 3, 2004–05

Weight of Male Newborns in Northern Iran Golalipour, M.J.; Heydari, K. (Gorgan)

Case Report 149 Arnold-Chiari Syndrome (Type I) Taşdemir, N.; Tacar, O.; Demirant, A.; Taşdemir, S. (Diyarbakır)

Commentary 152 Do Olfactory Ensheathing Cells Play a Role in the Defense of

the Brain against Infection? Chuah, M.I.; Vincent, A.J.; West, A.K. (Tasmania)

IV

Neuroembryol and Aging Vol. 3, 2004–05

Contents