Organization and development of thalamocortical pathways in the rabbit auditory system.
Item Type text; Dissertation-Reproduction (electronic)
Authors de Venecia, Ronald Kent.
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 28/09/2021 07:46:24
Link to Item http://hdl.handle.net/10150/187176 INFORMATION TO USERS
This ~uscript )las been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely. event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and contim1jng from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manlls.aipt have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
UMI A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor. MI4S106·1346 USA 3131761-4700 800:521·0600
ORGANIZATION AND DEVELOPMENT OF
THALAMOCORTICAL PATHWAYS
IN THE RABBIT AUDITORY SYSTEM
by
Ronald Kent de Venecia
A Dis~ertationSubmitted to the Faculty of the
DEPARTMENT OF CELL BIOLOGY AND ANATOMY
In Partial Fulfillment of the Requirements For the Degree of
DOCTOR OF PHILOSOPHY
In the Graduate College
THE UNIVERSITY OF ARIZONA
1995 OMI Number: 9534682
OMI Microform 9534682 Copyright 1995, by OMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 2 THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE
As members of the Final Examination Committee, we certify that we have
read the dissertation prepared by Ronald Kent de Venecia
entitled Organization and Development of Thalamocortical Pathways
in the Rabbit Auditory System
and recommend that it be accepted as fulfilling the dissertation
requirement for the Degree of ___D__ o_c_to_r __ o_f __ P_h_i_lo_s_o~p~h~y ______
Nathaniel T. McMullen, Ph.D. ~P.C7~ Leslie P. Tolbert, Ph.D. Date I pa~1 f· Jdt?!h::-- ¥1I19~
Date" ~l 'II 4ft!?.) Date
Final approval and acceptance of this dissertation is contingent upon the candidate's submission of the final copy of the dissertation to the Graduate College.
I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.
Dissertation Director Nathaniel T. McMullen 3
STATEMENT BY AUTHOR
This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.
Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.
SIGNED: -z::~ 4
ACKNOWLEDGEMENTS
I extend my deepest gratitude to my dissertation advisor, Nathaniel T. McMullen, whose support and guidance made this work possible. Nate has provided me with a solid foundation in both the neurosciences and the scientific method of research that will enable me to become a productive neurobiologist. He is more than just a mentor to me, he is also a true and trusted friend. Although I will miss the enjoyment of our daily interactions and his wry sense of humor, I look forward to continuing our professional and personal relationship in the future. I also thank the members of my advisory committee, Leslie Tolbert, Paul St. John, Nick Strausfeld and Raphael Gruener. Their advice, constructive criticism, and constant encouragement have both made me a better scientist and allowed me to complete this work. My sincere gratitude also extends to Chad Smelser, whose expert technical assistance in all phases of the experiments described herein made this project possible. I will always value his friendship and support. I thank the members of the Department of Cell Biology and Anatomy for their kindness and encouragement during my graduate career. I especially appreciate the support they gave to me and my family when we suffered the loss of my father. Finally, I especially thank my mother, Eleanor S. de Venecia, whose unfailing love and understanding provide me with strength and encouragement during my most difficult times. 5
DEDICATION
In memory of my father,
Nestor Flor de Venecia, M.D. 6
TABLE OF CONTENTS
LIST OF ILLUSTRATIONS ...... 7
LIST OF TABLES ~ ...... 10
ABSTRACT ...... 11
CHAPTER 1. INTRODUCTION ...... 13 Organization of the Medial Geniculate Body...... 13 Functional Maps of Auditory Cortex ...... 14 Multiple Parallel Pathways to Auditory Cortex ...... 16 Calcium-binding Proteins and Parallel Thalamocortical Pathways ...... 17 Development of Auditory Neocortex and Thalamocortical Projections ...... 19 Specific Aims ...... 26
CHAPTER 2. COMPLEMENTARY EXPRESSION OF PARVALBUMIN AND CALBINDIN D-28k DELINEATES SUBDIVISIONS OF THE RABBIT MEDIAL GENICULATE BODY ...... 28 Abstract ...... 28 Introduction ...... 30 Methods ...... 34 Results ...... 36 Discussion ...... 41
CHAPTER 3. THALAMOCORTICAL PATCHES IN AUDITORY NEOCORTEX79 Abstract ...... 79 Introduction ...... 80 Methods ...... 81 Results ...... 84 Discussion ...... 84
CHAPTER 4. PARVALBUMIN IS EXPRESSED IN A RECIPROCAL CIRCmT LINKING THE MEDIAL GENICULATE BODY AND AUDITORY NEOCORTE)92 Abstract ...... 92 Introduction ...... 93 Methods ...... 95 Results ...... 102 Discussion ...... 106 7
TABLE OF CONTENTS-Continued
CHAPTER 5. SINGLE THALAMOCORTICAL AXONS DIVERGE TO MULTIPLE PATCHES IN NEONATAL AUDITORY CORTEX ...... 139 Abstract ...... 139 Introduction ...... 140 Methods ...... 142 Results ...... 145 Discussion ...... 146
CHAPTER 6. DISCUSSION ...... 157 Ascending Pathways Relayed by the MGB to Auditory Cortex ...... 157 Descending Pathways From Auditory Cortex ...... 158 Summary and Discussion of Major Findings in Present Study ...... 160
REFERENCES ...... 169 8
LIST OF ILLUSTRATIONS
FIGURE 2.1 Posterior view of adult rabbit brain revealing MGB ...... 60
FIGURE 2.2 Low magnification view of the complementary staining patterns of PV and CB MGB ...... 62
FIGURE 2.3 PV and CB stained coronal sections near anterior pole of MGB ...... 64
FIGURE 2.4 PV and CB stained coronal MGB sections 1mm posterior to previous figure ...... 66
FIGURE 2.5 PV and CB stained coronal MGB sections l.4mm posterior to previous figure ...... 68
FIGURE 2.6 PV and CB stained coronal MGB sections 2.2mm posterior to previous figure ...... 70
FIGURE 2.7 PV and CB stained coronal MGB sections 2.9mm posterior to previous figure ...... 72
FIGURE 2.8 PV and CB stained sagittal MGB sections ...... 74
FIGURE 2.9 Summary diagram of PV and CB neurons in the MGB ...... 76
FIGURE 3.1 Coronal sections of thalamus and ipsilateral auditory cortex showing patch-like distribution of thalamocortical axons ...... 89
FIGURE 3.2 High magnification of patch and diagram of patch location in primary auditory field ...... 91
FIGURE 4.1 Rabbit auditory cortex stained with antibodies to parvalbumin ...... 118
FIGURE 4.2 Comparison of BDA-Iabeled thalamocortical patches and PV patches...... 120
FIGURE 4.3 High magnification view of BDA and PV patches ...... 122
FIGURE 4.4 Photomicrograph of BDA thalamocortical patch with Golgi-like labeling of Layer VIa corticothalamic neurons ...... 124 9
LIST OF ILLUSTRATIONS-Continued
FIGURE 4.5 Camera lucida reconstruction of layer VIa corticothalamic neurons ...... 126
FIGURE 4.6 Photomicrograph of PV + layer VIa corticothalamic pyramidal neurons ...... 128
FIGURE 4.7 Fluorescence photomicrographs of double-labeled pyramidal cells in layer VIa ...... 130
FIGURE 4.8 Coronal sections of MGB revealing PV immunoreactivity and neuronal morphology ...... 132
FIGURE 4.9 Retrograde double-labeling studies: cortical injection site and low magnification photomicrograph of MGV ...... 134
FIGURE 4.10 Retrograde double-labeling studies of MGV neurons ...... 136
FIGURE 4.11 Schematic of reciprocal circuits linking auditory neocortex and MGV ...... 138
FIGURE 5.1 Thalamocortical patches at postnatal day 1 150
FIGURE 5.2 Thalamocortical patches at postnatal day 3 and MGB projection neurons ...... 152
FIGURE 5.3 Computer microscope reconstruction of thalamocortical axon forming two patches ...... 154
FIGURE 5.4 High magnification view of patch shown in previous figure ...... 156 10
LIST OF TABLES
TABLE 2.1 Abbreviations ...... 77 11
ABSTRACT
Thalamocortical relations in the rabbit auditory system were investigated by calcium-binding protein immunohistochemistry and neuroanatomical tracing techniques.
The differential distribution of the calcium-binding proteins parvalbumin and calbindin delineated four major subdivisions of the medial geniculate body (MGB): the ventral, dorsal, medial and internal nuclei. In addition, several subnuclei that are not easily distinguished by routine Nissl stains alone were clearly identified. A comparison with previous studies of MGB connectivity suggests that parvalbumin expression is highest in subdivisions that receive substantial input from the central nucleus of the inferior colliculus and that project to primary auditory cortex (AI). In contrast, calbindin expression characterizes nuclei that receive input primarily from other sources and that project to secondary auditory cortex. Anterograde axonal tracing experiments demonstrated that thalamocortical axons originating from coincident cell groups in the ventral division of the MGB (MGV) terminated primarily within patches occupying the full depth of layer IV and the bottom of layer III in primary auditory cortex. Less dense zones of termination were located in layers I and VI. The intermittent distribution of the patches seen in the coronal plane was similar to the distribution of binaural interaction regions in AI. Furthermore, the patches were elongated in the rostral-caudal axis forming bands that parallel the isofrequency contours in AI. Double-labeling experiments employing PV-immunohistochemistry and neuroanatomical tracing indicate that PV is 12 expressed in biochemically distinct thalamocortical and corticothalamic pathways between the MGV and AI. These results are consistent with a model of MGV organization containing functionally distinct, parallel anatomical pathways to primary auditory cortex.
Finally, anterograde axonal tracing studies in neonatal animals indicate that the thalamocortical axons segregate into patches in the absence acoustically driven activity.
The morphology of individual axons and patches in neonates, however, suggests significant axonal remodeling during postnatal maturation. These results are discussed in terms of contribution of thalamocortical afferents to the formation of the functional architecture in auditory neocortex. 13
CHAPTER ONE
INTRODUCTION
Organization of the MGB
The mammalian medial geniculate body (MGB) is a complex of nuclei subserving auditory function in the dorsal thalamus. The complex is conventionally divided into three main subdivisions: the ventral, dorsal, and medial nuclei. The ventral division (or nucleus) primarily receives topographically organized projections from isofrequency laminae in the central nucleus of the inferior colliculus and is generally recognized as the principal thalamic relay for the ascending central auditory pathway to primary auditory cortex (Kudo and Niimi, 1978; Oliver and Hall, 1978a; 1978b; Andersen et aI., 1980a;
1980b; Calford and Aitkin, 1983). In addition, the ventral division receives reciprocal descending projections originating from pyramidal neurons in layer VI of AI (Andersen et aI., 1980a; Kelly and Wong, 1981; Wong and Kelly, 1981; Winer and Larue, 1987).
The dorsal division receives input from other sources, such as the paracentral nuclei of the inferior colliculus, the lateral midbrain tegmentum, and the superior colliculus
(Morest, 1965; Oliver and Hall, 1978a; 1978b; Andersen et aI., 1980a; LeDoux et aI.,
1985). The dorsal division also has reciprocal connections with nonprimary auditory cortical fields surrounding AI (Oliver and Hall, 1978b; Andersen et aI., 1980b; Scheel,
1980; Rouiller and de Ribaupierre, 1985). Unlike the ventral division, there is no evidence of spatial organization of projections to the dorsal division (Morest, 1965; 14
Oliver and Hall, 1978a; 1978b; Calford and Aitkin, 1983). The medial division receives
afferents from most, if not all, of the midbrain areas which project to the other divisions
(Oliver and Hall, 1978a; Winer, 1983a; Rouiller et aI., 1989). The term internal division
has often been used synonymously with the medial division in species that lack a
magnocellular population of neurons, a characteristic feature of the medial division in the
cat and primate (Jones, 1985).
Despite similarities in nomenclature, there is considerable disagreement regarding
the number and grouping of MGB subdivisions, and homology between MGB
subdivisions in different species is often unclear (see Oliver and Hall, 1978a; LeDoux,
1985; 1987; Morest and Winer, 1986; Winer, 1992). Although a variety of criteria have
been used to parcel the MGB, including afferent and efferent connectivity, the
electrophysiological characteristics of neurons, and neuronal morphology in Golgi stained preparations, the most commonly used criterion is regional cytoarchitecture as seen in
Nissl stained sections. The degree of cytoarchitectonic differentiation of the dorsal thalamus in Nissl stained sections, however, varies considerably among species and is a source for difficulty in comparative studies of MGB organization.
Functional Maps of Auditory Cortex
The auditory neocortex of mammals contains functional maps of sound frequency and binaural interactions (Brugge and Reale, 1985; Aitkin, 1990; Phillips et aI., 1991;
Clarey et aI., 1992). Although the overall organization of the thalamocortical pathways that subserve these maps have been inferred from functional (Imig and Morel, 1983; 15
1988), degeneration (Mesulam and Pandya, 1973; Sousa-Pinto, 1973; Niimi and Naito,
1974; Frost and Caviness, 1980; Vaughan, 1983), anterograde (Jones and Burton, 1976;
LeDoux et aI., 1985) and retrograde tracing experiments (Colwell and Merzenich, 1975;
Winer et aI., 1977; Imig and Morel, 1983; Middlebrooks and Zook, 1983; Morel and
Imig, 1987), there is little direct information concerning the morphology and laminar termination of auditory thalamocortical pathways in any species. Numerous studies conducted on the cat thalamocortical auditory system have resulted in a
"divergence/convergence" theory of geniculocortical organization (Merzenich et aI.,
1982; 1984). In this species, high frequency sound is represented caudally and low frequency sound rostrally in the primary auditory cortical field (AI). A striking feature of this topographical organization is that any given sound frequency is represented along a mediolaterally oriented line, an isofrequency contour. Cortical injections of HRP into physiologically defined isofrequency loci have shown that restricted areas of the ventral division of the MOB (MOV) diverge to widespread areas along an isofrequency contour
(Colwell and Merzenich, 1975; Merzenich et aI, 1982). These results suggest that the tonotopic map of auditory cortex derives from the preferential arborization of thalamocortical afferents along an isofrequency contour. The second functional map of the auditory cortex consists of elongated regions in which neurons driven by stimulation of the contralateral ear are either facilitated (EE) or inhibited (EI) by simultaneous stimulation of the ipsilateral ear (Imig and Adrian, 1977). In the cat, EE/EI binaural interaction bands are oriented orthogonal to isofrequency contours in the form of alternating bands (Imig and Adrian, 1977; Middlebrooks et aI., 1980; Kelly and Sally, 16
1988). Experiments in which retrograde tracers were placed within electrophysiologically-defined binaural loci demonstrate that EE and EI bands receive convergent projections from spatially segregated neuronal groups within MGV
(Middlebrooks and Zook, 1983).
The "divergence/convergence" scheme of auditory thalamocortical organization has been questioned in favor of a simple "point-to-point" topographic projection from
MGV to AI (Redies et aI., 1989; Bradner and Redies, 1990). A point-to-point model of auditory thalamocortical pathways predicts a "nonpatchy" distribution of thalamocortical afferents originating from a single locus in the thalamus. Thus, focal injections of anterograde tracers into MGV would label a continuous cortical band of terminal axons whose long axis is parallel to isofrequency contours and whose tangential extent would be a direct function of injection size.
Multiple parallel pathways to auditory cortex.
Another important aspect of sensory thalamocortical relations is the existence of multiple, independent parallel pathways to primary sensory cortex (Diamond, 1983;
Winer, 1991; 1992). The demonstration of divergent thalamic projections to isofrequency contours and convergent thalamic projections to binaural bands indicate the existence of separate parallel pathways from MGV to AI in the cat (Middlebrooks and Zook, 1983).
Furthermore, these data suggest that there are multiple types of MGV neurons which project to AI. In support of this organization are retrograde transport studies which have demonstrated at least two types of MGV relay neurons to AI: cells with tufted dendritic 17
arbors, considered to be the principal relay cell in the MGV, and less common smaller cells with oval or drumstick-shaped somato-dendritic profiles (Winer, 1984; 1991).
However, direct evidence for morphologically distinct classes of thalamocortical axons arising from the MGV is lacking. Further evidence for the existence of multiple independent parallel pathways to auditory cortex is provided by retrograde tracing experiments indicating that 5-20% of cells in MGV project to more than one auditory cortical field (Morel and Imig, 1987).
Calcium-binding proteins and parallel thalamocortical pathways.
An emerging theme in the field of thalamocortical relations is that thalamic relay neurons belonging to separate parallel pathways (e.g. lemniscal vs nonlemniscal), and which project to different layers of primary sensory neocortex, can be further distinguished by their expression of calcium binding proteins, such as parvalbumin and calbindin (Rausell and Jones, 1991a; 1991b; Diamond et at., 1993). Studies of the visual and somatosensory system have demonstrated that single thalamic subdivisions give rise to multiple parallel pathways to sensory neocortex (Jones, 1985). These parallel pathways are best illustrated by the primate geniculocortical system where parvocellular neurons project to layers IVA and IVCb of area 17, magnocellular layers project to IVCa
(Hubel and Wiesel, 1972; 1977) and intercalated cells project primarily to cytochrome oxidase blobs in lamina III (Usrey et aI, 1992; Lachica and Casagrande, 1992).
Immunocytochemical studies have revealed that the principal cells projecting to lamina
IV express the calcium binding protein parvalbumin whereas calbindin is expressed by 18
the intercalated layers (Jones and Hendry, 1989; Diamond et al., 1993). Each of these parallel channels to area 17 is thought to encode a different aspect of visual perception
(Livingstone and Hubel, 1987). A similar organization has recently been described in the primate somatosensory thalamus where chemically distinct compartments of the VPM nucleus relaying different components of somatosensory information project to separate layers of the somatosensory cortex (Rausell and Jones, 1991a; 1991b). Large to medium parvalbumin-positive cells of cytochrome-oxidase (CO) rich rods project to the middle layers of somatosensory cortex. Calbindin-positive cells in the CO-poor VPM matrix project to lamina I (Rausell and Jones, 1991a; 1991b). These studies demonstrate that functionally distinct compartments of specific thalamic nuclei can be differentiated on the basis of calcium-binding protein expression and by differential projections to separate layers of sensory neocortex (Jones and Hendry, 1989; Rausell and Jones, 1991b). Recent evidence suggests that similar parallel pathways from the MGV to AI also exist
(Hashikawa et aI, 1991). Hashikawa et al. (1991), using the retrograde transport of fluorescent dyes in combination with calcium binding protein immunocytochemistry, have shown that parvalbumin-positive MGV cells project primarily to lamina IIIIIV, whereas calbindin-positive MGV cells project to lamina I.
Parvalbumin (PV) and calbindin (CB) belong to a family of low molecular weight calcium-binding proteins that includes calmodulin, S-100, calretinin, and troponin C
(Andressen et al., 1993). Both PV and CB are widely distributed throughout the central nervous system (Celio, 1990), and are present in all major sensory and motor pathways
(Jones and Hendry, 1989). The precise physiological function of PV and CB, however, 19 is unknown. They may effect calcium-dependent neuronal properties, such as excitability, and neurotransmitter release, by buffering transient increases in intracellular calcium
(Baimbridge etal., 1992; Andressen et aI., 1993; Chard et aI., 1993). PV has been localized in "fast-firing" GABAergic neurons (Kawaguchi et aI., 1987; Kawaguchi and
Kubota, 1993) and in cells with high oxidative metabolism (Braun et aI., 1985a,b). The expression of CB in a wide variety of neuronal subtypes, however, has precluded an obvious correlation with specific neuronal properties (Andressen et aI., 1993). Regardless of their physiological function, the differential and often complementary expression of
PV and CB has been used to delineate distinct nuclei in the central nervous system (Jones and Hendry, 1989; Celio, 1990; Braun et aI., 1991) and for determining patterns of connectivity under normal (DePelipe and Jones, 1991; Hashikawa et aI., 1991) and experimental conditions (Rausell et aI., 1992b; Cellerino et aI., 1992; Bliimcke et aI.,
1994).
Development of auditory neocortex and thalamocortical projections .
In contrast to the visual and somatosensory cortex, almost nothing is known about thalamocortical development in the auditory system, in particular, the relationship between afferent axon innervation and the neocortical cytoarchitectonic differentiation.
In the visual and somatosensory neocortex, a common organizational scheme is the grouping of afferents and cortical processing circuitry to form functional units or modules. This modular organization is exemplified by the barrels of the primary somatosensory cortex of rodents (Woolsey and Van der Loos, 1970), the ocular 20
dominance strips, orientation columns and the color coding blobs of the primary visual
cortex of primates (Hubel and Wiesel, 1969; 1972; Livingstone and Hubel, 1984; T'so
and Gilbert, 1988). While each neocortical area has distinctive morphological and
functional characteristics, they share some common features such as six primary cell
layers, characteristic cell types, interlaminar connection patterns and the laminar origin
of cortical outputs (Jones, 1984). Based on these and other similarities, it has been
suggested that genetic factors specify a common basic structure for all cortical areas
(Rakic, 1976; 1988). Epigenetic factors, such as local cell-cell interactions, afferent
innervation, and efferent connectivity, then operate on the basic framework to determine
a number of area-specific differences (O'Leary and Koester, 1993).
All cortical neurons are generated within the proliferative neuroepithelium
adjacent to the cerebral ventricles (Sidman et aI., 1959). Immature neurons migrate away from the ventricles along radial glia to form cortical layers II-VI below the marginal zone
(future layer I) in an inside-out sequence (Rakic, 1972; 1974). Because migration of cortical neurons from the ventricular surface is largely radial, it has been suggested that a "proto-map" of neocortical areas is present in the ventricular neuroepithelium (Rakic,
1988).
Several lines of evidence indicate that epigenetic factors determine the differentiation of cortical sensory areas from an immature neocortex that is anatomically uniform and functionally pluripotent (O'Leary and Koester, 1993). Experiments altering the modality of thalamocortical afferents to a specific area indicate the existence of a substantial degree of cross-modal cortical plasticity. By misrouting retinal axons to 21 somatosensory or auditory thalamic nuclei at an early age, somatosensory and auditory cortex can be induced to process visual information: neurons in these cortices have response properties to visual stimuli resembling those observed in visual cortical neurons
(Metin and Frost, 1989; Sur et aI., 1988). Experiments involving transplantation of fetal neocortical tissue also suggest a significant degree of cortical plasticity in early development. Fetal visual cortex transplanted into primary motor cortex will establish efferent connections appropriate for the host tissue, suggesting that fmal cortical neuronal connectivity is not predetermined (Stanfield and O'Leary, 1985; O'Leary and Stanfield,
1989). Scblaggar and O'Leary (1991) have recently found that "barrels" form in late fetal visual cortex transplanted to neonatal somatosensory cortex in the rat. These experiments also suggest that differences among neurons arise late in development, and may involve local signals that are region-specific. A source of these signals that is unique to each sensory neocortical area is the thalamocortical afferents that innervate it. Previous developmental studies involving peripheral manipulations that alter afferent input to the visual cortex suggest that thalamic afferent innervation plays a critical role in the development of normal cortical morphology at the cellular level (Coleman and Riesen,
1968; Valverde, 1968; 1976; Le Vay et aI., 1980; Heumann and Rabinowicz, 1982). The primary targets of thalamocortical axons are pyramidal and nonpyramidal cells in lamina
III/IV (White, 1978; Peters et aI.,1979; White and Rock, 1981). In the mouse somatosensory cortex, the cytoarchitectural and dendritic alterations in lamina III/IV that accompany neonatal vibrissae damage (van der Loos and Woolsey, 1973; Ryugo et al.,
1975; Harris and Woolsey, 1981) are secondary to the anomalous organization of 22 thalamocortical afferents (Killackey et aI, 1986; Jensen and Killackey, 1987). Recent studies of auditory neocortex also suggest that thalamocortical afferent innervation influences the dendritic growth and orientation of lamina III/IV neurons during normal development and after unilateral neonatal deafening (McMullen and Glaser, 1988;
McMullen et aI., 1988a;b).
The hypothesis that thalamocortical afferents induce differentiation of sensory neocortical areas and the emergence of powerful anterograde axonal tracing techniques have created a growing interest in their ontogeny. Recent experiments have focused on the development of the visual and somatosensory thalamocortical projections of several species. These studies indicate that thalamocortical afferents reach the cortex long before laminar differentiation occurs. Ghosh and Shatz (1992) have found that visual thalamocortical axons accumulate and extend widespread terminal branches within the visual subplate over a two week period before penetrating the deep layers of the cortical plate. These findings confirm the results of previous studies in cats and monkeys demonstrating the existence of a prolonged "waiting" period during which the generation and migration of layer IV neurons occurs (Rakic, 1976; 1977; Shatz and Luskin, 1986).
In rodents, visual and somatosensory thalamocortical axons also appear in the intermediate zone before layer IV neurons have completed migration into the overlying cortical plate. However, within a day the axons begin to steadily invade each layer as it differentiates from the cortical plate (Lund and Mustari, 1977; Catalano et aI., 1991;
DeCarlos et aI., 1991; Agmon et aI, 1993; 1995). Thus, in all mammalian sensory systems studied to date, thalamocortical afferents are in a position to affect differentiation 23 of cortical plate neurons.
Morphological analysis of thalamocortical afferents suggests that significant interactions among axons and their cortical targets shape internal circuitry. After visual thalamocortical axons have entered layer IV, there is a rapid elaboration of terminal branches within this layer (Naegele et aI., 1988; Florence and Casagrande, 1990; Ghosh and Shatz, 1992). Terminal arbors may grow to twice the width normally found in the adult, indicating considerable remodelling of arbors to attain adult morphology. This remodelling appears to correspond to the sharpening of topographic projections in the cortex (LeVay et aI., 1978; Naegele et aI., 1988), which appears to involve competition among afferents (Shatz and Stryker, 1978; LeVay et aI., 1980; Stryker and Harris,
1986). In the rodent somatosensory system, ventrobasal afferents also display rapid growth upon entering layer IV. However, instead of exhibiting diffuse growth, these axons appear to segregate into discrete clusters as they enter this layer. New branches are then added within their appropriate cluster (Erzurumlu and Jhaveri, 1990; Schlaggar and O'Leary, 1991; Agmon et aI., 1993; 1995). These findings suggest that ventrobasal afferents govern the formation of barrels in primary somatosensory neocortex.
Developmental studies have shown that the adult patterns of thalamocortical patches arise from an initially diffuse projection that is refined during maturation (Ghosh and Shatz, 1992; Senft and Woolsey, 1991; LeVay et aI., 1980; LeVay et aI., 1978;
Rakic, 1977; Rakic, 1976). Although retinal ganglion cell activity is critical for this refinement process in the visual system (Stryker and Harris, 1986; for review see ref.
Goodman and Shatz, 1993), in other centers (e.g. the rodent trigeminal system, Chiaia 24 et aI., 1992; Henderson et aI., 1992) peripherally-driven neuronal activity is not required for central pattern formation indicating that sensory systems differ in their dependence on functional activity.
The rabbit has proven to be a useful model for auditory development. The cytoarchitectural lamination seen in the adult is absent until the end of the first postnatal week (Rice, 1985; McMullen et aI, 1988b). Quantitative analysis of auditory neocortical neurons have shown that dendritic growth of presumptive thalamocortical target neurons in lamina IIIIIV is predominantly a postnatal event. Furthermore, the most dramatic changes in the dendritic growth of lamina III/IV nonpyramidal neurons correlate well with the onset of functional activity within the auditory pathway (McMullen et aI,
1988b).
Previous developmental studies of the rabbit auditory neocortex suggest a correlation between anatomical and functional maturation. Low frequency sound is represented ventrally, and high frequency sound dorsally in the auditory neocortex of rabbits (McMullen and Glaser, 1982a). Neocortical neurogenesis occurs between embryonic day 15 (E-15) and 25 (gestation is 31 days) and proceeds in a ventral to dorsal gradient (Fernandez and Bravo, 1974). From a functional standpoint, rabbits are born deaf. Evoked cortical responses appear first to low frequency sound at P-5 (Konig and
Marty, 1974). The behavioral onset of hearing occurs at the end of the first postnatal week (P-6.5), and is limited to sound frequencies of less than 1 kHz (Foss and Flottorp,
1974). Hearing at higher frequencies appears over the next few days. During this time
(PD 6-9), there is a reversal of cortical evoked potential surface polarity, and a twofold 25 decrease in cortical response latency indicating a concomitant maturation of auditory thalamocortical afferents (Konig and Marty, 1974). Coincident with the onset of hearing, there is a dramatic elaboration of dendrites on lamina III/IV nonpyramidal neurons to levels double that of adults (McMullen et al., 1988b), suggesting that thalamocortical input stimulates dendritic growth either by afferent activity or chemically mediated synaptic contact (White, 1978; Peters et aI., 1979; White and Rock, 1981). The dramatic elaboration of supernumerary dendrites is followed by a period of significant dendritic loss, which may be correlated with remodeling of afferent axon arbors.
The goal of the present study is to characterize the organization and development of thalamocortical pathways that contribute to the functional architecture in the primary auditory cortex of the rabbit. The rabbit was selected as an experimental model because its altricial status in terms of hearing and degree of cortical development at birth, and its size make it an excellent preparation for developmental studies involving iontophoretic injections of neuroanatomical tracers. Furthermore, rabbits have a lissencephalic cortex, and few auditory cortical fields, which help simplify anatomical studies of the organization of auditory thalamocortical pathways. Previous electrophysiological mapping studies have shown that the primary auditory cortex of the rabbit is tonotopically organized with neurons responsive to high frequency sound located dorsally and neurons responsive to low frequencies located ventrally (McMullen and Glaser, 1982a). 26
Specific Aims
The present work has four specific aims:
1. To determine the expression of the calcium-binding proteins parvalbumin and
calbindin within the subdivisions of the rabbit MGB.
2. To determine the organization of thalamocortical projections from the MGV to AI in
adult rabbits.
3. To determine if multiple biochemically distinct pathways connect the MGV and
primary auditory cortex.
4. To examine the early postnatal ingrowth of thalamocortical axons into auditory cortex.
Aim 1 is an essential prerequisite for the other studies because very little is known
about the anatomical and functional organization of the rabbit MGB. Because calbindin
and parvalbumin have been found to delineate functionally distinct compartments of thalamic sensory relay nuclei, immunohistochemical methods will be used to examine the expression of these proteins within the MGB. A differential distribution of these proteins will provide indirect evidence for functional differences between nuclei, and possible associations with the multiple ascending parallel auditory pathways.
Aims 2-4 more directly address the primary goal of the present study which is to characterize the organization and development of thalamocortical pathways that contribute to the functional architecture in the primary auditory cortex of the rabbit. For Aim 2, sensitive neuroanatomical tracers will be used to examine the topographical organization 27 of the thalamocortical axons originating in the MGV and tenninating in AI in adults.
Neuroanatomical tracers will then be used in combination with parvalbumin immunohistochemistry to provide initial biochemical evidence for more than one class of MGV projection neuron (Aim 3). Finally, in Aim 4 the morphological development of MGV axons tenninating in auditory cortex will be examined in early postnatal animals to gain insight into mechanisms underlying the fonnation of auditory thalamocortical circuitry. 28
CHAPTER TWO
COMPLEMENTARY EXPRESSION OF PARVALBUMIN AND CALBINDIN D-28k
DELINEATES SUBDIVISIONS OF THE RABBIT MEDIAL GENICULATE BODY
ABSTRACT
Previous cytoarchitectonic studies of the medial geniculate body (MOB) recognize
three major subdivisions: the ventral, dorsal, and medial divisions. Although additional
parcelling of the MOB is warranted based on Oolgi and connectional studies, further
subdivisions are frequently difficult to discern with routine Nissl stains. We report that
immunohistochemistry for the calcium-binding proteins parvalbumin (PV) and calbindin
D-28k (CB) clearly delineate subdivisions of the rabbit medial geniculate complex. PV
and CB immunohistochemistry display a remarkable complementary staining pattern in the MOB: PV-immunoreactivity predominates in the ventral and medial divisions, whereas CB immunoreactive cells characterize the dorsal and internal divisions. The ventral nucleus is strongly PV-immunoreactive due to dense neuropil labeling and moderately labeled PV + somata. Although CB-immunoreactivity is distinctly lacking in the neuropil of the ventral nucleus, islands of CB + somata are present, especially in the posterior one-third of the nucleus. PV-immunohistochemistry also reveals a small medial 29
nucleus characterized by medium and large PV + somata, and thick PV + axons and puncta. Numerous PV + fibers in the brachium of the inferior colliculus can be traced
into the ventral and medial nucleus. The large wedge-shaped internal nucleus, composed of densely labeled CB+ cells, separates the dorsal and ventral nuclei along its anterior extent, and expands posteriorly to encapsulate the caudal pole of the MGV. The dorsal nucleus is characterized by large, multipolar CB+ neurons with radiate dendrites. Nearly the entire MGB complex is encapsulated by a fibrous marginal zone containing a few
CB + cells. In addition, the differential expression of PV and CB distinguishes subnuclei that are difficult to discern in Nissl stained sections, including the deep dorsal and superficial dorsal nuclei in the dorsal division and a ventrolateral component in the ventral division. The validity of these chemoarchitectonic subdivisions is further supported by their close correspondence to MGB nuclei recently defmed by AChE and
NADH-diaphorase activity in the rabbit (Caballero-BIeda et aI., 1991). Although studies of MGB connectivity in the rabbit are few, a comparison with studies in other species suggests that PV expression is highest in subdivisions that receive a substantial input from the central nucleus of the inferior colliculus and that project to primary auditory cortex. In contrast, CB expression characterizes nuclei that receive input primarily from other sources, such as the paracentral nuclei of the inferior colliculus, the lateral tegmentum and the spinal cord, and that project to secondary auditory areas. The ability of calcium-binding protein immunohistochemistry to delineate distinct neuronal compartments across indistinct cytoarchitectonic borders makes it a powerful tool for guiding future connectional and physiological studies of the MGB. 30
INTRODUCTION
The mammalian medial geniculate body (MGB) is a complex of nuclei subserving
auditory function in the dorsal thalamus. Regional parcellations of the MGB in a variety
of species generally recognize three main subdivisions: the ventral, dorsal, and medial
divisions (for review see Jones, 1985; Winer, 1991). A tripartite organization of the
MGB was proposed by Ramon y Cajal (1966). His account of the MGB in the cat,
rabbit, guinea pig, and dog includes a superior lobe (dorsal nucleus) composed of widely
separated neurons, an inferior lobe (ventral nucleus) with more closely packed cells, and
an internal (medial) nucleus containing magnocellular neurons. Ramon y Cajal (1966)
also identified finer subdivisions in the dorsal (superficial and deep tiers) and ventral
nuclei (ovoidal, and marginal regions). The Golgi studies of the cat MGB by Morest
(1964; 1965a,b) confirmed Cajal's observations, and further characterized dorsal, superficial dorsal, deep dorsal, and suprageniculate nuclei in the dorsal division, as well as lateral, ovoidal and ventrolateral nuclei in the ventral division. Additional support for these subdivisions has derived from physiological and connectional studies (Andersen et al., 1980a,b; Calford, 1983; Calford and Aitkin, 1983). The terminology proposed by
Morest (1964) to identify subdivisions of the MGB has been extended to a variety of species, such as the rat (LeDoux, 1987; Winer and Larue, 1987; Clerici and Coleman,
1990; Clerici et al., 1990), tree shrew (Oliver and Hall, 1975; 1982), opossum (Morest and Winer, 1986; Winer et al., 1988), mustached bat (Winer and Wenstrup, 1994a,b), and human (Winer, 1984). 31
Despite similarities in nomenclature, there is considerable disagreement regarding
MGB subdivisions and their boundaries, and homology between MGB subdivisions in different species is often unclear (see Oliver and Hall, 1978a; LeDoux, 1985; 1987;
Morest and Winer, 1986; Winer, 1992). A primary reason for the difficulty in comparing studies of MGB organization is that the degree of cytoarchitectonic differentiation of the dorsal thalamus in Nissl stained sections varies considerably among species. For example, clearly discernible subnuclei in the dorsal nucleus of the tree shrew (Oliver and
Hall, 1975; 1978a) are not sharply differentiated in the cat (Andersen et al., 1980a;
Calford, 1983), or rat (LeDoux, 1987; Clerici and Coleman, 1990). In addition, species differences in the configuration of the diencephalon have led to further variations in the number and nomenclature of the fmer subdivisions of the MGB (Jones, 1985).
Regional parcellations of the rabbit MGB also recognize dorsal, ventral, and internal (medial) subdivisions (Rose, 1942; Tarlov and Moore, 1966; Giolli and
Gutherie, 1971; Geeraedts, 1978; Holstege and Collewijn, 1982; Jones, 1985). Although the ventral and dorsal nuclei in these studies correspond well to Cajal' s inferior and superior lobes, descriptions of the internal nucleus do not include a distinct magnocellular component (Rose, 1942; Tarlov and Moore, 1966).
Recently, chemoarchitectonic criteria have been used to characterize the subdivisions of the medial geniculate complex in several species (Jones and Hendry,
1989; Caballero-BIeda et al., 1991a,b; Hashikawa et al., 1991, 1994; Campbell and
Pandya, 1994; McMullen et aI., 1994a). In the rabbit, Caballero-BIeda et al. (1991a,b) have shown that the differential distribution of acetylcholinesterase and NADH- 32
diaphorase activities clearly distinguish MGB subdivisions. We have recently found that
parvalbumin immunohistochemistry delineates the major subdivisions of the rabbit MGB,
as well as primary auditory cortex (McMullen et al., 1994a). Differential expression of
the calcium-binding proteins parvalbumin, calretinin, and calbindin-D28k in the monkey
MGB also allows subdivisions to be immunohistochemically identified (Jones and
Hendry, 1989; Hashikawa et al., 1991, 1994; Campbell and Pandya, 1994).
Parvalbumin (PV) and calbindin (CB) belong to the "E-F hand" family of
calcium-binding proteins that includes calmodulin, 8-100, calretinin, and troponin C
(Andressen et al., 1993). Although the precise physiological function of PV and CB is
unknown, they may contribute to calcium homeostasis by buffering transient increases
in intracellular calcium, thereby effecting calcium-dependent neuronal properties such as excitability, neurotransmitter release and resistance to excitotoxic cell death (Baimbridge et al., 1992; Andressen et al., 1993). The ability of PV and CB to buffer depolarization induced changes in intracellular calcium has been recently demonstrated in rat dorsal root ganglion neurons (Chard et al., 1993). Immunohistochemical studies of the neuroanatomical distribution of these proteins have attempted to elucidate their intracellular function by correlating their expression with the physiological properties of cells that contain them. Although PV has been localized in "fast-firing" GABAergic neurons (Kawaguchi et al., 1987; Kawaguchi and Kubota, 1993) and in cells with high oxidative metabolism (Braun et al., 1985a,b), the expression of CB in a wide variety of neuronal subtypes has precluded an obvious correlation with specific neuronal properties
(Andressen et al., 1993). Rather than being essential for basic properties of neurons, it 33
has been suggested that CB may have a modulatory role in controlling neuronal activities
(Baimbridge et aI., 1992). Regardless of their physiological function, the differential and
often complementary expression of PV and CB delineates distinct nuclei in the central
nervous system (Jones and Hendry, 1989; Celio, 1990; Braun et aI., 1991) and serves
to distinguish subsystems processing different types of information (Rausell and Jones,
1991a,b; Hashikawa et aI., 1991; Bennett-Clarke et aI., 1992; Rausell et aI., 1992).
In the present study, we have used parvalbumin and calbindin
immunohistochemistry to delineate the subdivisions of the medial geniculate complex in the rabbit. The complementary staining patterns of these calcium-binding proteins allow the boundaries of the main MGB subdivisions to be more clearly delimited than with
Nissl staining alone. Moreover, these immunohistochemical techniques differentiate finer subdivisions of the dorsal and ventral nuclei, as well as a magnocellular nucleus, not recognized in earlier cytoarchitectonic studies of the rabbit MGB (Rose, 1942; Tarlov and Moore, 1966; Geeraedts, 1978; Holstege and Collewijn, 1982). We have sought to provide a chemoarchitectonic atlas of the MGB that will aid further comparative and connectional studies of these subdivisions. A preliminary report of the results has appeared in abstract form (McMullen et aI., 1994b). 34
METHODS
A total of ten adult (3-4 kg) New Zealand white rabbits were obtained from
commercial suppliers. All animals received terminal anesthetic doses of ketamine (50
mg/kg, Lm.) and sodium pentobarbital (50 mg/kg, Lv.) before intracardial perfusion with
0.1 M sodium phosphate buffered saline (PBS, Ph 7.4) followed by 4%
paraformaldehyde in PBS. Brains were immediately removed from the skull and postfixed
overnight in the same fixative. Thalamic blocks containing the MGB were carefully
dissected from the overlying cerebrum and the lower brainstem, hemisected, and placed
in ascending sucrose solutions (up to 30 %) for cryoprotection. Serial frozen sections (40
",m thick) through the MGB were cut in either coronal or sagittal planes, and collected
in PBS. Every third section was stained with 0.1 % methylene blue. The adjacent sections were processed by indirect immunoperoxidase histochemistry to examine the distribution of either parvalbumin (PV), or calbindin D-28k (CB).
Both the anti-PV (Swiss Antibodies #235) and anti-CB (Swiss Antibodies #300) monoclonal antibodies used in these experiments have been well characterized. The former reacts specifically with parvalbumin in tissue originating from human, monkey, rat, mouse and fish (Celio et aI, 1988), and the latter with calbindin D-28k originating from human, monkey, rabbit, rat, mouse, and chicken tissue (Celio et aI., 1990).
Moreover, neither antibody cross-reacts with the other nor with other known calcium binding proteins as evaluated by radioimmunoassay and immunoenzymatic labeling of immunoblots (Celio et aI., 1988, 1990). 35
All immunohistochemical procedures were performed on free-floating sections
using a rocker table for gentle agitation. Dilutions and rinses were done with 0.1 M PBS,
pH 7.4. Sections were fIrst submerged in 1 % HzOz for 15 minutes to suppress
endogenous peroxidase activity, then placed into 3% normal horse serum (NHS; Vector
Laboratories) with 1 % Triton X-lOO for 60 minutes to block nonspecifIc labeling and to
increase antibody penetration. The sections were incubated in primary mouse monoclonal
antibody (1:5000-1:15,000 dilution with 3% NHS) recognizing either PV or CB for 72
hours at 4°C, followed by biotinylated horse anti-mouse IgG (1 :200 dilution with 3 %
NHS) for 2 hours, and avidin-biotin-peroxidase complex (Vector Standard ABC kit) for
90 min. at room temperature. Antibody labeling of PV and CB was visualized using the
cobalt-nickel DAB intensifIcation method of Adams (1981). All sections were mounted
on gelatin-coated slides, dehydrated in ascending alcohols, cleared in xylene, and coverslipped with Permount for light microscopic examination. No specifIc staining was observed in control experiments in which sections were incubated in primary antibody preadsorbed with either HPLC-purifIed rat parvalbumin (Swiss antibodies) or 28 kD calbindin (Swiss antibodies).
Photographs were taken with a Zeiss Ultraphot photomicroscope equipped with
Luminar objectives (40, 63 and 100 nun lenses) using TMAX-lOO (4 X 5 inch) sheet fIlm. Outline diagrams of the MOB subdivisions as revealed by the differential expression of PV and CB were drawn on transparent plastic sheets placed over the photographs. The borders between subdivisions were confIrmed by comparing the Nissl stained sections with the adjacent sections stained immunohistochemically for PV or CB. A total of 5 36
frontal levels and 1 sagittal level were selected to demonstrate the major features of the
MGB.
RESULTS
The medial geniculate body in the rabbit is located in the most posterior region
of the dorsal thalamus. The posterior half of the MGB protrudes laterally to form a
prominent bulge, the medial geniculate tubercle, on the ventrolateral surface of the
diencephalon. A small depression in the tubercle approximates the border between the
ventral and dorsal divisions (Fig. 1). The anterior half of the MGB lies internal to the
lateral geniculate nucleus, and fibers of the superior thalamic radiations separate the
anterior pole from the ventrobasal complex. The MGB is bounded medially by the anterior pretectal nucleus, while its lateral and dorsolateral margins are surrounded by fibers of the optic tract and the brachium of the superior colliculus (Fig. 2).
Parvalbumin- and calbindin-immunohistochemistry in conjunction with Nissl cytoarchitectonics consistently delineate four main subdivisions in the rabbit MGB: the ventral, dorsal, internal and medial nuclei (Fig. 2). In addition, differential immunohistochemical labeling also clearly reveals a ventrolateral region in the ventral nucleus, as well as superficial and deep dorsal components in the dorsal nucleus. Nearly the entire MGB is enveloped along its dorsal, lateral, and ventral edges by a fibrous marginal zone that contains sparsely distributed CB + cells whose dendrites are oriented parallel to the outer margins of the MGB complex. The typical ovoid appearance of the 37
large ventral and dorsal subdivisions is evident in both coronal sections (Figs. 2-7), and
sagittal sections (Fig. 8).
Ventral nucleus. The prominent ventral nucleus (MGV) extends throughout the anterior
two-thirds of the MGB complex. In Nissl stained sections, MGV is readily distinguished
by the high packing density of its small- and medium-sized cells that are arranged in
parallel laminae whose orientation is dorsomedial to ventrolateral (Figs. 4, 5). In sections
stained by PV-immunohistochemistry, the MGV stands out as an intensely PV
immunoreactive region composed of PV + somata, fibers and terminals. Because the
MGV neuropil lacks significant CB-immunoreactivity, in contrast to surrounding
structures such as the internal nucleus (dorsally and caudally), and the marginal zone
(ventrally and laterally), CB-immunohistochemistry provides a remarkable "negative
image" of the MGV (Figs 4, 8). Many of the cells in MGV (ca. 50%) are PV +, and the
neuropil contains a high concentration of PV + terminal boutons and fibers. Although
both large and small PV + terminals are found throughout the nucleus, large PV +
boutons are especially concentrated within an ovoidal region in the dorsal part of the
nucleus (Fig. 5). Thick and thin PV + axons are also distributed throughout the MGV.
Many of the thicker axons can be traced dorsomedially out of the nucleus into the superior thalamic radiations. At the anterior pole of the MGV, obliquely oriented axon fascicles partially disrupt its oval profile in coronal sections (Fig. 3). Although PV + neurons predominate in the MGV, islands of CB + neurons are present, especially in the posterior one-third of the nucleus (Figs. 5, 6). At the caudal pole of MGV, there is a 38
transition zone where CB + cells of the internal nucleus encroach, making the MGV
borders more difficult to delimit in coronal sections (Fig. 6).
Ventrolateral to the strongly labeled PV + portion of the MGV, there is a crescent
shaped region, the ventrolateral subnucleus (VL) , containing a high proportion of
moderately labeled CB+ cells and CB+ terminal-like labeling (Figs. 4, 5). Unlike the
main body of the MGV, there are relatively few PV + cells in VL. Scattered PV + fibers
in the VL neuropil help to delimit its ventrolateral border since its CB-immunoreactive
neurons merge with the spindle-shaped CB+ cells in the adjacent marginal zone.
Although difficult to discern in Nissl stained sections, VL can be distinguished from
MGV by the slightly lower packing density of its cells that tend to be oriented parallel
to the ventrolateral edge of the MGV.
Dorsal Nucleus. The large, ovoid dorsal nucleus (MGD) occupies approximately the same volume as the MGV and lies superior to this nucleus throughout the anterior two thirds of the MGB. The MGD is bounded anterolaterally by the lateral geniculate nucleus, and posteriorly by the suprageniculate pretectal nucleus (Fig. 8). In Nissl stained sections, MGD is clearly distinguishable from MGV by the larger size and lower packing density of its cells. CB-immunohistochemistry labels large, widely spaced multipolar neurons with radiate dendritic fields throughout the MGD. Moderately stained CB+ fibers and terminal-like labeling are present in the MGD neuropil. In contrast to the
MGV, the MGD contains only a scattered population of PV + neurons within a loose meshwork of thick and thin PV + fibers. 39
The differential expression of PV and CB selectively delineates deep dorsal and
superficial dorsal subnuclei within the dorsal division. Situated at the ventromedial border
of MOD, the deep dorsal nucleus (DD) is difficult to distinguish from the main body of the MOD in Nissl stained sections. In PV-immunohistochemically stained sections, however, this region is demarcated by a high concentration of PV + fibers and terminal like labeling in its neuropil (Fig. 4). In contrast to the sparsely distributed PV + fibers in the main body of MOD, the PV + fibers in DD have a distinct dorsomedial to ventrolateral orientation. CB-immunohistochemistry also reveals a small population of closely spaced CB+ neurons with a similar orientation in DD. The superficial dorsal nucleus (DS) is a small ovoid region in the dorsomedial sector of MOD that contains intensely labeled CB+ stellate neurons, and a dense concentration of CB+ fibers and terminal-like labeling (Fig. 5). The DS can be distinguished from the main body of the
MOD by the higher density of CB+ stellate neurons and the dorsoventral orientation of the CB+ dendrites and axons within the nucleus. The distinct lack of PV + neurons and the absence of PV + terminal-like labeling in its neuropil serve to further differentiate DS from the MOD.
Internal Nucleus. The wedge-shaped internal nucleus (MOl) stands out as an intensely
CB + region separating the MOV from the MOD along its anterior-posterior extent (Figs.
4, 5). The majority of MOl neurons are CB-immunoreactive and the MOl neuropil is filled with CB+ fibers and terminal-like labeling. Sagittal sections reveal that the MOl enlarges posteriorly to encapsulate the caudal pole of the MOV and forms the caudal tip 40
of the medial geniculate complex (Fig. 8). The ventromedial edge of MGI borders the
medial nucleus of the MGB except at the posterior pole of the complex where the medial
nucleus is replaced by the PV + fibers of the brachium of the inferior colliculus. (Figs.
4-7). In Nissl stained sections, the packing density of the medium-sized neurons in MGI
is intermediate to that of the MGV and MGD.
In the anterior one-half of the MGB complex, the dorsomedial edge of MGI abuts
another intensely CB-immunoreactive nucleus, the suprageniculate nucleus (SG). SG is
distinguishable from MGI by the higher packing density of its CB + neurons which form
a column separating MGD from the anterior pretectal nucleus. Furthermore, the SG
neuropil has a highly distinctive reticular appearance in CB-immunohistochemically
stained sections because the dense concentration of terminal-like labeling in the neuropil
is punctuated by the circular profiles of large unstained axon fascicles (Fig. 4). Sagittal
sections reveal that the axon fascicles traverse the SG with an anterodorsal to
posteroventral orientation.
Medial Nucleus. The medial nucleus (MGM) is situated medial to MGV and ventral to
MGI, and extends nearly the full anterior-posterior length of the medial geniculate complex. Anteriorly, the MGM has a narrow columnar shape which expands in the posterior one-third of the MGB (Figs. 3-6). PV-immunohistochemistry reveals a mixture of intensely labeled large multipolar cells and moderately labeled small to medium-sized cells. The large multipolar cells are widely dispersed throughout the nucleus but are more abundant in the posterior half where they are surrounded by dense fascicles of PV + 41
axons from the brachium of the inferior colliculus (Figs. 5, 6). CB
immunohistochemistry reveals a sparse distribution of small to medium-sized CB+ cells,
and few thin fibers in the MGM. A small region immediately dorsal to the MGM lacks
PV-immunoreactivity. In contrast, this region contains both CB+ cells and neuropil
labeling, but at a reduced level relative to the main body of the MGI (Figs. 4, 5).
Although this region has been included in the MGI, it may represent a separate CB+
portion of the MGM.
DISCUSSION
Although the medial geniculate body is primarily regarded as the thalamic relay
for the tonotopically organized ascending auditory pathway to the cerebral cortex, its
roles are multiple and diverse. It is composed of several anatomically and physiologically
distinct nuclei that are believed to form the basis for functionally distinct parallel ascending systems to the auditory cortex (for review see Jones, 1985; Winer, 1991). In the present study, four major subdivisions, the ventral, dorsal, internal, and medial
(magnocellular) nuclei, are distinguished by the differential distribution of PV- and CB immunoreactivity. Furthermore, the complementary staining of MGB nuclei provided by
PV- and CB-immunohistochemistry also delineates a ventrolateral component in the ventral nucleus, and superficial and deep dorsal regions of the dorsal nucleus. The ventral, dorsal, and internal nuclei correspond well with previous tripartite organizational schemes of the rabbit MGB (Rose, 1942; Giolli and Gutherie, 1971; Geeraedts, 1978). 42
The magnocellular medial nucleus is not usually described as a separate subdivision of
the rabbit MGB, however, a similar nucleus was described in the chemoarchitectonic
study of Caballero-BIeda et al. (1991a).
The distribution of PV- and CB-immunoreactive neurons and fibers in the rabbit
MGB is summarized in Figure 9. A comparison with previous studies of MGB
connectivity suggests that the differences in calcium binding protein expression in MGB
subdivisions parallel differences in their connectivity with the brainstem and cerebral
cortex (see below). In general, PV-immunoreactive labeling of cells and neuropil is most
intense in those regions (Le. the ventral and medial nuclei) that receive a major projection from the central nucleus of the inferior colliculus and that project to primary auditory cortex. In contrast, CB-immunoreactivity is highest in regions (Le. the dorsal and internal nuclei) that receive projections primarily from other sources, such as the paracentral nuclei of the inferior colliculus, the lateral tegmentum and the spinal cord, and that project to secondary auditory cortical areas. In the following discussion, our observations will be correlated with previous knowledge of the MGB in the rabbit, and with similar subdivisions in mammals that have been more extensively studied.
Ventral nucleus. The MGV contains the highest density of PV-immunoreactive cells and fibers in the rabbit medial geniculate complex. In Nissl stained sections the neurons in the MGV are densely packed and arranged in parallel laminae. A similar laminar organization is characteristic of the ventral nucleus in other species (Ramon y Cajal,
1966; Morest, 1964; 1965a; Oliver and Hall, 1978a; Oliver, 1982; Clerici and Coleman, 43
1990; Winer and Wenstrup, 1994a,b), and presumably represents the anatomical substrate for isofrequency domains within the nucleus (Morest, 1965a; Calford, 1983; Imig and
Morel, 1985; Rodrigues-Dagaeff et aI., 1989). In all mammals studied, the ventral division of the MGB is considered to be the principal thalamic nucleus in the tonotopically organized lemniscal pathway conveying acoustic information from the cochlea to primary auditory neocortex (cat: Kudo and Niimi, 1978, 1980; Andersen et aI., 1980a,b; Calford and Aitkin, 1983; Rouiller and de Ribaupierre, 1985; Imig and
Morel, 1984; rat: Winer and Larue, 1987; Scheel, 1988; Roger and Arnault, 1989;
Romanski and LeDoux, 1993; tree shrew: Casseday et al., 1976; Oliver and Hall,
1978a,b; primates: Moore and Goldberg, 1966; Fitzpatrick and Imig, 1978; Hashikawa et al., 1991). A limited number of connectional studies indicate that the ventral nucleus of the MGB in the rabbit, like other species, receives a large projection from the central nucleus of the inferior colliculus (Tarlov and Moore, 1966), and projects to primary auditory cortex (McMullen and de Venecia, 1993; de Venecia and McMullen, 1994).
The selective expression of calcium binding proteins also suggests homology between the rabbit MGV and the ventral nucleus of other mammals. Similar to the rabbit
MGV, PV -immunoreactivity characterizes the ventral division of the MGB in the monkey
(Jones and Hendry, 1989; Hashikawa et al., 1991; 1994; Campbell and Pandya, 1994).
Using retrograde neuroanatomical tracing in combination with PV immunohistochemistry ,
Hashikawa et al. (1991) have provided evidence indicating that PV + neurons in the
MGV project to midcortical layers of primary auditory cortex. Recent studies in our laboratory have shown that focal injections of sensitive anterograde tracers in the MGV 44 label axons that terminate primarily in layers III/IV in primary auditory cortex
(McMullen and de Venecia, 1993; de Venecia and McMullen, 1994). Furthermore, PV immunohistochemistry reveals a zone of PV + terminal-like labeling in layers III/IV of primary auditory cortex that may originate, in part, from MGV neurons (McMullen et aI., 1994a). Preliminary double-labeling studies in our laboratory have found that MGV neurons labeled after retrograde transport of tracers injected into auditory cortex also express PV (de Venecia et aI., unpublished observations).
The intense PV-immunoreactivity that characterizes the MGV is due to the high concentration of PV + fibers and terminal-like labeling in the neuropil. Although the source of these fibers and terminals is yet to be established, it is likely that they originate in part from the central nucleus of the inferior colliculus and auditory neocortex, areas that provide substantial projections to the MGV (Tarlov and Moore, 1966; Andersen et aI., 1980a,b; Roger and Arnault, 1989; Rouiller and deRibaupierre, 1985). In support of a collicular origin of the PV + fibers and terminals is the presence of large fascicles of PV + fibers in the brachium of the inferior colliculus and PV + cells within the central nucleus of the inferior colliculus. Moreover, recent double-labeling studies in our laboratory indicate that PV + pyramidal neurons in layer VI of auditory cortex project to MGV (McMullen et aI., 1994a). An additional source of PV + fibers and terminals may be the reticular complex which has been shown to have reciprocal connections with
MGV in the rat and cat (Montero, 1983; Rouiller and de Ribaupierre, 1985; Rouiller et aI., 1985). Like other species (rat: Seto-Oshima et aI., 1989; Celio, 1990; monkey: Jones and Hendry, 1989), the reticular complex of the rabbit is intensely PV-immunoreactive. 45
CB-immunohistochemistry reveals a "negative-image" of the MGV because of the paucity of immunoreactive labeling in its neuropil. Moderately labeled CB + neurons, however, are present in small numbers relative to the numerous PV + cells in the MGV.
These CB + cells frequently form clusters and appear more prevalent in the posterior one-third of the nucleus. In monkeys, CB+ neurons in the ventral nucleus of the MGB project to layer I of primary auditory cortex and represent a parallel calbindin-positive thalamocortical pathway distinct from the parvalbumin-positive projection to midcortical layers (Hashikawa et aI., 1991). The possibility that CB+ cells in the rabbit MGV relay a separate parallel calbindin-positive projection to AI merits further investigation.
CB- and PV - immunohistochemistry specifically delineate a ventrolateral subregion in the MGV, VL, that contains CB+ cells, fibers and terminal-like labeling, but relatively few PV + cells. The VL is difficult to distinguish in Nissl sttined sections, however, because the size and shape of its cells are similar to those in the main body of the MGV. In addition, the packing density of cells in VL is only slightly lower than that of the MGV. In their study of the differential distribution of AChE and NADH diaphorase activities in the rabbit MGB, Caballero-BIeda et al. (1991) also recognized a crescent shaped ventrolateral portion of MGV characterized by decreased NADH activity and a lower cell density relative to the medially located ovoidal portion the nucleus. These data reveal the potential of chemoarchitectonic methods to provide a more consistent characterization of neuronal compartments in comparison to Nissl stains alone.
In the cat and rat, a ventrolateral subdivision has also been distinguished cytoarchitectonically from the MGV by the lower packing density of its cells (Morest) 46
1964; Winer, 1985; LeDoux et al., 1987). A ventrolateral subnucleus containing oval to
elongate cells has also been described in the MOB of the opossum although the packing
density of the cells is increased relative to MGV (Winer et al., 1988). In the cat, the
ventrolateral nucleus receives little input from the central nucleus of the inferior
colliculus (Calford and Aitkin, 1983). In the cat and rat, the ventrolateral nucleus
projects to secondary auditory cortical fields (Andersen et al., 1980a; Morel and Imig,
1987; Scheel, 1988). The caudomarginal nucleus of the tree shrew may be homologous
to the ventrolateral nucleus of the cat since it is also more or less coextensive with the
ventral nucleus, receives input primarily from paracentral nuclei of the inferior
colliculus, and projects to nonprimary auditory cortex (Oliver and Hall, 1978a,b). In the
cat, VL neurons display broad tuning characteristics and long latency responses that are
not tonotopically organized (Calford, 1983; Calford and Aitkin, 1983; Imig and Morel,
1985; Morel and Imig, 1987). Because of the connectional and physiological properties
of the ventrolateral nucleus in the cat, it has been suggested that this nucleus may be
more related to the dorsal nucleus, and should be considered part of the "diffuse pathway" to field All, in contrast to the "cochleotopic" pathways to fields AI and AAF
(Andersen et al., 1980a; Calford and Aitkin, 1983; Winer, 1985). A similar suggestion has been made for the caudomarginal nucleus in the tree shrew (Oliver, 1982). While data for VL in the rabbit are lacking, the selective expression of calbindin in VL cells may reflect connectional differences relative to the PV + portion of the MGV.
Dorsal nucleus. Previous studies of neuronal morphology in the dorsal division of the 47
MGB have recognized a number of subnuclei, including dorsal, deep dorsal, superficial dorsal and suprageniculate nuclei, (Morest, 1964; Oliver and Hall, 1978a; Oliver, 1982;
Clerici and Coleman, 1991; Winer and Morest, 1983b; Winer, 1985). Connectional studies indicate that each subdivision receives projections from different regions of the inferior colliculus (Morest, 1965b; Andersen et aI., 1980b; Calford and Aitkin, 1983;
Rouiller and de Ribaupierre, 1985; Oliver and Hall, 1978a) and has different thalamocortical and corticothalamic connections (Andersen et aI, 1980; Imig and Morel,
1984; 1985; Oliver and Hall, 1978b; Rouiller and de Ribaupierre, 1985; Roger and
Arnault, 1989; Arnault and Roger, 1990). The neurons in each division also display different auditory response properties (Calford, 1983).
In Nissl stained sections, the rabbit MGD consists of widely spaced medium-sized cells and has no conspicuous laminar organization. Similar cytoarchitectonic features distinguish the dorsal nucleus in other mammals (rat: Clerici and Coleman, 1990; Winer and Larue,1987; LeDoux et aI., 1985; cat: Ramon y Cajal, 1960; Morest, 1964; mustached bat: Winer and Wenstrup, 1994a; tree shrew: Oliver and Hall, 1978; opossum: Winer et aI., 1988; monkey: Burton and Jones, 1976). CB immunohistochemically stained sections reveal a scattered population of immunoreactive neurons whose dendritic morphology is similar to the stellate neurons described in Golgi studies of the MGB in the rat (Clerici et aI., 1991; Winer and Larue, 1987), cat (Ramon y Cajal, 1966; Morest, 1964, 1965b; Winer and Morest, 1983b), mustached bat (Winer and Wenstrup, 1994b), tree shrew (Oliver, 1982), and opossum (Morest and Winer,
1986). Connectional studies in a variety of species indicate that the dorsal nucleus 48
receives substantial projections from diverse sources including the lateral tegmentum, the paracentral nuclei of the inferior colliculus, the superior colliculus, and the spinal cord, but oI'lly a minor contribution from the central nucleus of the inferior colliculus (Morest,
1965b; Rouiller and deRibaupierre, 1985; LeDoux et aI., 1987). This diversity of input is reflected in the physiological characteristics of dorsal nucleus neurons: they are typically broadly tuned to stimulus frequency, they lack a precise tonotopic arrangement, and they have long response latencies (Calford, 1983; Imig and Morel, 1985; Rouiller and de Ribaupierre, 1985).
Although difficult to distinguish from the main body of the MGD in Nissl stained sections, the differential distribution of PV- and CB-immunoreactivity in MGD clearly identifies two subnuclei, the deep dorsal and superficial dorsal. DD is most readily recognized in sections stained for PV because of the relatively dense concentration of
PV-immunoreactive fibers and terminal-like labeling in its neuropil. The dorsomedial to ventrolateral orientation of the PV + fibers and cells in DD also serves to further differentiate it from the MGD. In contrast to the deep dorsal component of the MGD, the superficial dorsal nucleus lacks PV + cells and terminal-like labeling. DS is characterized by closely spaced CB-immunoreactive stellate neurons whose dendrites are oriented dorsolateral to ventromedial, an orientation that parallels CB + fibers within the nucleus. Topographically, DD is similar in location to the deep dorsal nucleus described in the rabbit (Caballero-BIeda et aI., 1991), cat (Morest, 1964; Winer and Morest, 1983), rat (LeDoux et aI., 1987; Clerici and Coleman, 1990; Clerici et aI, 1990), tree shrew
(Oliver and Hall, 1978), opossum (Winer et aI., 1988), and human (Winer, 1984). DS 49
is similar in location to the superficial dorsal nucleus described in the cat (Morest, 1964;
Winer and Morest, 1983), opossum (Winer et al., 1988), mustached bat (Winer and
Wenstrup, 1994a,b), and human (Winer, 1984).
The deep dorsal nucleus receives inputs from the central nucleus of the inferior
colliculus in the cat (Kudo and Niimi, 1980; Calford and Aitkin, 1983; Rouiller and de
Ribaupierre, 1985) and has reciprocal connections with primary auditory cortex in the
cat (Andersen et al., 1980a) and rat (Arnault and Roger, 1989). Physiological studies
indicate that deep dorsal nucleus neurons in the cat have acoustic response properties that
are intermediate to those of neurons in the main body of the dorsal division and in the
ventral division (Calford, 1983). A distinctive feature of the cat deep dorsal nucleus is that its neurons respond best to high frequency acoustic stimuli (Calford, 1983; Rouiller and de Ribaupierre, 1985). Connectional and physiological studies of the superficial dorsal nucleus are lacking. Previous studies either have not distinguished this region as separate from the dorsal nucleus, or have considered it part of the lateral posterior nucleus (Calford, 1983; Calford and Aitkin, 1983).
Internal nucleus. Although an internal division is not usually described as a distinct component of the mammalian MGB, the differential expression of PV and CB clearly demarcates a wedge shaped internal nucleus (MGI) that separates the MGV from MGD.
MGI stands out as an intensely CB+ region notable for its dense population of CB+ cells and distinct lack of PV + cells. Sagittal sections demonstrate that the MGI occupies a large volume of the MGB, especially posteriorly where it expands to encapsulate the 50 posterior pole of the MOV and forms the caudal tip of the MOB. Caballero-BIeda et al.
(1991) have also described an internal nucleus in the rabbit MOB with a topographic location and cytoarchitecture similar to the MOl of the present study. Their demonstration that the MOl is also biochemically distinct (containing weak AChE and moderate NADH diaphorase activities) from other MOB subdivisions is further support for the existence of MOl as a separate subdivision.
Previous cytoarchitectonic parcellations of the rabbit MOB also recognize a wedge-shaped internal subdivision located between the ventral and dorsal divisions similar to MOl in the present study (Rose, 1942; Tadov and Moore, 1966; Oiolli and
Outhrie, 1971; Oeeraedts, 1978; Holstege and Collewijn, 1982; see also Jones, 1985).
The internal nucleus in these studies is usually assumed to correspond to the medial magnocellular division of other mammals, even though a distinct magnocellular population is not present within the nucleus (Rose, 1942; Tadov and Moore, 1966;
Jones, 1985). Tadov and Moore (1966) found a large projection to the internal nucleus from the inferior colliculus, but the axonal degeneration techniques used in their study did not permit a precise determination of the neurons of origin (central vs. paracentral nuclei). The internal nucleus also receives a small projection from the deep layers of the superior colliculus (Tadov and Moore, 1966; Holstege and Collewijn, 1982). Whether the CB+ fibers in the MOl originate from the inferior or superior colliculi, however, is unknown.
The lack of connectional and physiological data for the rabbit MOl prevents an accurate assessment of homology with MOB subdivisions in other species. The location 51
of the MGI relative to the MGV suggests that portions of the dorsal division would be
the homologue of MGI in other species. In the rabbit, the MGI encapsulates the MGV
posteriorly to form the caudal pole of the MGB. Caudal portions of the dorsal division
similarly form the posterior pole of the MGB in the cat (Morest, 1964; Andersen, 1980a;
Calford, 1983; Winer and Morest, 1983b), rat (Roger and Amault, 1989; Clerici and
Coleman, 1990), tree shrew (Oliver and Hall, 1978a) and monkey (Burton and Jones,
1976; Fitzpatrick and Imig, 1978).
Intense CB + labeling is also present in the cells and neuropil of the
suprageniculate nucleus. The SG adjoins the MGI at its dorsomedial edge and separates
the medial border of the MGD from the anterior pretectal nucleus. SG is distinguishable
from MGI, however, by the columnar arrangement of its cells, and by the presence of
large unstained axonal fascicles. These fascicles interrupt the homogeneous CB+
terminal-like labeling in the neuropil to give the region a reticular appearance in sections
immunohistochemically stained for CB. In many species, SG is usually considered a
component of either the dorsal division of the MGB or of the posterior group of thalamic
nuclei because it receives multimodal sensory inputs from deep layers of the superior
colliculus (Benevento and Fallon, 1975; Casseday et aI., 1976; Graham, 1977; Oliver
and Hall, 1978a; Holstege and Collewijn, 1982), the dorsal midbrain tegmentum (Oliver
and Hall, 1978), the inferior colliculus (leDoux et aI., 1987), and spinal cord (leDoux et aI., 1987; Schroeder and Jane, 1971). In the monkey and cat, SG projects primarily
to insular cortex (Burton and Jones, 1976), with the cat having additional projections to
secondary auditory areas (Heath and Jones, 1971). In the opossum, SG projects 52
extensively to auditory cortex and to neighboring association areas (Neylon and Haight,
1983).
Medial nucleus. The intensely PV + medial division is situated dorsomedial to the MGV
and ventromedial to MGD, and extends nearly the entire length of the MGB. MGM contains predominately large PV + cells, but small PV + cells are also present. PV + fibers and terminals fill the neuropil and contribute to the intense PV-immunoreactivity that characterizes this nucleus. CB-immunohistochemistry also reveals a small population of CB + cells. The topographic location and cytoarchitecture of the MGM is consistent with the classic description of the magnocellular medial nucleus in other species (Ramon y Cajal, 1966; Morest, 1964; Winer and Morest, 1983a; Oliver and Hall, 1975, 1978a;
Oliver, 1982; Clerici and Coleman, 1990). Connectional and physiological studies in other species indicate that the medial (magnocellular) nucleus receives multimodal sensory input (Wepsic, 1966; Love and Scott, 1969; LeDoux et aI., 1987), including a major contribution from the central nucleus of the inferior colliculus (Moore and
Goldberg, 1966; Kudo and Niimi, 1978; 1980; Andersen et aI., 1980b), and has widespread connections with primary and secondary auditory neocortex (Oliver and Hall,
1978a,b; Andersen et aI., 1980a; LeDoux et aI., 1985; Scheel, 1988; Redies et aI., 1989;
Rouiller et al., 1989). Because they observed a weak tonotopic organization in the medial division of the cat MGB, Rouiller et aI. (1989) suggested that this nucleus should be included in the tonotopic system of connections between the auditory thalamus and cerebral cortex as initially defined by Andersen et aI. (1980a). 53
Caballero-BIeda et al.(1991a,b) have recently described a biochemically distinct mediorostral nucleus in the rabbit MOB that has topographic and cytoarchitectural features identical to the MOM in the present study. The mediorostral nucleus has moderate AChE and NADH-diaphorase activities in its neuropil and a distinct population of large stellate cells that are intensely stained by NADH-diaphorase histochemistry.
These large neurons are more abundant at the posterior pole of the nucleus, similar to the graded distribution of the large PV + stellate cells in the MOM of the present study.
A gradient in the distribution of large cells has also been noted in the medial division of the MOB in the rat (Clerici and Coleman, 1990), cat (Winer and Morest,
1983), and monkey (Hashikawa et aI., 1991). The uneven distribution of medial division cell types in the monkey parallels the differential distribution of parvalbumin and calbindin, with smaller CB-immunoreactive cells more prevalent ventromedially and large
PV + cells dorsally (Hashikawa et aI., 1991). This difference in calcium-binding protein expression may also reflect differences in both function and connectivity. In the rat
MOB, Scheel (1980) reported that the caudal one-third of the medial division, where magnocellular neurons are most abundant (Scheel, 1980; Clerici and Coleman, 1990), projects to primary auditory cortex, whereas more rostral areas of the division project to secondary auditory areas. In the tree shrew, the medial division has been divided into mediorostral and mediocaudal nuclei that differ not only in cytoarchitectonic composition
(the latter containing the largest cells) but also in their connections with the brainstem and cerebral cortex (Oliver and Hall, 1978a,b). Although a preferential distribution of large PV + stellate cells in the caudal MOM was noted in the present study, there was 54 no gradient of PV and CB expression within the nucleus: PV-immunoreactivity was characteristic of the nucleus throughout its rostrocaudal extent. There is, however, a small region surrounding the dorsal MGM that contains only moderately labeled CB + cells and neuropil, and lacks PV-immunoreactivity. Whether this region represents a separate CB + component of the MGM or a subregion of the MGI is an issue that may be resolved by future connectional studies of the rabbit MGB.
Comparative studies calcium-binding protein distribution in the auditory thalamus.
In contrast to the extensive number of anatomical and physiological studies of the cat MGB (for review see Winer, 1992), to our knowledge there have been no studies of the distribution of PV and CB in this species. The few detailed studies of the auditory thalamus in other mammals, however, suggest considerable species differences in the distribution of calcium-binding protein expression. The differential expression of PV and
CB within the ventral, dorsal, and medial divisions of the rabbit is similar to that of the monkey MGB (Jones and Hendry, 1989; Hashikawa et aI., 1991; Campbell and Pandya,
1994). In contrast to both the rabbit (present study) and monkey (Jones and Hendry,
1989; Hashikawa et aI., 1991; Campbell and Pandya, 1994), Celio (1990) found no PV immunoreactive cell bodies in the rat MGB. PV + neuropil labeling, however, was highest in the ventral and medial divisions (Celio, 1990). In addition, the strongest CB immunoreactivity was present in the cells and fibers in the dorsal division (Celio, 1990).
Other studies of the rat MGB have also reported a widespread distribution of CB+ neurons in all divisions but only few weakly labeled cells in the ventral division (Sequier 55
et aI., 1990; Puelles et aI., 1992; Friauf, 1994). The detailed study of the rat MOB by
Friauf (1994) reported a pattern of CB-immunoreactive staining similar to the rabbit
MOB with the· strongest CB + labeling of cells and neuropil in the peripeduncular and
lateral posterior nuclei, and intennediate levels of staining in the dorsal and medial divisions, the marginal zone, and the suprageniculate, and posterior intralaminar nuclei.
Interestingly, the calcium binding-protein, calretinin, has a similar widespread distribution in the rat MOB, with a paucity of expression in the ventral division and the ventral part of the dorsal division (Arai et aI., 1991).
In comparison to the rabbit, both PV and CB have a more diffuse distribution in the MOB of the mustached bat (Wenstrup, 1994; Zettel et aI., 1991). Heavily labeled
PV + and CB+ neurons are found widely distributed in the ventral, dorsal, and medial divisions. The mustached bat is, however, similar to the rabbit and monkey in that PV immunohistochemistry labels cells and neuropil most heavily in regions receiving input from the central nucleus of the inferior colliculus (Wenstrup, 1994). In contrast to the widespread distribution of CB+ cells, CB+ staining of the neuropil is heaviest in the dorsal division, particularly the superficial dorsal nucleus, but is low in the ventral and medial divisions and virtually absent from the posterior complex and suprageniculate nucleus (Zettel et aI., 1991). Unlike the rabbit, rat and monkey, the suprageniculate nucleus in the mustached bat is characterized by PV + cells and fibers, and receives a large projection from the central nucleus of the inferior colliculus (Wenstrup, 1994;
Wenstrup et aI., 1994). 56
Functional implications of the differential distribution of calcium-binding proteins.
In recent years, calcium-binding proteins have been increasingly used as specific
neuronal markers (Andressen et al., 1993). The differential distribution of PV and CB
in neuronal subsystems often reflects dissimilar functional and structural characteristics
of neurons, as well as their distinct patterns of connectivity. Both interspecies differences
in the distributions of PV and CB, and the overlapping distribution of these proteins in
some regions of the central nervous system, however, indicate that caution should be
exercised when making generalizations concerning the functional significance of calcium
binding protein expression (for review see Baimbridge et al., 1992; Andressen et al.,
1993). For example, although CB-immunoreactivity selectively labels monaural pathways
in the auditory brainstem of the bat (Zettel et al., 1990) and the chinchilla (Kelley,
1992), this association does not hold true in the rat (Friauf, 1994).
The expression PV or CB has been strongly correlated with the expression of high
and low levels, respectively, of the metabolic enzyme cytochrome oxidase in distinct
neuronal circuits in the auditory and visual systems of the zebra finch (Braun et aI,
1985a,b; 1991), the somatosensory system of monkeys and rodents (Rausell and Jones,
1991a,b; Bennett-Clarke et al., 1992), and the monkey visual system (Celio, et aI, 1986;
Hendry et al., 1989; Van Brederode et al., 1990). Furthermore, in the rodent trigeminal brainstem complex, PV and CB antibodies stain subpopulations of neurons that differ morphologically and have distinct (single-vibrissa and multiple-vibrissae, respectively) receptive field properties (Bennett-Clarke et al., 1992).
Several studies have demonstrated that PV and CB are differentially expressed in 57
functionally distinct thalamocortical pathways in the visual and somatosensory systems
of the monkey, the prosimian galago and the tree shrew (Jones and Hendry, 1989;
Rausell and Jones, 1991; Diamond et aI, 1993). Studies combining neuroanatomical
tracing and calcium-binding protein immunohistochemistry in the auditory system also
provide evidence for separate, chemically distinct parallel projections from the medial
geniculate body to superficial and midcorticallayers of auditory cortex (Hashikawa et al. ,
1991). Hashikawa et al. (1991) have speculated that the "calbindin system" in the
auditory pathways from the midbrain to the cerebral cortex may be more diffusely
organized than the "parvalbumin system" .
The results of the present study indicate that PV is expressed predominantly in
MGB nuclei that receive a substantial projection from the central nucleus of the inferior
colliculus and that project to primary auditory cortex. In contrast, CB-immunoreactivity
identifies nuclei that receive a majority of their input from other sources and that project
to nonprimary auditory cortical fields. Furthermore, the expression of PV in both thalamocortical and corticothalamic projection neurons in the rabbit suggests that PV is expressed in reciprocal circuits connecting the MGV and auditory cortex (McMullen et al., 1994a). The functional significance of the differential expression of these calcium binding proteins, however, is unclear.
Concluding remarks
A comparison of previous anatomical and physiological data on the MGB with the results of calcium-binding protein immunohistochemistry indicates that the MGB of the 58
rabbit, like other mammals, is composed of a "principal" ventral nucleus, a
magnocellular medial nucleus, and a dorsal nucleus composed of several subdivisions.
In addition, we have identified a distinct internal nucleus that is not usually recognized
in studies of MOB organization. The validity of these subdivisions, however, is supported
by their close correspondence to nuclei identified by Caballero-BIeda et al. (1991a,b) on
the basis of the differential distribution of AChE and NADH-diaphorase activity in the
rabbit MOB. As these authors have noted, the presence of four chemoarchitectonically
distinct nuclei does not imply that the organization of the rabbit MOB is more complex than other species, but likely reflects the relative insensitivity of cytoarchitectonic methods for parcelling this thalamic region. Indeed, several researchers have indicated that the inability to clearly identify individual nuclei and their borders in routine Nissl stained sections of the MOB has hampered the advancement of our understanding of
MOB connectivity and function (Oliver and Hall, 1978a; LeDoux et al., 1987; Caballero
BIeda et al., 1991a,b). The ability of calcium binding protein immunohistochemistry to identify distinct neuronal compartments across unclear cytoarchitectonic borders makes it a powerful tool for guiding future connectional studies of the MOB and for resolving many comparative issues concerning the organization of the auditory thalamus. 59
Figure 2.1
Oblique posterior view of an adult rabbit brain. The posterior cerebrum overlying the caudal diencepalon has been removed by dissection midway through primary auditory cortex (AI). The caudal half of the MGB protrudes from the lateral surface of the diencepalon to form the medial geniculate tubercle. A shallow depression in the tubercle approximates the boundary between the dorsal (D) and ventral (V) divisions of the MGB. b = brachium of the inferior colliculus. Scale bar = 5 mm. · GO 61
Figure 2.2
Low magnification view of two serial coronal hemisections of the thalamus midway through the anterior-posterior extent of the MOB. The complementary staining patterns provided by (A) parvalbumin- and (B) calbindin-immunohistochemistry clearly delineate MOB subdivisions and surrounding nuclei. The anterior half of the MOB is bordered laterally by the optic tract (ot) , medially by the anterior pretectal nucleus
(APT), and ventrally by the peripeduncular and posterior intralaminar nuclei (not labeled). A higher magnification view of these sections is shown in Figure 4. Scale bar
= 1 mm. 62
\ \ I, \
~ \ 63
Figure 2.3
A. Outline diagram of the MGB near the anterior pole of the complex drawn from serial coronal sections shown in B-D. B. Methylene blue stain. C. PV immunohistochemical stain. D. CB-immunohistochemical stain. Both ventral (V) and medial (M) nuclei stain intensely for PV. In contrast, both the suprageniculate (SG) and dorsal nucleus (D) are CB-immunoreactive. Axon fascicles containing, in part, PV + fibers can be traced from the ventral nucleus into the thalamic radiations (tr). Scale bar
= 500/Lm. 64 65
Figure 2.4
A. Outline diagram of MOB approximately 1 mm posterior to level shown in Figure 3 drawn from serial coronal sections shown in B-D. B. Methylene blue stain. C. PV immunohistochemical stain. D. CB-immunohistochemical stain. The ventral nucleus is readily identified in Nissl stained sections by the high density of its cells, many of which express PV. The dorsal nucleus is characterized by widely spaced CB immunoreactive stellate cells, and is separated from the ventral nucleus by the intensely
CB + internal nucleus (I). PV + fibers delineate the deep dorsal nucleus (Dd) at the ventromedial edge of the dorsal nucleus. The superficial dorsal nucleus (Ds) can be distinguished by its lack of PV-immunoreactivity and strong CB+ labeling. The medial nucleus contains large PV + cells and fibers, and appears compact relative to more posterior levels. Scale bar = 500/-tm. 56
') \ \
1 67
Figure 2.S
A. Outline diagram of MGB approximately 1.4 nun posterior to level shown in
Figure 3 drawn from serial coronal sections shown in B-D. B. Methylene blue stain.
C. PV-immunohistochemical stain. D. CB-immunohistochemical stain. The medial
nucleus, which enlarges posteriorly, contains a distinct population of large PV + cells
widely dispersed among PV + fibers. In the ventral nucleus, CB-immunohistochemistry reveals islands of CB + cells and a moderately labeled, crescent shaped region, the ventrolateral nucleus (VL), at its ventrolateral margin. A dorsal region in the ventral nucleus shows increased PV-immunoreactivity due to a high concentration of large PV + puncta and fibers. Scale bar = 500 /-tm. 68 69
Figure 2.6
A. Outline diagram of MGB approximately 2.2 mm posterior to the level shown
in Figure 3 drawn from serial coronal sections shown in B-D. B. Methylene blue stain.
C. PV- immunohistochemical stain. D. CB-immunohistochemical stain. At the posterior pole of the ventral nucleus, there is a transition zone where CB + cells of the internal nucleus encroach making the borders between the nuclei more difficult to delimit. In addition, islands of CB + cells are more prevalent posteriorly in the ventral nucleus. Large PV + cells in the medial nucleus are surrounded by PV + fibers from the brachium of the inferior colliculus. Scale bar = 500 I'm. 70 71
Figure 2.7
A. Outline diagram of the posterior pole of the MOB (approximately 2.9 mm posterior to the level shown in Figure 3) drawn from serial coronal sections shown in B
D. B. Methylene blue stain. C. PV-immunohistochemical stain. D. CB immunohistochemical stain. The posterior pole of the MOB is formed primarily by the
CB + internal nucleus. The internal nucleus is bordered medially by the PV + brachium of the inferior colliculus. The suprageniculate pretectal nucleus (SOP) demarcates the superior border of the MOB. Scale bar = 500 p.m. MT 73
Figure 2.8
A. Outline diagram of MGB approximately 1.7 mm from its lateral surface drawn from sagittal sections shown in B-D. B. Methylene blue stain. C. PV immunohistochemical stain. D. CB immunohistochemical stain. The CB + internal nucleus (I) encapsulates the PV + ventral nucleus posteriorly, and fonns the caudal pole of the MGB. The suprageniculate pretectal nucleus (SGP) demarcates the caudodorsal border of the MGB. Scale bar = 500 JLm. 74 75
Figure 2.9
Summary diagram of the distribution of PV and CB immunoreactive somata (A) and fibers (B) in the MOB of the rabbit. 76
Somata Fibers MZ u - ff ., , Os If '\ "' ~ ~ -, IFsITA D I / d\ ( / ~ ( ..... l..... I ~ T7 f': r7 (~~ ~ 1'-.... "" r...'J 1\ ~
~ :::i!~~ VL PV IIII~ ...... CB o CB 77
TABLE 2.1
ABBREVIATIONS
AI primary auditory cortex
APT anterior pretectal nucleus
bic brachium of the inferior colliculus
bsc brachium of the superior colliculus
CB calbindin
D dorsal division of MOB
Ds superficial dorsal division of MOB
Dd deep dorsal division of MOB
I internal division of MOB
IC inferior colliculus
LON lateral geniculate nucleus
LP lateral posterior complex
M medial division of MOB
MOB medial geniculate body
MOD dorsal division of MOB
MOl internal division of MOB
MOM medial division of MOB
MOV ventral division of MOB 78
MT midbrain tegmentum
MZ marginal zone
OT nucleus of the optic tract ot optic tract
PIN posterior intralaminar nucleus
PP peripeduncular nucleus
PV parvalbumin
SC superior colliculus
SO suprageniculate nucleus
SOP suprageniculate pretectal nucleus tr thalamic radiations
V ventral division of MOB
VPM ventral posteromedial nucleus
VL ventral lateral division of MOB
ZI zona incerta 79
CHAPTER THREE
THALAMOCORTICAL PATCHES IN AUDITORY NEOCORTEX
ABSTRACT
Thalamocortical afferents to the primary auditory cortex of the rabbit were labeled by the iontophoretic injection of the anterograde tracers PHA-L or biocytin into the ventral division of the medial geniculate body (MGV). Single injections of either tracer into the MGV labeled multiple "patches" of afferent axons in lamina III/IV of the ipsilateral auditory cortex. Serial section analysis revealed that single patches were elongated in the rostral-caudal axis forming bands approximately 2.0 mm in length. The orientation of the bands was similar to the isofrequency contours of the tonotopic maps derived from prior electrophysiological experiments. Within the coronal plane, the topography of the patches is remarkably similar to the intermittent distribution of binaural interaction subclasses described in physiological studies. Our results are consistent with a model of MGV organization containing functionally-distinct, parallel anatomical pathways to AI. 80
INTRODUCTION
The auditory neocortex of mammals contains maps of sound frequency and
binaural interactions (Brugge and Reale, 1985; Aitkin, 1990; Phillips et al., 1991; Clarey
et al., 1992). The thalamocortical circuits that subserve these maps have been inferred
from functional (Imig and Morel, 1983; 1988), degeneration (Mesulam and Pandya,
1973; Sousa-Pinto, 1973; Frost and Caviness, 1980; Vaughan, 1983), anterograde (Jones
and Burton, 1976; LeDoux et al., 1985) and retrograde tracing experiments (Winer et al., 1977; Imig and Morel, 1983; Middlebrooks and Zook, 1983; Morel and Imig, 1987).
Numerous studies conducted on the cat thalamocortical auditory system have resulted in a "divergence/convergence" theory of geniculocortical organization (Merzenich et aI.,
1982; 1984). For example, injections of retrograde tracers into physiologically-defined frequency loci have shown that restricted areas of the ventral division of the MGB project to divergent cortical sites along an isofrequency contour (Merzenich et al., 1982). These results suggest that the cortical tonotopic map derives from successive sheet-like arborizations of thalamocortical axons along an isofrequency contour. A second functional map within auditory cortex consists of alternating patches of binaural interaction columns in which neurons driven by stimulation of the contralateral ear are either facilitated (EE) or inhibited (EI) by simultaneous stimulation of the ipsilateral ear
(Imig and Adrian, 1977). In the cat, EE/EI binaural columns are oriented orthogonal to isofrequency contours in the form of alternating bands (Imig and Adrian, 1977;
Middlebrooks et al., 1980; Kelly and Sally, 1988) each receiving convergent input from 81
spatially-segregated cell groups within the MGV (Middlebrooks and Zook, 1983). More
recently, this "divergence/convergence" scheme of auditory thalamocortical organization
has been questioned in favor of a traditional "point-to-point" topographic projection from
MGV to AJ(Redies et aI., 1989; Bradner and Redies, 1990).
In contrast to other sensory thalamocortical systems, there have been remarkably few studies of auditory geniculocortical pathways using sensitive anterograde tracers. We report that single injections of the anterograde tracers PHA-L or biocytin into the MGV label multiple "patches" of thalamocortical afferents to the auditory cortex. In the coronal plane, the patches have an intermittent distribution similar to the binaural bands described in electrophysiological studies (Imig and Adrian, 1977; Middlebrooks et aI.,
1980). Furthermore, single patches are elongated along the rostral caudal axis similar to the isofrequency contours in this cortex (McMullen and Glaser, 1982a).
METHODS
Experiments were performed on a total of eighteen normal young adult New
Zealand White rabbits (2-3 kg), obtained from local suppliers or from our breeding colony. All animals were anesthetized with ketamine (44 mg/kg Lm.) and xylazine (10 mg/kg, Lm.) and placed in a modified Kopf stereotaxic device. Electrode placement within the MGV was performed stereotaxically with coordinates determined from pilot studies, stereotaxic atlases (Paxinos and Watson, 1986; Shek et aI., 1986), and conventional parcellations of the MGB (Morest, 1964; LeDoux et aI., 1985; Winer, 82
1985; Winer, 1992). The anterior-posterior (AP), medial-lateral (ML) and dorsal-ventral
(DV) coordinates were computed relative to bregma (with bregma and lambda in the
same horizontal plane). Final stereotaxic coordinates for MGV injections were: AP: -7.9
mm, ML: + 5.5 mm and DV: -12.5 (measured from surface of skull). Borosilicate glass
capillary tubing, pulled to outside tip diameters of 10-30 urn using a Brown and Flaming
electrode puller, were filled with 2.5% phaseolus vulgaris leucoagglutinin (PHA-L,
Vector Labs; Gerfen and Sawchenko, 1984) in O.lM phosphate buffered saline (PBS) or
a 5% solution of biocytin in 0.05M Tris buffer(King et aI., 1989). The electrodes were
stereotaxically advanced into the MGB through small holes drilled through the skull.
Iontophoretic injections of tracers were made using cathodal (PHA-L) or anodal
(biocytin) current of 2-4 uA for 15 min. (50% duty cycle) using a Midgard CS-3 constant
current generator. At the conclusion of the injection, the electrode was removed, the
skull incision was sutured and chloromycetin (100 mg) administered intravenously. After
a survival period ranging from 48 hr (biocytin) to 10 days (PHA-L), the rabbits were
again deeply anesthetized and intracardially perfused. For animals receiving injections
of PHA-L, the pH-shift method of fixation suggested by Gerfen and Sawchenko (1984) was used. It consisted of a 500 ml wash of isotonic saline followed by 1 L of 4 % paraformaldehyde in 0.1 M acetate buffer (pH 6.5) at 4°C. This was followed by 1 L of 4% paraformaldehyde and .05% glutaraldehyde in 0.1 M borate buffer at pH 9.5 at
4°C. For biocytin experiments, animals were intracardially perfused with O.lM PBS followed by 4% paraformaldehyde in PBS (PH 7.4). The brains were then removed and fixed overnight at 4°C in the same fixative. After cryoprotection in graded sucrose 83
solutions (to 30%), serial frozen sections were cut at 80-100 urn and collected in PBS
(PH 7.4). In some cases, sections were incubated for 60 min. in 10% methanol and 1 %
> ••••• H202 to suppress endogenous peroxidase activity.
Immunoperoxidase visualization of PHA-L labeled thalamocortical axons was performed using the method developed by Gerfen and Sawchenko (1984). Sections were incubated for 24 hrs in PBS containing 2% normal rabbit serum (NRS) and 0.3% Triton
X-I00 followed by incubation in goat anti-PHA (1:2000 dilution) for 48 hrs at 4°C on a rocker table. The sections were transferred to a solution of biotinylated rabbit anti-goat
IgG followed by avidin-biotin horseradish peroxidase complex (ABC, Vector Labs).
Biocytin-labeled axons were visualized using a protocol suggested by Usrey et al (1992).
Sections were incubated in goat anti-biotin (1:20,000, Vector Labs) with 1 % NRS in PBS for 24 hrs at 4°C. The sections were then transferred to a solution of biotinylated rabbit anti-goat (1:200) with 1 % NRS in PBS followed by ABC complex (Vector). PHA-L- and biocytin-labeled axons were visualized histochemically using the nickel-cobalt diaminobenzidine method of Adams (1981). All sections were mounted on gelatinized slides and coverslip were applied using Permount. In some cases, sections were lightly counterstained with methylene blue or cresyl violet. Reconstructions of injection sites within the MGB and labeled axons in auditory neocortex were carried out with the aid of an image-combining computer microscope system (Glaser et aI., 1983) and
Neurolucida software (Microbrightfield Inc., Baltimore, Maryland). 84
RESULTS
Both anterograde· tracers· yielded very similar results: iontophoretic injections of
PHA-L (Fig. 1A) or biocytin (Fig. 1D) into the MGV labeled multiple (ca. 3-5)
"patches" of thalamocortical axons within the ipsilateral auditory cortex (Fig. 1B; Fig.
1C). The patches were approximately 500 urn in diameter and terminated broadly within
lamina III/IV with occasional axonal fascicles reaching to the lamina II/III border. Each
patch derived from the convergence of exceedingly fme terminal axons « 1.0 urn)
studded with bouton-like swellings (ca. 2.5 urn dia., Fig 2A). Although the tangential
extent of the patches in the coronal plane was limited, an examination of serial sections
revealed that single patches frequently extended for 2 mm or more along the rostral
caudal axis (Fig. 2B). Thus, in a plane tangential to the pial surface, the patches took on the form of elongated bands whose major axis was oriented rostral-caudally. In addition
to axonal patches within lamina III/IV, axons labeled by MGV tracer injections were found within the outer half of lamina I where they coursed tangentially for long distances
(Fig 1C).
DISCUSSION
A simple "point-to-point" organization of auditory geniculocortical pathways predicts that a focal injection of tracer into MGV would label a cortical band of terminal axons whose long axis was parallel to isofrequency contours and whose tangential extent 85 would be a direct function of injection size. Indeed, terminal fields in the present study were elongated in the rostral-caudal axis, an orientation which parallels the isofreqency contours of this species' tonotopic map (McMullen and Glaser, 1982a). However, each injection also labeled intermittent patches orthogonal to the isofrequency axis. Results of the present study suggest that coincident cell groups within MGV give rise to divergent thalamocortical pathways to AI.
An alternative explanation of our results is that the large MGV injections labeled divergent projections to multiple cortical auditory fields (Aitkin, 1990). Studies involving retrograde axonal transport methods have shown that single MGV neurons project to multiple tonotopic fields (Morel and Imig, 1987). However, labeled patches were always located within a cortical region defined by prior electrophysiological (McMullen and
Glaser, 1982a) and cytoarchitectonic studies (McMullen and Glaser, 1982b) as the large primary field or AI. Ongoing developmental studies have confirmed that even highly localized MGV injections (ca. 100 urn dia.) yield multiple thalamocortical patches suggesting that they are a fundamental feature of auditory thalamocortical organization
(de Venecia and McMullen, 1992).
Although a patchy distribution of thalamocortical axons is a characteristic organizational feature of the visual and somatosensory systems (Lorente de No, 1922;
Hubel and Wiesel, 1972; for review, see Jones, 1985), this is the first demonstration of a patchy distribution of thalamocortical axons in the auditory system. Patchy connections between the MGV and the AI in the cat have been predicted on the basis of retrograde tracer experiments (Andersen et aI., 1980a; Merzenich et aI., 1982) and patterns of 86 cortical degeneration following large MGV lesions (Vaughan and Foundas, 1982). Single,
PHA-L labeled axons arising from the monkey MGV terminate in the form of patches within AI similar to those described in the present paper (Hashikawa et al., 1992).
Similar afferent patches have also been described within the cat MGV following injections of anterograde tracers within the inferior colliculus (Andersen et al., 1980b) suggesting that the patches represent the cortical terminus of ascending parallel pathways originating in the auditory brainstem (Middlebrooks and Zook, 1983).
One physiological function having a patchy organization within AI is the binaural interaction bands in which the responses of neurons to contralateral sounds are either facilitated (EE) or inhibited (EI) by simultaneous stimulation of the ipsilateral ear (Imig and Adrian, 1977). We have also identified alternating EEtEI regions during tangential penetrations of the rabbit primary field similar in size and orientation to the anatomical patches labeled by anterograde tracers (Glaser and McMullen, 1980). In AI of the cat,
EE and EI neurons are segregated within AI in the form of alternating "bands"
(Middlebrooks et al., 1980). Detailed studies by Middlebrooks and Zook (1983) using injections of retrograde tracers into electrophysiologically-defined binaural bands demonstrated that neuronal groups projecting to either EE or EI bands are spatially segregated within the MGV (Middlebrooks and Zook, 1983). Their finding that restricted loci within subdivisions of the MGV converge on multiple binaural subdivisions within AI is consistent with the findings of the present study. Studies combining electrophysiological mapping of binaural classes with anterograde labeling will be necessary to determine if the thalamocortical patches represent binaural response-specific 87 pathways to AI. This strategy has been used successfully in delineating the differential projections of binaural bands in commissural and association pathways (Imig and Brugge,
1978; Imig and Reale, 1980; 1981).
It should be noted that the orientation of the anatomical patches described in the present study differs from the binaural EE/EI patches in the cat where they repeat along the isofrequency dimension (Imig and Adrian, 1977; Middlebrooks et ai., 1980;
Middlebrooks and Zook, 1983). However, the thalamocortical patches in rabbits are consistent with the binaural maps of two other lissencephalic mammals: the rat (Kelly and
Saslly, 1988) and the gerbil (Caird et ai., 1991). In the gerbil auditory cortex, 2-DG labeled EE patches are arranged parallel to the tonotopic axis with individual patches elongated to form bands parallel to isofrequency strips (Caird et al., 1991). When corrected for the 900 rotation of the gerbil's tonotopic map relative to the rabbit's, the topography of the binaural patches is comparable to the thalamocortical patches described in the present paper. In summary, our results are consistent with a model of MGV organization containing functionally-distinct, parallel anatomical pathways to AI, a model which can embrace both topographic (Brandner and Redies, 1990) and divergence/convergence theories of auditory thalamocortical circuitry (Merzenich et al.,
1982; 1984). This model is currently being tested in our laboratory by reconstructing the cortical terminal fields of single MGV axons. 88
Figure 3.1
A. Coronal section through thalamus and posterior auditory neocortex. Large
injection of PHA-L is restricted to MGV. Three "patches" of PHA-L labeled
thalamocortical axons are visible in ipsilateral auditory neocortex (open arrow). Inset.
Reconstruction of injection site shown in A. Hatched area shows spread of tracer within
MGV. Black area indicates location of filled neurons at injection site. B. Higher
magnification of thalamocortical axonal "patches" shown in A. MGV axons terminate
primarily within lamina III/IV. Methylene blue counterstain. C. Three "patches" of
thalamocortical axons within ipsilateral auditory neocortex resulting from biocytin
injection shown in D. Axons are also present within superficial lamina I. D. Coronal
section through brainstem showing biocytin injection restricted to the MGV. Methylene
blue counterstain. V, ventral division of MGB; M, medial division of MGB; D, dorsal division of MGB; A, anterior pretectal nucleus; SN, substantia nigra. In A bar = 2.0
mm; In Band C, bar = 250 urn; In D, bar = 1.0 mm. a9
.'
1•
. .... '\~.' ,; .,. .. "
o 90
Figure 3.2
A. Higher magnification photomicrograph of thalamocortical patch within lamina
III/IV showing the fine caliber of tenninal axonal branches and boutons. Bar = 50 urn.
B. Serial reconstruction of thalamocortical patches labeled by MGV injection shown in
Fig. 1D. Tenninal axons within lamina III/IV were manually marked at lOOX with the
aid of a computer microscope system. Every third section (each 100 urn thick) is shown.
Arrow denotes section shown in Fig. 1e. Bar = 1.0 mm. 91
·... . . '.~.\f":"']~: . -- v-<"- 92
CHAPTER FOUR
PARVALBUMIN IS EXPRESSED IN A RECIPROCAL CIRCUIT LINKING
THE MEDIAL GENICULATE BODY AND AUDITORY NEOCORTEX
ABSTRACT
Recent studies of the rabbit auditory forebrain showed that antibodies directed against the calcium-binding protein parvalbumin (PV) specifically demarcate auditory neocortex and the ventral division of the medial geniculate body (McMullen et aI, 1994b). The auditory cortex was characterized by two PV- immunoreactive bands: dense terminal-like labeling within lamina III/IV and a prominent band of PV + somata in lamina VIA. In some cases, PV immunocytochemistry of auditory cortex revealed patches of terminal-like label that are remarkably similar to thalamocortical patches labeled by the injection of anterograde tracers into MGV. The presence of PV + patches and the PV + somata in lamina VIa suggested the existence of a reciprocal PV + circuit linking AI and the MGV.
In the present study, double labeling experiments in adult rabbits were carried out to provide evidence for this circuit. Focal injections of the tracers biocytin or biotinylated dextran amine (BDA) into the MGV labeled thalamocortical afferent patches within lamina III/IV and retrograde-labeled corticothalamic neurons in lamina VIa of the ipsilateral auditory cortex. Adjacent sections stained with antibodies against PV revealed 93 terminal-like PV-immunoreactive patches in III/IV and PV + somata in VIa that were precisely in register with those labeled by BDA injections into the MGV. Serial section reconstruction of BDA-Iabeled corticothalamic neurons in VIa revealed pyramidal cells with tangentially-oriented basal dendrites and sparsely-branched apical dendrites that ascended to lamina I. Fluorescent double-labeling studies demonstrated that a subpopulation of corticothalamic neurons also express PV. PV-negative corticothalamic neurons were also found. Discrete injections of BDA into auditory cortex labeled a column of neurons in the ipsilateral MGV whose orientation paralleled the fibrodendritic laminae characteristic of this subdivision. Retrograde double-labeling experiments showed that most MGV relay neurons also express PV. Small numbers of PV negative relay neurons were also found. These studies provide evidence for the existence of multiple, chemically-coded pathways linking primary auditory cortex and the ventral division of the MGB.
INTRODUCTION
The calcium binding protein parvalbumin (PV) has been shown to be a useful chemical marker for the auditory neocortex and medial geniculate body in a variety of species including the monkey (Jones and Hendry, 1989; Morino-Wannier et aI., 1992;
Campbell and Pandya, 1994), cat (Hendry and Jones, 1991; Wallace et aI, 1991), rabbit
(McMullen et aI, 1994a; de Venecia et aI, 1995) and bat (Vater, 1994). In the cerebral cortex, a subset of GABA-ergic nonpyramidal neurons which includes basket cells, 94
chandelier neurons and small stellate cells are immunoreactive for PV (Celio, 1986;
Hendry et aI., 1989; DeFelipe et aI., 1989; Lewis and Lund, 1990; Van Brederode et
aI., 1990; Bliimcke et aI., 1990; McMullen et aI, 1994a; 1994b). Calcium binding
protein immunocytochemistry also distinguishes different classes of relay neurons in
dorsal thalamic nuclei. An emerging theme in several sensory systems is that thalamic
relay neurons belonging to distinct functional systems (e.g. lemniscal vs nonlemniscal),
and which project to different layers of the sensory neocortex, can be further
distinguished by their expression of calcium binding proteins (Rausell and Jones, 1991a;
1991b). For example, in the primate somatosensory thalamus, chemically distinct
compartments of the VPM nucleus relaying different components of somatosensory
information project to separate layers of the somatosensory cortex (Rausell and Jones,
1991a; 1991b). Large to medium PV + cells of cytochrome-oxidase (CO) rich rods
project to the middle layers of somatosensory cortex. Calbindin-positive cells in the CO
poor VPM matrix project to lamina I (Rausell and Jones, 1991a; 1991b). These studies
demonstrate that functionally distinct compartments of specific thalamic nuclei can be
differentiated on the basis of calcium-binding protein expression and by their projections
to separate layers of sensory neocortex ( Jones and Hendry, 1989; Rausell and Jones,
1991b). There is evidence that a similar parallel pathways from the MGV to AI also
exists. Hashikawa et aI. (1991), using the retrograde transport of fluorescent dyes in
combination with calcium binding protein immunocytochemistry, showed that PV + MGV cells project primarily to lamina III/IV, whereas calbindin-positive MGV cells project to
lamina I. 95
Recent studies of the rabbit auditory forebrain showed that the auditory cortex is
characterized by two PV- immunoreactive bands: dense terminal-like labeling within
lamina III/IV and PV + somata in lamina VIa. Although a portion of the terminal
labeling within III/IV is likely to arise from the high density of PV + basket cells within
this layer, a portion may originate from neurons in the MGV whose neurons also express
PV (McMullen et aI, 1994b). In contrast, the band within lamina VIa is composed of
lightly-labeled PV + pyramidal cells that constitute 15% of the cells in this layer.
Double-labeling experiments revealed that a portion of the PV + pyramidal cells in VIa
project to the MGV (McMullen et aI, 1994b) suggesting that PV is expressed in a
reciprocal circuit linking auditory cortex and MGV. In the present study, double labeling experiments were carried out to provide direct evidence for a chemically-coded reciprocal circuit linking AI and the MGV.
METHODS
Tracer injections. Experiments were performed on adult New Zealand White rabbits obtained from commercial suppliers. All animals were anesthetized with ketamine (40 mg/kg Lm.) and xylazine (10 mg/kg Lm.) during surgical procedures. Focal injections of biocytin or biotinylated dextran amine (BDA) were made into the MGV in 10 animals for anterograde labeling of thalamocortical axons and retrograde labeling of corticothalamic projection neurons. BDA was injected into AI in both cerebral hemispheres of 2 animals to label medial geniculate projection neurons in a retrograde 96 manner. A modified Kopf stereotaxic apparatus was used to advance glass electrodes through small trephine holes into the MGV or AI. Stereotaxic coordinates for the MGV and AI were determined from previous anterograde-labeling and electophysiological studies of the auditory cortex (McMullen and Glaser, 1982a; McMullen and de Venecia,
1993; de Venecia and McMullen, 1994). Borosilicate glass capillary tubing (1 mm O.D.) was pulled on a Flaming/Brown P-87 micropipette puller to manufacture electrodes with outside tip diameters of 20-30 urn. Electrodes were filled with either a 5 % solution of biocytin (Molecular Probes) in 0.05 M Tris buffer, pH 7.6 (King et aI., 1989) or 10% biotinylated dextran amine (MW = 10,000; Molecular Probes) in O.OlM phosphate buffer, pH 7.25 (Veenman et aI., 1992). Iontophoretic injections of either tracer were made by passing 2-4 uA positive current (50% duty cycle) for 10-20 minutes with a Midgard CS-3 constant current generator. The electrodes were left in place for at least 10 minutes after discontinuing the current to prevent spread of the tracer along the electrode tract during withdrawal. Once the injections were completed, scalp incisions were sutured and the animals were allowed to recover. The post-injection survival period for MGV injections ranged from 48 hours (biocytin;N=2) to as long as 11 days (BDA; N=8). Because retrograde transport of BDA with short survival periods results in punctate cytoplasmic labeling of neurons, a survival period of 4 days after cortical injections of BDA was used for double labeling studies of MGV projection neurons. After the appropriate post injection survival period, the animals were again deeply anesthetized and transcardially perfused with 0.1 M phosphate buffered saline (PBS; pH 7.4), followed by 4% paraformaldehyde in PBS. The brains were removed from the skull and stored overnight 97 in the same fIxative. Tissue blocks of the auditory cortex and of the thalamus were then dissected and cryoprotected in graded sucrose solutions (up to 30 %). Serial frozen sections were cut in the coronal plane with a sliding microtome and collected in PBS. All histochemical and immunohistochemical procedures were performed on free-floating sections using a platform rotator for gentle agitation.
Localization of injection sites. Thalamic or cortical sections (80-100 urn thick)
containing the injection sites were incubated for 15 minutes in 1 % H20 2 to suppress endogenous peroxidase activity. Biocytin and BDA were localized by avidin-biotin horseradish peroxidase histochemistry (Vector Elite ABC Kit) using nickel-cobalt intensifIcation of the diaminobenzidine reaction product (Adams, 1981). In many cases, sections were counterstained with 1 % aqueous methylene blue to help confIrm the location and extent of the injections.
Screen for biocytin or DDA labeling in AI. The auditory cortex ipsilateral to MGV injections was frozen sectioned at a thickness of 30um. Serial coronal sections were collected in PBS, and every tenth section was taken to screen for the presence of biocytin- or BDA-Iabeled thalamocortical axons and corticogeniculate neurons. Screening sections were placed in Vector Elite ABC solution (180 ul each of reagents A and B per
10 ml of 1 % Triton X-100 in PBS, pH= 7.4) and reacted with 0.01 % H202 in a nickel cobalt DAB solution (Adams, 1981). Sections containing labeled axons and cells were selected for immunohistochemical localization of biocytinlBDA or the calcium-binding 98
protein parvalbumin, and for double-labeling studies.
Anti-biotin immunohistochemistry. A more sensitive immunoperoxidase method was
used to visualize biocytin or BDA labeled axons and neurons in the cortex ipsilateral to
MGV injections (McMullen and de Venecia, 1993). Sections were treated with 1 % H202
for approximately 15 minutes to suppress endogenous peroxidase activity followed by 3 %
normal rabbit serum (NRS) in 1 % Triton X-100 to block nonspecific antibody labeling
and to increase antibody penetration. Sections were then incubated for 48 hrs at 4°C in
goat anti-biotin antibody (Vector) diluted 1: 10,000 in PBS containing 3 % NRS, followed
by biotinylated rabbit anti-goat IgG (Vector) diluted 1:200 in 3% NRS-PBS for 2 hours
at room temperature, and then by Vector Standard ABC solution (90 ul each of reagents
A and B per 10 ml of PBS) for 90 minutes. The sections were reacted using heavy metal
intensification of DAB (Adams, 1981), mounted onto gelatinized slides and coverslipped with Permount. Camera lucida drawings of BDA-Iabeled neurons were made with a Zeiss
Standard microscope equipped with a drawing tube and a Zeiss 63X oil-immersion objective (N.A. 1.25).
Parvalbumin immunohistochemistry. Cortical sections adjacent to those processed for biocytin or BDA-Iabeled thalamocortical axons, as well as sections of auditory cortex and
MGB from normal animals, were processed with a monoclonal antibody (Swiss
Antibodies #235) that recognizes fish, mouse, rat, monkey, and human forms of PV
(Celio et aI., 1988). The sections were treated with 1 % H20 2 for 15 minutes then 99 incubated 60 minutes in 0.1 M PBS, pH=7.4, containing 1 % Triton X-100 and 5% normal horse serum (NHS). Incubation in primary antibody diluted 1: 5000 in PBS containing 0.3% Triton X-100 and 5% NHS was carried out at 4°C for approximately
48 hours. The sections were then incubated in biotinylated horse anti-mouse IgG (Vector) diluted 1:200 in PBS containing 0.3% Triton X-100 and 5% NHS for 2 hours at room temperature followed by avidin-biotin-peroxidase complex (Vector Standard ABC kit) for
90 minutes. Visualization of PV immunoreactivity was completed by using the nickel cobalt method of Adams (1981). No immunoreactive labeling was present in sections incubated in anti-PVantisera preadsorbed with high-performance liquid chromatography purified rat PV (Swiss Antibodies). Because PV-immunoreactivity was examined in cortical sections from animals that had received anterograde tracer injections into MGV, adjacent control sections were processed by avidin-biotin-peroxidase histochemistry without antibody incubation to demonstrate that the patches of PV immunoreactivity were not due to the presence of tracer in axons and cells.
Immunofluorescent double labeling of PV containing corticogeniculate neurons.
Sections containing biocytin or BDA-Iabeled corticogeniculate projection neurons were incubated for 1 hour in a solution of 5 % normal rabbit serum (NRS) and 1 % Triton X-
100 in 0.1M PBS, pH=7.4. The sections were then incubated for 48 Hours at 4°C in a cocktail of goat anti-biotin IgG (1: 1000; Vector) and mouse monoclonal anti-parvalbumin
IgG (1:1000; Swiss Antibodies #235) in PBS containing 0.5% Triton X-100 and 5%
NRS. After rinsing in PBS, the sections were incubated for 2 hours in the dark at room 100
temperature in a secondary antibody cocktail consisting of tetramethylrhodamine
isothiocyanate-conjugated affInity purifIed F(ab')2 fragments of rabbit anti-goat IgG
(H + L) (Jackson Immunoresearch) and fluorescein isothiocyanate-conjugated affInity purifIed F(ab')2 fragments of rabbit anti-mouse IgG (H + L) (Jackson Immunoresearch) diluted 1:200 and 1:100, respectively, in PBS containing 5% NRS and 0.3% Triton X-
100. The sections were mounted onto clean glass slides and coverslipped with
Vectashield medium (Vector) for epifluorescence microscopy. Photomicrographs were taken with a Leitz Aristoplan epifluorescence microscope equipped with a Leitz Orthomat camera and narrow band emission fIlters (Omega Optical) for fluorescein and rhodamine.
No specifIc labeling for biocytin or BDA was seen in adjacent control sections incubated in the primary cocktail then in the secondary FITC-conjugated anti-mouse antisera only, or in sections incubated in anti-PV antisera only and the secondary antibody cocktail. In addition, no specifIc labelin.g of PV + neurons in AI (McMullen et aI., 1994a) was observed in control sections incubated in the primary antibody cocktail then in the secondary rhodamine-conjugated anti-goat antisera only, or in sections incubated in the anti-biotin antisera only and the secondary cocktail. The specifIcity of the anti-PV antisera was further confIrmed by the lack of immunoreactive labeling in sections incubated in antisera preadsorbed with high-performance liquid chromatography purifIed rat PV (Swiss Antibodies). Furthermore, the results of the immunofluorescence double-labeling experiments were confIrmed in experiments using a DAB-based double labeling protocol developed for the brightfIeld microscopic examination of PV- immunoreactivity in MGV neurons projecting to auditory cortex. 101
Double labeling of MGV projection neurons. The thalamus at the level of the MGB ipsilateral to BDA injections was frozen sectioned at a thickness of 30 urn. Serial coronal sections were treated with 1 % H20 2 for approximately 15 minutes to eliminate endogenous peroxidase activity followed by 1 % Triton X-100 and 5% normal horse serum (NHS) to increase reagent penetration and to block nonspecific antibody labeling.
Sections were then incubated for 48 hours at 4°C in a cocktail of Vector Standard ABC solution (120 ul each of reagents A and B per 10 ml diluent) and mouse monoclonal anti
PV antibody (Swiss Antibodies #235) diluted 1:5000 in PBS containing 0.3% Triton X-
100 and 3% NHS. After thorough rinsing in PBS, the sections were reacted for 6-8 minutes with DAB and 0.003% H202 in the presence of nickel and cobalt (Adams, 1981), yielding a punctate black reaction product in the cytoplasm of BDA containing neurons.
The sections were then rinsed 5 X 10 minutes in PBS and incubated for 2 hours at room temperature in biotinylated horse anti-mouse IgG (Vector) diluted 1:200 with 3% NHS-
0.3% Triton X-1OO in PBS. After thorough rinsing, the sections were incubated for 90 minutes at room temperature in Vector Elite ABC solution (120 ul each of reagents A and B per 10 ml of PBS). Sections were then preincubated for approximately 5 minutes in DAB (10 mg per 20 ml PBS) before reacting for 4-5 minutes in fresh DAB (same concentration) with 0.001 % H202, to yield a brown reaction product in PV immunoreactive neurons. Sections were mounted onto gelatin-coated slides and coverslipped with Permount for brightfield microscopic examination. With this protocol, the cytoplasm of double labeled neurons is characterized by a black punctate product on a brown background. The presence of cells containing only black punctate BDA-Iabeling 102
serve as an internal control for the specificity of the brown PV-immunoreactivity. In
addition, no brown labeling was seen in control sections incubated in the initial cocktail
solution containing ABC only (no PV-antisera). Brown labeling was also absent in
sections incubated in the anti-PV/ABC cocktail but not in the biotinylated anti-mouse
antisera.
RESULTS
Parvalbumin immunocytochemistry labels terminal-like patches in AI. A coronal section through the auditory cortex of a normal adult rabbit labeled by PV immunohistochemistry is shown in Fig 1. Three patches of terminal-like immunoreactivity can be seen in midcortical layers corresponding to lamina III/IV. In addition, a prominent band of PV + somata is present within lamina VIa. This pattern of PV immunoreactivity is remarkable similar to cortical labeling obtained by tracer injections into the MGV (see below).
Thalamocortical axonal patches coincide with PV + patches in AI. Focal injections of biocytin or BDA into the MGV label multiple patches of afferent axons in ipsilateral auditory cortex (Figs. 2A, 3A, 4). The thalamocortical patches occupy the full extent of layer IV and the bottom of layer III. Labeled axons are also prominent in the upper half of layer I and, less extensively, in VIa. Immunohistochemical localization of PV in sections adjacent to those containing biocytin- or BDA-Iabeled axonal patches revealed 103
PV + terminal-like patches in layers IIIIIV that were in close register with the
thalamocortical patches (Figs. 2B, 3). The prominent band of PV + somata in lamina
VIa also corresponds to the band of retrograde-labeled corticothalamic somata seen in
Fig. 2A. The lack of axonal labeling in control sections processed by avidin-biotin
peroxidase histochemistry without antibody incubation (Fig. 2C) and the presence of
PV + patches in III/IV in normal brains (Fig. 1) indicates that the terminal-like PV
labeling cannot be solely attributed to the presence of tracer within terminal axons and
cells. Moreover, double-labeling immunofluorescence studies confIrmed that the biocytin
and BDA-Iabeled thalamocortical patches were located within patches of PV + terminal
like labeling in the same section (unpublished observations). A higher magnifIcation view
of the correspondence between thalamocortical and PV patches is shown in Fig. 3. The
correspondence between the PV + patches and the axonal patches suggests that the
terminal-like labeling in the PV + patches is due in part to thalamocortical axons
originating from MGV. The registration of BDA labeled neurons in lamina VIa with
PV + somata in lamina VI suggests that PV is also expressed in corticothalamic
projection neurons.
Morphology of corticothalamic projection neurons. Although the pyramidal morphology of the layer VIa corticothalamic projection neurons was apparent after either biocytin or BDA injections, nearly complete fIlling of the cells to attain a Golgi-like staining was best achieved using BDA and long (9-11 days) postinjection survival periods. Because there have been no detailed descriptions of identifIed corticothalamic 104
projection neurons in the auditory system of any species, the backfilled neurons were
reconstructed from serial sections to characterize their morphology. Results of one such
experiment are shown in Fig. 4. Camera-Iucida drawings of three reconstructed cells are
shown in Fig. 5. The size and the prominent apical dendrites of these cells indicate that
they are small to medium pyramidal neurons. The moderately branched basal dendrites
had a preferential tangential orientation within layer VIa, and a spine density similar to
the lower apical dendritic branches. The horizontal extent of the basal dendritic field was
slightly less than the width of the apical dendritic field. A striking feature of these
pyramidal neurons was their sparsely branched apical dendrites which extended into
upper layer I (Fig. 5). Along their course, the apical dendrites produced a number of
branches in layers VIa, and V, and an occasional branch in layer III. There was a notable
lack of branches, however, within layer IV. The density of spines on the apical dendrites
and side branches was highest in layers V and IV, and reduced in the supragranular
layers.
PV expression in corticothalamic projection neurons. The registration of BDA labeled neurons in lamina VIa with PV + somata in lamina VI suggests that a population of corticothalamic projection neurons in AI expresses PV. The prominent band of PV + somata in lamina VIa consisted of large multipolar nonpyramidal cells embedded within a popUlation of lightly-labeled cells with a pyramidal-like morphology (Fig. 6).
Immunofluorescent double-labeling experiments revealed that at least some of these PV + cells were indeed corticothalamic pyramidal cells whose unbranched apical dendrites 105 reached lamina I (Fig. 7). The expression of PV in layer VIa pyramidal neurons was confirmed by DAB-based double-labeling experiments which allowed a permanent label visible by brightfield microscopy (data not shown). Both double-labeling techniques also revealed a large population of corticothalamic projection neurons that did not express PV
(Fig. 7). These data suggest that there are multiple parallel corticothalamic pathways linking AI with the MGV.
Parvalbumin expression in thalamocortical neurons of the MGV neurons. The correspondence between the axonal patches labeled by anterograde tracers injected into the MGV and the patches of PV + terminal-like labeling in layers III/IV of AI suggests that PV is expressed in MGV neurons whose axons terminate in patches. A PV-labeled coronal section through the MGB is shown in Fig. SA. PV immunocytochemistry clearly delineated the four major subdivisions of the MGB: the ventral (MGV), dorsal (MGD), internal (MGI) and medial (MGM) nuclei (de Venicia et aI, 1995). The entire MGV was intensely PV-immunoreactive due to dense labeling of both somata and neuropil which was filled with terminal-like puncta (Fig. SB). A significant proportion ofMGV cells (ca.
50%) were PV + (Fig. SB). Thick PV + axons were also present within the MGV and appeared to exit its dorsal-medial border. Results from the MGV double-labeling experiments are shown in Figs. 9 and 10. Focal injections ofBDA into AI labeled a band of neurons in MGV extending from dorsomedially to ventrolaterally (Fig. 9), an orientation which presumably reflects the isofrequency contours within this nucleus
(Calford and Aitkin, 19S3). Double labeling immunocytochemistry revealed that a large 106
proportion of MGV neurons labeled by the retrograde transport of BDA injected into
auditory cortex were also immunoreactive for PV (Fig. lOA). Although the vast majority
ofMGV neurons containing BDA were PV-immunoreactive, single BDA-Iabeled neurons
were also observed (Fig. lOB). A schematic diagram summarizing the reciprocal circuits
linking auditory cortex and the MGV is shown in Fig. 11.
DISCUSSION
Thalamocortical pathways
Thalamocortical axons labeled by focal injections of biocytin or BDA into the
MGV terminated primarily in patches in layers III/IV of AI, confirming the findings of
our earlier anterograde axonal tracing studies (McMullen and de Venecia, 1993; de
Venecia and McMullen, 1994). Serial section analysis in these previous studies further
demonstrated that the axonal patches seen in coronal sections are elongated (up to 2 mm
in length) in the anterior-posterior axis (McMullen and de Venecia, 1994) forming bands
that parallel the isofrequency contours in AI (McMullen and Glaser, 1985). Also
consistent with previous studies (McMullen and de Venecia, 1993; de Venecia and
McMullen, 1994) a less dense band of termination was seen in layer I. Single axon reconstructions in early postnatal animals have shown that terminal branches in layer I
arise from axons that contribute to the layer III/IV patches (de Venecia and McMullen,
1994). Although single axon reconstructions have yet to be carried out in the adult rabbit, these data suggest that single MGV axons project to III/IV as well as lamina I. Further 107
analysis of the pattern of thalamocortical afferent termination in the present study
indicates that axons originating from MGV also terminate sparsely in upper layer VI.
Thalamocortical fibers originating from primary sensory nuclei have also been shown to terminate in upper layer VI of somatosensory (Jones, 1975; Hendry and Jones, 1983a,b) and visual neocortex (LeVay and Gilbert, 1976; Peters and Saldanha, 1976).
PV-expression in the rabbit auditory cortex is characterized by terminal-like labeling in lamina III/IV and a continuous band of PV immunoreactive somata in layer
VI. PV + terminal labeling within midcorticallayers also delineates the auditory cortex of the monkey (Hendry et aI., 1990; Hashikawa et aI., 1991; Morino-Wannier et aI.,
1992) and the cat (Hendry and Jones, 1991; Wallace et aI., 1991). A novel finding in the present study is the coincidence of thalamocortical afferent patches with patches of PV + terminal-like labeling in layers III/IV of AI. Double-labeling experiments demonstrated that the majority of the PV + neurons in the MGV project to primary auditory cortex.
In addition, numerous PV + axons and terminals were found in the upper portions of layers I and VI, a distribution coextensive with the pattern of thalamocortical axon termination in these layers. These results are evidence that PV is expressed in a thalamocortical pathway originating in the MGV and terminating in layers III/IV, and possibly layers I and VI.
The double labeling experiments also revealed a smaller population of MGV projection neurons that lack PV-immunoreactivity. These neurons likely represent a separate parallel pathway from MGV to AI that may express the calcium-binding protein calbindin (Hashikawa et aI., 1991; McMullen et aI., 1994a; de Venecia et aI., 1995). 108
Double labeling studies in the monkey indicate that PV + relay neurons in the MOB
project to midcortical layers (Hashikawa et aI., 1991), and calbindin positive relay
neurons project to layer I. An immunohistochemical investigation of the expression of
calbindin and parvalbumin in the rabbit MOB found that these proteins were differentially
distributed in MOB subdivisions (de Venecia et aI., 1995; Chapter 2). Although PV
immunoreactive cells predominate in the MOV, there are islands of calbindin
immunoreactive neurons that are especially prominent in the posterior one third of the
nucleus. These CB+ neurons may represent a separate, chemically distinct pathway from
MOV to AI. Their projection pattern, however, is unknown.
Additional evidence that the PV + terminal-like labeling within layers III/IV originates from MOV projection neurons is provided by studies of PV expression in the
sensory systems of other mammals. PV + terminal-like labeling in midcortical layers is characteristic of the primary sensory neocortical areas of the rat (Bruckner et aI., 1994;
Van Brederode et aI., 1991), gerbil (Bruckner et aI., 1994), cat (Stichel et aI., 1987;
Hendry and Jones, 1991), and monkey (Jones and Hendry, 1989; Blumcke et aI., 1990; van Brederode, 1990; Blumcke and Celio, 1992). Somatosensory and visual relay neurons in the ventral posterior and dorsal lateral geniculate nuclei have been shown to express PV in the monkey, and double labeling studies have demonstrated thalamocortical axons labeled by anterograde tracer in the white matter below visual and somatic sensory areas of the cerebral cortex (Jones and Hendry, 1989). Electron microscopic studies demonstrating PV + asymmetric synaptic contacts within lamina IV of the monkey visual and somatosensory cortex provide further evidence for a PV + thalamocortical pathway 109
(Blumcke et aI., 1991; DeFelipe and Jones, 1991). Relay neurons in functionally distinct
compartments of the monkey somatosensory thalamus differentially express parvalbumin
and calbindin, and project to different layers of somatosensory cortex (Raussell and
Jones, 1991a,b; Rausell et aI., 1992).
The PV + terminal-like labeling in layers III and IV can also be attributed, in
part, to the local arborization of PV + cortical neurons. The most common type of PV
immunoreactive neuron in layers III and IV of rabbit auditory cortex is the large basket
cell with vertically oriented dendrites (McMullen et aI., 1994). The axons of these cells
extend tangentially oriented branches with vertical collaterals that form pericellular arrays
around unstained somata in these layers. Colocalization studies demonstrate that almost
all PV + cells in sensory neocortex also express the neurotransmitter gamma
aminobutyric acid (GABA) (Celio, 1986; Hendry et aI., 1989; van Brederode et aI.,
1989; 1990; Hendry and Jones, 1991; Morino-Wannier et al., 1992). In the sensory
neocortex of a variety of species, presynaptic terminals on pyramidal neuron somata form
symmetrical synaptic junctions and tend to be GABAergic indicating that basket cells play
an inhibitory role in local cortical processing (Ribak, 1978; Hendrickson et aI., 1981;
Keller and White, 1987; Houser et al., 1983; DeFelipe et aI., 1986; for review see Jones and Hendry, 1984). Basket cells form about half of their synaptic connections with the proximal dendrites and somata of pyramidal cells, and the remainder with the cell bodies and dendrites of nonpyramidal cells (Somogyi et al., 1983; Freund et al., 1986; Kisvardy et al., 1987). Direct evidence that basket cells are GABAergic has been provided by the immunohistochemical localization of GABA in synaptic terminals of basket cells filled 110 by intracellular injection of HRP (Somogyi and SoltEsz, 1986; Kisvardy et al., 1987).
Electron microscopic studies of thalamocortical connectivity indicate that every neuron having a dendrite in layer IV of primary sensory neocortex forms some proportion of its synapses with thalamocortical afferents (for review, see White, 1989). Although a direct demonstration of thalamocortical synapses onto an identified basket cell is not available, numerous studies in several species have confirmed that nonpyramidal neurons in sensory neocortex receive thalamocortical synapses (White, 1989). Mitani et al. (1985a,b) identified specific morphological cell types receiving direct MOB projections by combining intracellular recording and labeling of AI neurons after subcortical electrical stimulation. Stellate, tufted and fusiform cells in layer IV were found to receive monosynaptic input from fast-conducting MOB fibers and horizontal cells in layer I received inputs from slower conducting fibers.
The patchy distribution of both thalamocortical axons and PV + terminal-like labeling suggests a modular organization of circuits within the rabbit primary auditory cortex. Indirect evidence for PV + modular cortical circuits in the rabbit neocortex has been provided by the demonstration of barrel-like patches of PV-immunoreactivity within the vibrissae representation in primary somatosensory cortex (McMullen et al., 1994).
Additional evidence derives from correlative light and electron microscopic observations in area 17 and area 3b in the monkey (DeFelipe and Jones, 1991). Patchy aggregations of PV + terminal-like labeling in layers IV and VI reflect the preferential distribution of thalamocortical afferents in a series of microzones. These zones are separated by regions of neuropil that lack thalamocortical afferent synapses (DeFelipe and Jones, 1991). 111
Furthermore, the PV + terminals within the neuropil surrounding the patches form asymmetric synapses and apparently originate from local inhibitory neurons (DeFelipe and Jones, 1991). The patchy distribution of excitatory thalamocortical afferents surrounded by zones of inhibitory synapses may contribute to the formation of functional columns within these cortical areas (DeFelipe and Jones, 1991). The greater horizontal extent of the PV + patches compared to the thalamic afferent patches suggests that a similar organization of excitatory and inhibitory zones may be present in the neuropil of
AI in the rabbit.
Corticothalamic pathways
Previous retrograde tracing studies of sensory neocortex have demonstrated that corticothalamic axons arise exclusively from pyramidal neurons located primarily in layer
VI and to a lesser extent in layer V (Jacobson and Trojanowski, 1975; Jones and Wise,
1977; Wise and Jones, 1977; for review see Jones, 1985; White, 1989). In the cat auditory system, large injections of HRP into the MGB label continuous sheets of pyramidal neurons throughout the full depth of layer VI and upper layer V of AI (Kelly and Wong, 1981; Wong and Kelly, 1981; Rouiller and de Ribaupierre, 1985). A similar study in the hamster also found a continuous sheet of labeled pyramidal neurons throughout layer VI, but only occasionally were labeled cells observed in upper layer V
(Ravizza et al., 1976). In contrast, corticothalamic neurons labeled by tracer injections into the MGV in the present study formed a band of labeled neurons preferentially located in layer VIa. 112
The results of the present study are the first to indicate a preferential distribution
of corticothalamic projection neurons in layer VIa of auditory cortex. A preferential distribution of corticothalamic neurons within layer VIa has also been observed in the visual cortex of the grey squirrel (Robson and Hall, 1975). In contrast, corticothalamic neurons occupy the full depth of layer VI of the visual cortex in the cat (Gilbert and
Kelly, 1975) and the monkey (Lund et al., 1975). Within layer VI of monkey visual cortex, however, there is evidence for a segregation of corticothalamic neurons projecting to different layers of the dorsal lateral geniculate such that pyramidal neurons projecting to the parvocellular layers are preferentially distributed in the upper half of VI and those projecting to the magnocellular layers chiefly occupy the deeper half (Lund et al., 1975).
There is no evidence for a comparable dissociation of corticothalamic neurons in monkey somatosensory cortex, where they are found throughout layer VI (Jones and Wise, 1977).
Corticothalamic neurons are located mainly in the upper half of layer VI in the somatosensory cortex of the mouse (Hersch and White, 1981), throughout VI in the rat
(Wise and Jones, 1977b), and also to a lesser extent in layer V in both species (Wise and
Jones, 1977; Hersch and White, 1981).
In the present study, the population of layer VIa pyramidal neurons labeled by tracer injections into the MGV was coextensive with a population of PV + pyramidal neurons in AI. Double-labeling studies demonstrated that many of the PV + pyramidal neurons project to MGV. Direct evidence for a PV + corticothalamic pathway has not been presented in any other system. A significant number of the backfilled neurons in layer VIa, however, contained no detectable PV-immunoreactivity. Although this may 113 result from a lack of sensitivity of the PV-immunohistochemical techniques, it is likely that the non-immunoreactive corticothalamic neurons represent a separate parallel pathway from AI to the MGV. Separate parallel auditory corticothalamic pathways have been postulated to explain the differential effects of cooling the auditory cortex on acoustically evoked discharge patterns in the MGV (Ryugo and Weinberger, 1976).
Anatomical evidence for multiple corticothalamic pathway in the auditory system has been presented in the rat in the form of two distinct types of axon terminals labeled by injections of the anterograde tracer Phaseolus vulgaris-Ieucoagglutinin into AI (Rouiller and Welker, 1991). Small boutons were observed in the ventral and dorsal divisions of the MGB, the reticular nucleus, and the lateral part of the posterior nucleus (POI) whereas giant terminals were only present in a restricted zone of the dorsal division
(Rouiller and Welker, 1991). Consistent with these results, Andersen et al. (1980a) found terminal-like labeling in the ventral, medial and deep dorsal nuclei of the MGB, the reticular formation, and the POI after discrete injections of tritiated amino acids into physiologically defined loci in AI in the cat.
The Golgi-like labeling of corticothalamic neurons from MGV tracer injections revealed a population of small pyramidal neurons with thin apical dendrites that reached lamina I. In terms of morphology and laminar location, these neurons were very similar to double-labeled as well as PV-negative corticothalamic neurons. A characteristic feature of these pyramidal neurons was their thin apical dendrites that reached the pial surface.
There are no other detailed morphological descriptions of corticothalamic neurons in auditory cortex in any species. The morphology of these neurons differs markedly from 114 the morphology of thalamocortical neurons reported in layer VI of cat visual cortex
(Katz, 1987) which have apical dendrites that rarely extend beyond layer III.
Corticothalamic neurons in the monkey somatosensory cortex are similar to those in the visual cortex in that their apical dendrite gives off significant numbers of side branches within layer IV before terminating in or below layer II or III (Hendry and Jones, 1983).
In the mouse somatosensory cortex, Hersch and White (1981) reported that most layer
VI corticothalamic cells have apical dendrites that terminate within or just below layer
IV; only a few extend up to layer I. Recently, corticogeniculate neurons with long thin apical dendrites extending into layer I in rat visual cortex have been labeled after lateral geniculate nucleus injections ofBDA or biocytin in combination with the neurotransmitter
N-methyl-D-aspartate (Jiang et aI., 1993).
Results of the present study demonstrate that the apical dendrites of corticothalamic neurons projecting to MGV lie within the patchy axonal terminal field of MGV neurons projecting to layers III/IV, as well as layers I and VI, of AI. The proximity of the apical dendrites of corticothalamic neurons to the thalamocortical axons suggests that they may receive thalamocortical synapses, thus forming a monosynaptic feedback loop. Although a direct demonstration of thalamocortical synapse formation with corticothalamic neurons is lacking in the auditory system, numerous electron microscopic studies of thalamocortical connectivity indicate that every neuron class that extends dendrites into layers III/IV receives thalamocortical synapses (for review, see
White, 1989). Furthermore, an electron microscopic study by Hersch and White (1981) in mouse somatosensory cortex combining anterograde axonal degeneration and 115
retrograde neuronal labeling with HRP found synapses between thalamocortical axon
terminals and the apical dendrites of layer VI corticothalamic neurons. In the visual
system, physiologically identified corticothalamic cells are activated by electrical
stimulation of the optic tract at latencies commensurate with their receiving monosynaptic
input from thalamocortical axons (Gilbert, 1977; Harvey, 1978; Builler and Henry,
1979).
The functional role of the auditory corticothalamic projection remains uncertain.
Physiological experiments in anesthetized cats showed that a weak electrical stimulation of the auditory cortex prior to acoustic stimulation could either facilitate, inhibit, or have no effect on the response of MGV neurons (Watanabe et aI., 1966). Ryugo and
Weinberger (1976) found that reversible cooling of AI in unanesthetized cats differentially affected the discharge pattern of multiple-unit clusters in MGV. Cortical cooling decreased reverberatory discharges, but had no effect on the discharge rate of nonreverberatory clusters. The background activity of only nonreverbaratory clusters, however, was increased by cortical cooling. The short-latency response was unchanged for either group. These results suggest that the corticothalamic projection effects MGV by two different parallel mechanisms: one that facilitates discharges and a second that dampens evoked cell discharge. These effects could be mediated by local neuronal circuits within the MGV and by the differential connectivity of separate parallel corticothalamic pathways.
The MGV contains two general types of cells, projection neurons and inhibitory interneurons, which differ in their arrangement of synapses established with inferior 116 colliculus and corticothalamic axons (Morest, 1964; 1965). Inferior colliculus axons provide the dominant input to the MOV (Andersen et aI., 1980b; Oliver and Hall,
1978a). The projection neurons receive synapses from inferior colliculus axons on their somata and proximal dendrites. Corticothalamic axons terminate mainly on distal dendrites (Morest, 1965). On the other hand, corticothalamic fibers terminate closer to the somata of interneurons than afferent fibers from the inferior colliculus (Morest, 1965;
1975). Whether both relay neurons and interneurons receive corticothalamic input from the same class of neurons is unknown.
In conclusion, results in the present study indicate that calcium binding proteins delineate multiple reciprocal pathways linking auditory neocortex and MOV. These probes will be useful tools for guiding future connectional studies of the MOB with the brainstem and auditory neocortex. 117
Figure 4.1
Coronal section through auditory cortex stained with antibodies to PV. Note the three patches of terminal-like labeling within lamina III/IV and the prominent band of PV + somata in lamina VIa. Bar = 500 urn. 118
'.
. . .. ., ••••
, • • " '" V .. , tJ, • . , .. ~ . "" ,~ V. .. • " . , ." ... .. --... .' .... ,to • . .- . .• • .' ..
...... -- 119
Figure 4.2
Coronal section through auditory cortex showing correspondence between thalamocortical patches in lamina IIIIIV and corticothalamic neurons in VIa with PV immunoreactivity in adjacent section. A. Two thalamocortical patches labeled by focal injection of biocytin into ipsilateral MGV. A band of retrograde-labeled corticothalamic somata is also present in lamina VIa. B. Adjacent PV-immunostained section reveals patchy terminal-like labeling in III/IV and a prominent band of PV + somata in VIa. C. Control section from same animal processed with ABC histochemistry without antibody preincubation. Bar
= 500 urn. A 120
B
c
.. ' / . .. . 121
Figure 4.3
Higher magnification view of thalamocortical and PV-immunoreactive patches shown in
Figure 2. A. Biocytin-labeled thalamocortical patch in lamina III/IV. A band of retrograde-labeled corticothalamic pyramidal neurons is restricted to VIa. B. Adjacent section stained with antibodies to PV showing corresponding patch of PV immunoreactivity in III/IV and band of PV + somata in VIa. Note the concentration of
PV + nonpyramidal cells within the terminal-like patch in III/IV and the II apical II dendrites of some PV + cells in VIa. Bar = 200 urn 122
.. .:...,,~ Figure 4.4 Photomicrograph of thalamocortical afferent patch in III/IV and lamina VI corticothalamic cells following long-term survival after BDA injection into the ipsilateral MGV. Eleven-day survival after MGV injection results in Golgi-like labeling of corticothalamic neurons. Note the apical dendrites of corticothalamic neurons extending through the MGV terminal axons in III/IV. Bar = 200 um. 124 ,." ,." ... 125 Figure 4.5 Camera lucida, serial section reconstruction of BDA-Iabeled corticothalamic neurons in lamina VIa. Note the tangential orientation of basal dendrites and the relatively sparsely branched apical dendrites that extend to the pial surface. Exploded inset on right shows spines on apical dendritic segment within lamina IV. Bar = 50 urn. 126 I ...... D ...... _...... m ...... _.... : IV ...... ...... _...... - v ...... VIa ...... VIb 127 Figure 4.6 Photomicrograph of lightly-labeled, PV -immunoreactive pyramidal cells in lamina VIa. Arrows indicate apical dendrites. Several large PV + nonpyramidal cells are also present. Bar = 50 urn. 128 129 Figure 4.7 Fluorescence photomicrographs of double-labeled pyramidal cells in lamina VIa. A. Biocytin filled pyramidal cells in layer VIa retrograde labeled after an injection of biocytin into the ipsilateral MGV. Arrow denotes labeled pyramidal cell with unbranched apical dendrite that extends to lamina I. B. Identical section labeled with antibodies against parvalbumin. The corticogeniculate pyramidal cell shown in A is also immunoreactive for PV (arrow). Several other double-labeled cells are present. Asterisk in A denotes corticothalamic neuron that does not express PV. Bar in A = 50 urn. 130 131 Figure 4.8 Photomicrographs of coronal section through the medial geniculate body stained with antibodies to PV. A. The MGV (v) is strongly immunoreactive due to dense labeling of the neuropil and moderately-labeled somata. Bar = 500 urn. B. Higher magnification of MGV shown in A showing the PV -immunoreactive somata and neuropil. Bar = 100 urn. a: anterior pretectal area; ot, optic tract. 132 « 133 Figure 4.9 Retrograde double-labeling studies of MGV. A. Coronal section through auditory cortex showing iontophoretic injection of BDA into midcortical layers. Rhinal sulcus is at bottom center. Bar = 1.0 mm. B. Coronal section through MGV of same animal processed for both BDA- and PV immunocytochemistry. Arrows indicate column of retrograde-labeled cells within hypothetical "isofrequency slab". Boxed area is shown at higher magnification in Fig. IDA. Bar = 250 urn. 134 " ... _.. e. ,'. ~.----- '-"-'- -- 135 Figure 4.10 Retrograde double-labeling studies ofMGV. A. Photomicrograph of three double-labeled MGV projection neurons from Fig. 9B. Brown DAB reaction product represents PV immunolabeling, black punctate reaction product represents retrograde transport of BDA from ipsilateral auditory cortex. B. Photomicrograph ofPV-negative MGV neuron (left), PV-immunoreactive MGV neuron not labeled by BDA injection (middle) and double labeled MGV neuron (right). Bar = 25 urn. 136 137 Figure 4.11 Schematic diagram of reciprocal circuits linking auditory neocortex and MGV. See text for details. 138 Auditory Cortex I II-ill N v VIa VIb MGV ,. PV+ Neurons fj 0 Non-PV Neurons 139 CHAPTER FIVE SINGLE THALAMOCORTICAL AXONS DIVERGE TO MULTIPLE PATCHES IN NEONATAL AUDITORY CORTEX ABSTRACT Thalamic afferents originating in the ventral division of the medial geniculate body (MGV) terminate in patches within lamina IIIIIV of the primary auditory cortex in adult rabbits. Focal iontophoretic injections of the anterograde tracer biocytin were made into the MGV of neonatal rabbits to examine the morphological organization of auditory thalamocortical afferents prior to hearing onset. Thalamocortical afferents terminated in distinct patches as early as postnatal day 1 (PD-O = day of birth), 6 days before the behavioral onset of hearing. In contrast to thalamocortical afferents in adults, the terminal arbors of neonatal MGV axons occupied the entire depth of the cortical plate and lamina I. Serial section reconstructions revealed that single MGV axons in neonates branched to form multiple patches within the cortical plate. Collaterals also extended to lamina I where they coursed tangentially for several millimeters. An unusual feature of the neonatal thalamocortical patches was the contribution of descending collaterals from axons coursing in lamina I. The presence of distinct patches prior to hearing onset indicates that the segregation of auditory thalamocortical axons occurs in the absence of acoustically driven activity. The extensive postnatal remodeling of thalamocortical axons, 140 however, may indicate activity-dependent refmement of arbor size. INTRODUCTION A common feature of the visual and somatosensory neocortex in many mammals is a patchy distribution of thalamocortical axons (Jensen and Killackey, 1987; Humphrey et aI., 1985; Landry et aI., 1982; Gilbert and Wiesel, 1979; Hubel and Wiesel, 1972). The patchy organization of thalamocortical afferents corresponds to the distribution of functional modules within these sensory cortices (Landry et aI., 1982; Hubel and Wiesel, 1977; Woolsey and Van der Loos, 1970; Lorente de No, 1922). Developmental studies have shown that the adult patterns of thalamocortical patches arise from an initially diffuse projection that is refined during maturation (Ghosh and Shatz, 1992; Senft and Woolsey, 1991; Erzurumlu and Jhaveri, 1990; LeVay et aI., 1980; LeVay et aI., 1978; Rakic, 1977; Rakic, 1976). Although retinal ganglion cell activity is critical for this refinement process in the visual system (Stryker and Harris, 1986; for review see Goodman and Shatz, 1993), in other centers (e.g. the rodent trigeminal system, Chiaia et aI., 1992; Henderson et aI., 1992) peripherally-driven neuronal activity is not required for central pattern formation indicating that sensory systems differ in their dependence on functional activity. In contrast to the visual and somatosensory systems, there have been few detailed descriptions of thalamocortical afferent development in the auditory system (de Venecia and McMullen, 1992; Kageyama et aI., 1992; de Venecia and McMullen, 1991). Recent 141 studies in our laboratory employing sensitive anterograde tracers demonstrated that afferents originating from the ventral division of the rabbit m~dial geniculate body (MGV) terminate in patches in primary auditory cortex (McMullen and de Venecia, 1993). In the coronal plane, dense patches of terminal axonal arbors (ca. 500 urn in width) occupy lamina IV, and the lower part of lamina III. A secondary zone of termination is also present in lamina I. The patches are elongated in the rostral-caudal axis forming bands up to 2 mm long that run parallel to physiologically defined isofrequency lines (McMullen and Glaser, 1982a). While the functional significance of the thalamocortical patches remains obscure, their periodic distribution in the coronal plane is consistent with the alternating pattern of binaural interaction bands in AI (Merzenich et aI., 1984; Merzenich et aI., 1982; Glaser and McMullen, 1980; Middlebrooks et aI., 1980; Imig and Adrian, 1977). In the present study, anterograde labeling with biocytin was used to examine the organization of the auditory geniculocortical pathway prior to hearing onset. The rabbit was selected as a model to address this issue because of the late onset of cochlear function in this species (ca. 5 days after birth, Anggard, 1965), and the delay in behavioral onset of hearing until the end of the first postnatal week (Foss and Flottorp, 1974). We report that MGV axons terminated in patches within auditory cortex as early as postnatal day 1 (PD-l). Serial section reconstruction revealed that single thalamocortical afferents branched to form multiple terminal arbors within the cortex. These results indicate that neonatal thalamocortical patches derive, in part, from the branching pattern of single MGV neurons and that such patches in AI form in the 142 absence of extrinsic acoustic activity. METHODS Anterograde labeling of thalamic afferents to the presumptive auditory cortex of neonatal rabbits was produced by extracellular iontophoretic injections of biocytin into the MGV. Neonatal animals were obtained from our breeding colony of New Zealand White rabbits stocked from commercial suppliers. Results of the present study were based on bilateral injections made into the MGV in 12 animals ranging in age from PD-O (= day of birth) to PD-2. All animals were anesthetized with ketamine (20 mg/kg Lm.) and xylazine (2.5 mg/kg Lm.). Electrode placement into the MGV was accomplished with the aid of a modified Kopf stereotaxic device with stereotaxic coordinates determined from pilot studies and conventional parcellations of the MGB (Winer, 1992; Winer, 1991; Jones, 1985; LeDoux et al.,1985; Winer, 1985). Borosilicate glass capillary tubing (1 mm D.D.) was pulled on a Flaming/Brown P-87 micropipette puller to manufacture electrodes with outside tip diameters of 10-30 urn. Electrodes were filled with a 5% solution of biocytin in 0.05 M Tris buffer, pH 7.6 (King et aI., 1989) and advanced into the MGV through small holes in the skull. Iontophoretic injections of biocytin were made by passing 2-4 uA positive current (50% duty cycle) for 8-10 minutes with a Midgard CS-3 constant current generator. After the injections were completed, scalp incisions were sutured and the animals were allowed to recover. After 24 hours, the animals were again deeply anesthetized and transcardially perfused with 0.1 143 M phosphate buffered saline (PBS; pH 7.4), followed by 4% paraformaldehyde in PBS. The brains were removed from the skull and stored overnight in the same fixative. The cerebral cortex was dissected from the thalamus for separate immunohistochemical processing. After tissue blocks were cryoprotected in graded sucrose (to 30%), serial frozen sections were cut at 100 urn in the coronal plane with a sliding microtome and collected in PBS. Free-floating sections were incubated for 15 min. in 1 % H202 to suppress endogenous peroxidase activity. Biocytin injection sites in the MOB were localized with the aid of avidin-biotin-horseradish peroxidase histochemistry (Elite ABC Kit, Vector) and nickel-cobalt intensification of the diaminobenzidine reaction product{Adams, 1981). A more sensitive immunoperoxidase method was used to visualize labeled axons in the cortex ipsilateral to successful MOV injections (McMullen and de Venecia, 1993). Serial coronal sections through the presumptive auditory cortex were incubated for 48 hrs at 4°C in goat anti-biotin antibody (1:20,000; Vector) in PBS containing 1 % normal rabbit serum (NRS). The sections were then incubated in a solution of biotinylated rabbit anti goat IgO (I :200, Vector) for 2 hrs. ABC histochemistry with nickel-cobalt intensification was performed as described above. All sections were mounted onto gelatinized slides and coverslipped with Permount. Sections through the MOB were counterstained with 0.1 % methylene blue. Cortical laminae through the presumptive auditory cortex were determined from cresyl violet stained paraffin sections from the littermates of injected animals or from the hemisphere contralateral to a successful MOV injection. Camera lucid a drawings of labeled axons and neurons were made with a Zeiss Standard 144 microscope equipped with a drawing tube and a Zeiss 63X oil-immersion objective (N .A. 1.25). Serial reconstructions of thalamocortical patches as well as single axons were carried out with an image-combining computer microscope system (Glaser et aI., 1983) and Neurolucida software (Microbrightfield, BaIt., MD). Tissue section contours such as the pial surface, white matter and laminar boundaries were first digitized at low magnification. All biocytin-labeled axons in each section were then digitized at 630X in the form of short chords. For the reconstruction of population terminal fields (e.g. Fig. 2C), no attempt was made to follow individual axons through serial sections. Cut endings and pial landmarks were used to serially align each digitized section in order to fully reconstruct the thalamocortical patches. For the serial reconstruction of single axons, tracings began in the internal capsule to include early branch formation within the deep white matter (Humphrey et al., 1985). Axons were followed to their termination in each section. Endings were coded as boutons, growth cones or truncated at the obverse or reverse face of the section. The truncated endings were used for computer-assisted realignment with an adjacent section, and for the pursuit ofaxons into serial sections. A total of five neonatal thalamocortical axons were fully reconstructed from as many as eleven 100 urn thick serial sections. 145 RESULTS Examples of focal biocytin injections restricted to the MGV at PD-1 and PD-3 are shown in Fig 1B and 2C, respectively. Biocytin labeled neurons at the injection site were of the bushy type, characterized by their tufted dendrites (Winer, 1991) and axons which exited the dorsal-medial perimeter of the MGV (Fig. 2B). Focal injections of biocytin in MGV as small as 100 urn in diameter labeled patches of thalamocortical afferents in the auditory cortex of neonatal animals (Figs. 1 & 2). As in the adult, the patches were elongated in the rostral-caudal axis, forming bands up to 1-2 mm in length. In contrast to those of adults, thalamocortical axons in the PD-1 rabbits occupied the nascent lamina IV, the full extent of the cortical plate and lamina I (Figs. 1 & 2). By PD-3, the cortex had acquired an adult-like laminar organization; however, the terminal field of the thalamocortical axons still occupied the full depth of the granular and supragranular layers (Fig. 3). At both ages studied, the projection of thalamocortical axons to lamina I was far more extensive than in adults (Fig. 1-3). Serial section reconstruction revealed that single axons gave rise to branches in the white matter which diverged to multiple patches within the cortex (Figs. 1 & 3). The morphology of single axons reflected the popUlation reconstructions in that axonal arbors branched extensively within middle cortical layers to form terminal arbors that filled the entire depth of the cortical plate and lamina I (PD-1), or laminae I-IV (PD-3). An unusual feature of the thalamocortical projection in the neonates was the substantial contribution of descending collaterals of lamina I axons to patches within laminae II, III 146 and IV (Fig. 4). DISCUSSION The fIrst auditory cortical evoked responses in rabbits appear at postnatal age 4-6 (Konig and Marty, 1974; Klyvania and Obrastova, 1968), coincident with the appearance of the endocochlear potential, cochlear microphonic and summating potential (Anggard, 1965). The present study demonstrates that the segregation of thalamocortical axons into distinct patches, though morphologically immature, occurs prior to acoustically-driven cortical activity, and well before the behavioral onset of hearing at PD-7 (Foss and Flottorp, 1974). Prior to the onset of cochlear function, spontaneous electrical activity in the auditory system may playa role in this segregation process (Rubel, 1984; Rubel, 1978). Recent studies in the rodent somatosensory system, however, have shown that infraorbital nerve blockade from birth does not disrupt pattern formation in the brainstem, somatosensory thalamus, or barrel cortex (Henderson et aI., 1992). Furthermore, dLGN axons form patches even under binocular impulse blockade (Antonini and Stryker, 1993) indicating that factors other than activity can facilitate segregation (Bolz et aI., 1993). Several distinctive features of the immature thalamocortical patches indicate exuberant axonal growth during the early events of auditory geniculocortical maturation. In the neonate, the terminal fIelds of thalamocortical axons occupy the full depth of the granular and supragranular layers (the cortical plate at PD-1). In adults, thalamocortical 147 patches are restricted to lamina IV and the base of lamina III (McMullen and de Venecia, 1993). Moreover, the projection of neonatal MGV neurons to lamina I is far more extensive relative to that of adults. Ongoing developmental studies have revealed considerable immaturity of thalamocortical patches well past hearing onset at PD-7 (de Venecia and McMullen, 1992). The significant axonal remodelling of the thalamocortical patches during postnatal maturation suggests the possibility of activity-dependent refinement of thalamocortical arbors (Antonini and Stryker, 1993) and target neuron dendrites (Kageyama et aI., 1992; McMullen et aI., 1988a; McMullen et aI., 1988b). The projection of individual MGV neurons to lamina I as well as lamina III/IV is a novel finding not seen in other studies of thalamocortical development. Although Kato et al. (1984) reported exuberant lateral geniculate projections to lamina I of area 17 in kittens, their methods did not allow them to determine that the lamina I projections arose from collaterals of neurons projecting to midcortical layers. Even more striking are the descending arbors of lamina I axons to the patches in middle cortical layers. In adult rabbits, lamina I axons have relatively few branches which are largely confined to lamina I (McMullen and de Venecia, 1993). Furthermore, retrograde studies in the adult monkey auditory system have provided evidence that the thalamocortical projection to lamina III/IV and lamina I arise from chemically distinct cell types in the MGV (Hashikawa et aI., 1991). Ongoing developmental studies are attempting to determine whether the collateral branches to lamina I and their descending projections represent a transient growth phase, or indicate the existence of multiple classes of thalamocortical axons arising from the MGV (Usrey et aI., 1992; Humphrey, et aI., 1985). Finally, the patchy 148 projections of thalamocortical axons in the visual and somatosensory system have been shown to correspond to functional domains within sensory neocortex. Combined anatomical and electrophysiological studies will be necessary to determine the functional significance of thalamocortical patches in the auditory system. 149 Figure 5.1 Thalamocortical patches at PD-l. A. Camera lucida drawing of three thalamocortical patches labeled by biocytin injection within the MOV. The patches are formed by extensive branching of ascending axons within the nascent lamina IV and the cortical plate. Note the robust projection to lamina I whose axons contribute descending collaterals to deeper lying patches. Drawn from three 100 urn thick sections. Bar = 100 urn. B. Nissl-counterstained coronal section through ipsilateral MOB showing biocytin injection within the ventral subdivision (v). m: medial division; d, dorsal division. Bar = 200 urn. C. Computer microscope reconstruction (from nine 100 urn thick sections) of single axon labeled by biocytin injection into the MOV at PD-l. After exiting the internal capsule, the axon gives rise to multiple branches within the white matter (wm) which contribute to two patches. Inset shows cortical location of reconstruction. 150 '- ...... CP IV " ...... -.. . .. ' v .... - ...... ...... VI A CP IV V VI ...... '. WM ' ...... ", " ", ", ", ...... " ...... c soo 151 Figure S.2 Thalamocortical patches at PD-3. A. Nissl-counterstained coronal section through MGB showing biocytin injection (arrow) within ventral (v) division. m, medial division of MGB; d, dorsal division of MGB. Bar = 200 urn. B. Camera lucida reconstruction of all labeled cells at injection site. Two bushy neurons, whose axons exit the dorsal medial edge of MGV (arrows), are filled in a Golgi-like manner. The somata of nine lightly labeled neurons are also shown (stippling). Bar = 20 urn. C. Complete reconstruction, from twelve serial sections, of two thalamocortical patches labeled by biocytin injection shown in A. Thalamocortical axons within each patch span the full depth of the granular and supragranular layers. Note the tangential axons coursing within lamina I. 152 B II III IV --- V ---- \ \ \ \ \ \ \ \ \ \ \ VI \ \ , \ , ---"'" \ , , \ , --- \ \ \ \ , \ , , WM \ , \ , \ , \ \ \ \ \ \ \ \ \, ..;i' ~~ \~\.:: 500 c \~;\~>'" , "". " , , , , 153 Figure 5.3 Computer microscope reconstruction of single MGV axon contributing to the two thalamocortical patches shown in Fig 2C. Note axon collateral (arrow) which ascends unbranched through the granular and supragranular layers where it turns tangentially and courses for more than 2.0 mm in lamina I. Along its course, the axon gives rise to several descending branches which converge on patches formed by ascending branches. Inset shows cortical location of axon. 1 ...... -~- ... - ...... _... --- ...... v --- , ...... , .. .. ...... , \ \ \ \ , ,, 500 .. ----- , .. , .. , .. " ...... 155 Figure 5.4 Higher magnification view of thalamocortical patch shown in Fig. 2C and 3. thalamocortical patch is formed by the convergence of ascending branches (stippled) and the descending collaterals ofaxons coursing in lamina I (solid). 156 I , I , I I I I I I , I I ,I I ,I I , I , , , I I , ....c , I I , I S I I > , , I = , I I , , ,I , I , I I 157 CHAPTER SIX DISCUSSION Ascending pathways relayed by the MGB to auditory cortex. The medial geniculate body is the major auditory nucleus of the thalamus and is conventionally divided into three main subdivisions: the ventral, dorsal and medial divisions. In addition, several smaller nuclei are usually recognized in the MGB (Morest, 1964; 1965; Winer, 1985), including the ventrolateral nucleus in the ventral division, the deep dorsal and superficial dorsal nuclei in the dorsal division, and the suprageniculate nucleus. Each division receives a distinct set of ascending projections from the inferior colliculus and surrounding midbrain nuclei. The majority of ascending input into the MGB originates from the inferior colliculus, which provides a site for the summation of lower auditory brainstem processing as well as an opportunity for further processing. Multiple binaural and monaural pathways that have diverged from the cochlear nucleus converge within the inferior colliculus. The existence of multiple cell types in the inferior colliculus could allow for several types of auditory information to be transmitted in parallel to the MGB (Oliver and Hall, 1978a, Andersen, 1980a; Oliver, 1982). A useful scheme for the organization of parallel auditory pathways originating in the midbrain and relayed through the MGB to auditory cortex has been proposed by Oliver (1982). This scheme is based on the results of connectional studies in a variety of species, especially 158 the cat (Morest, 1964; 1965; Andersen et aI., 1980a,b; Calford and Aitkin, 1983) and the tree shrew (Casseday et al., 1976; Oliver and Hall, 1978a,b). Four major pathways are recognized. First, the central pathway originates in the central nucleus of the inferior colliculus, ascends through the brachium of the inferior colliculus, and terminates in the ventral division of the MGB. The ventral division in tum projects to the primary auditory cortex. Second, the pericentral pathways begin in the dorsal cortex and paracentral nuclei of the inferior colliculus and project to the deep dorsal, dorsal, and ventrolateral nuclei. Third, the lateral tegmental system (Morest, 1965) originates in the lateral midbrain tegmentum, primarily the sagulum, and the deep layers of the superior colliculus, and terminates in the dorsal division and suprageniculate nuclei (Morest, 1965; Oliver and Hall, 1978a; LeDoux, 1985). After synapses in the MGB, both the pericentral and lateral tegmental pathways continue to nonprimary auditory areas surrounding AI. Finally, the widespread pathway originates from nuclei throughout the inferior colliculus, and is relayed by neurons in the medial division of the MGB to both primary and nonprimary auditory cortex. Descending Pathways from Auditory Cortex. Each of the nuclei in the MGB also receive distinct patterns of corticothalamic input, as revealed by anterograde and retrograde tracing methods (Winer et al., 1977; Oliver and Hall, 1978b; Andersen et al., 1980a; Rouiller and de Ribaupierre, 1985; Morel and Imig, 1987; Rouiller et al., 1989; Morel and Kaas, 1992). In general, the 159 pattern of thalamocortical projections from each subdivision is matched by a similar reciprocal pattern of corticothalamic projections. Connections between AI and the MGV and, to a lesser extent, the MOM are topographic and tonotopically organized, but connections between nonprimary auditory cortex and MGD are more diffuse and lack tonotopicity (Andersen et aI., 1980; Calford and Aitkin, 1983; Rouiller and de Ribaupierre, 1985; Rouiller et aI., 1989; Morel and Kaas, 1992). Andersen et al. (1980) have proposed that there are two functionally distinct and segregated systems of reciprocal connections between the auditory cortex and the MGB of the cat: 1) a cochleotopic system arising from sharply tuned neurons in rostral subdivisions of the MGB (i.e. the ventral, deep dorsal and medial divisions) and terminating in tonotopically organized cortex, and 2) a diffuse system containing broadly tuned neurons in more posteriorly located nuclei (i.e. dorsal, ventrolateral and medial division) that project to auditory cortical fields lacking a precise tonotopic organization. The cochleotopic system thus includes the central pathway, and the diffuse system the pericentral and lateral tegmental pathways recognized by Oliver (1982). Although the widespread pathway may be included in either the cochleotopic or diffuse systems, the presence of a tonotopic organization in the medial division suggests that it should be placed in the former (Rouiller et al., 1989; Morel and Kaas, 1992). In addition to the direct corticothalamic connections, medial geniculate neurons may be indirectly influenced by the auditory cortex through corticofugal projections that terminate in the dorsal cortex and pericentral nucleus of the inferior colliculus. Corticofugal axons from AI also terminate in the reticular nucleus, which projects to the 160 ventral and dorsal subdivisions (Rouiller and Welker, 1991). Although the functional significance of these feedback loops is not clearly understood, they are probably involved in the modulation and gating of auditory information ascending to the cortex. The organizational schemes proposed by Andersen et aI. (1980) and Oliver (1982) provide a general outline for studying connections between the auditory thalamus and cortex; however, a detailed knowledge of the antomical and functional properties of the cells that give rise to the seperate parallel pathways is necessary for understanding the role of these pathways in hearing. Using the rabbit as a model, experiments conducted in the present study were designed to elucidate the detailed organization and development of parallel pathways connecting primary auditory cortex and the MGB, particularly the ventral division. Summary and Discussion of Major Findings in Present Study Experiments in Chapter 2 employed immunohistochemical techniques to examine the distribution of the calcium-binding proteins parvalbumin and calbidin in the rabbit MGB. These proteins have served as useful markers for relay neurons that belong to functionally distinct compartments in the thalamic sensory nuclei of other species (Jones and Hendry, 1989; Celio, 1990; Hashikawa and Jones, 1991; Rausell and Jones, 1991a,b; Andressen et aI., 1993; Diamond et aI., 1993). The differential expression of PV and CB clearly delineated not only the major subdivisions of the rabbit MGB, but also smaller nuclei that are difficult to discern with routine Nissl stains alone. The power 161 of calcium-binding protein immunohistochemistry to delineate subdivisions is illustrated in the present study by the identification of the CB + internal division as separate from the predominantly PV + medial division, a distinction not made in previous organizational schemes based primarily on the Nissl cytoarchitecture in the rabbit MGB (Jones, 1985). Further connectional studies are needed to confirm the homology of the MGB nuclei identified in the present study with subdivisions in other mammals. A comparison with previous connectional studies in other mammals, however, suggests that PV is predominantly expressed in nuclei (i.e. the MGV, Dd and MGM) belonging to the cochleotopic system as outlined by Andersen et al. (1980), and corresponding to the central and widespread pathways as described by Oliver (1982). Each of the nuclei in these pathways receive a substantial input from the central nucleus of the inferior colliculus. In contrast, CB-immunoreactivity characterized nuclei (i.e. the MGD, MGI, VI, Ds and SG) that belong to the diffuse system of Andersen et aI., and correspond to the pericentral and lateral tegmental pathways (Oliver, 1982). These results are consistent with the suggestion by Hashikawa et al. (1991) that there may be a "calbindin system" of auditory pathways from the midbrain to the cerebral cortex that is more diffusely organized than a "parvalbumin system". In Chapter 3, anterograde axonal tracing experiments in adult rabbits demonstrated that thalamocortical axons originating from the MGV terminate in AI primarily in patches occupying the full extent of layer IV and the bottom of layer III, and less densely in layer I. Further analysis of the data indicates that another zone of less dense termination may be present in layer VI. Although the functional significance of the 162 patches in III/IV is yet to be clearly established, they are elongated in the rostral-caudal dimension similar to the orientation of isofrequency contours in AI. Furthermore, the topography of the patches in the coronal plane is similar to the intermittent distribution of binaural interaction (EE/EI) regions in rabbit AI (Glaser and McMullen, 1980). The rostral-caudal bands of axonal termination labeled by focal injections of tracer in MGV is consistent with a point-to-point topographic organization of thalamocortical projections along isofrequency contours (Redies et aI., 1989; Brandner and Redies, 1990). The intermittent distribution of the patches in the coronal plane, however, suggests that coincident cell groups in MGV give rise to divergent projections to AI. A divergence/convergence model of thalamocortical projections from coincident MGV neurons has been proposed to account for the periodic distribution of binaural EE or EI bands oriented orthogonal to isofrequency contours in AI of the cat (Middlebrooks and Zook, 1983). It should be noted that in contrast to the cat, where EE/EI patches repeat along isofrequecy contours, EE/EI regions are arranged in parallel with isofrequency contours (Glaser and McMullen, 1980) in AI of the rabbit. A similar arrangement of binaural patches is seen in two other lissencephalic species, the rat (Kelly and Sally, 1988) and the gerbil (Caird et aI., 1991). Our results would be consistent with a model of MGV organization containing functionally distinct, parallel anatomical pathways to AI that includes both point-to-point topograhic and divergence/convergence elements. In Chapter 4, evidence for multiple parallel thalamocortical pathways connecting MGV to AI is provided by experiments combining immunohistochemical localization of parvalbumin and neuroanatomicallabeling ofaxons and projection neurons. Two major 163 findings indicate that PV is expressed in a subpopulation of MGV neurons that project to auditory cortex. First, thalamocortical afferent patches coincide with patches of PV + terminal-like labeling in layers III/IV of AI. Second, a subpopulation of neurons in MGV that project to AI are also labeled by PV-immunohistochemistry. The selective expression of PV has been associated with thalamocortical pathways that are functionally distinct from and parallel to pathways expressing CB in the visual and somatosensory systems of several species (Jones and Hendry, 1989; Celio, 1990; Rausell and Jones, 1991a,b; Diamond et al., 1993). The correspondence between the PV + patches and the afferent patches is consistent with the study by Hashikawa et al. (1991), who combined retrograde neuroanatomical tracing and calcium-binding protein immunohistochemistry in the monkey auditory system to provide indirect evidence for a PV + pathway projecting from MGV to midcorticallayers in AI, and a CB+ pathway projecting to layer I from the MGV and other MGB subdivisions. The presence of both projection neurons that are not PV-immunoreactive, and islands of CB + cells in the rabbit MGV (Chapter 2) are consistent with the existence of a CB+ pathway from MGV to AI. Double-labeling experiments also found that PV is expressed in a subpopulation of corticothalamic projection neurons labeled by the retrograde transport of tracer injected into the MGV. The expression of PV in subpopulations of thalamocortical and corticothalamic projection neurons suggests that there is a PV + feedback loop connecting MGV and AI. Since PV is expressed in fast-spiking neurons that can maintain non adapting high frequency firing rates (Kawaguchi and Kubota, 1993), one can speculate that this loop serves to modulate the activity of neurons with reverbaratory response 164 characteristics in the MGV (Ryugo and Weinberger, 1976). A patchy organization of thalamocortical afferents corresponds to the distribution of modular cortical processing units in the mammalian visual (Hubel and Wiesel, 1972; 1977; Livingston and Hubel, 1984; Antonnini and Stryker, 1993) and somatosensory cortex (Woolsey and Van der Loos, 1970; Jensen and Killackey, 1987). In the present study, two different techniques, PV-immunohistochemistry and anterograde axonal labeling, have provided indirect evidence for a modular organization of neuronal circuitry in rabbit AI. A patchy distribution of thalamocortical afferents labeled by PHA-L injections into MGV has also been noted in the auditory cortex of the rat (Romanski and LeDoux, 1993), the monkey (Hashikawa et aI., 1992), and ferret (Angelucci and Sur, 1993). These results suggest that a modular organization of neural circuits is a common feature of the auditory cortex in many mammals. A general consequence of a patchy organization is that different combinations of inputs are focused onto different neurons. Interestingly, afferent patches or bands have also been described in the central nucleus of the inferior colliculus following injections of anterograde tracers into brainstem nuclei belonging to the three general classes of ascending auditory brainstem pathways (Le. direct monaural pathways, indirect binaural pathways, and multisynaptic pathways) that terminate in the central nucleus (Oliver, 1984; 1987; Shneiderman et aI., 1988; Oliver and Shneiderman, 1991). In addition, afferents from the central nucleus terminate in focused bands in the MGV (Andersen et aI., 1980b) suggesting that the patches in AI represent the terminus of ascending parallel pathways originating in the auditory brainstem. The patches in the central nucleus and 165 the MGV may be related to both the tonotopic organization, and the monaural and binaural response properties of neurons within these nuclei (Oliver and Shneiderman, 1991). The functional significance of this organization, however, needs to be determined by more detailed studies of the auditory system that combine both anatomical and physiological observations at the cellular level. In experiments reported in Chapter 5, discrete injections of biocytin into MGV labeled thalamocortical afferents that terminated in patches in the auditory cortex of rabbits at early postnatal ages. The presence of distinct patches prior to the onset of cochlear function indicates that segregation of the axons occurs in the absence of acoustically driven cortical activity. Spontaneous activity in the auditory system, however, may playa role in the segregation process. Considerable evidence suggests that neural activity is essential for the formation of clustered horizontal connections, although it may not be required for the formation of radial connections within the cerebral cortex. Developmental studies of the visual system have shown that the adult patterns of thalamocortical patches arise from an initially diffuse projection that is refined during maturation by retraction of exuberant axonal growth (Stryker and Harris, 1986; Ghosh and Shatz, 1992). Retinal-activity blockade during a critical period early in development prevents the horizontal segregation of lateral geniculate axons into ocular dominance patches by disrupting activity-dependent synaptic competition between geniculocortical axons (Shatz, 1990; Goodman and Shatz, 1993). Binocular blockade of retinal activity, however, does not prevent LGN axons form extending into and arborizing within their appropriate target layers in visual cortex 166 (Stryker and Harris, 1986; Antonini and Stryker, 1993). The terminal arbors of lateral geniculate axons in binocularly deprived animals are similar in complexity and extent to those of normal animals, and can have a patchy organization although eye-specific columnar segregation is not evident (Antonini and Stryker, 1993). These results indicate that patterned neural activity is required for the formation of precise patterns t.1J.alamocortical afferent connections. Axonal segregation, however, can also be facilitated by other factors such as cell surface, extracellular matrix, and diffusable molecules (Goodman and Shatz,1993; Bolz et aI., 1993). This is illustrated by recent studies of thalamocortical axon development in the mouse somatosensory "barrel" cortex. In contrast to the visual geniculocortical system, there appears to be a high degree of topological precision in the development of the neonatal thalamocortical projection: axons segregate into discrete clusters upon entering layer IV and new branches are then added within their appropriate clusters (Agmon et aI., 1993; Agmon et aI., 1994). Furthermore, infraorbital nerve blockade from birth does not disrupt vibrissa related pattern formation in the somatosensory brainstem, thalamus, or barrel cortex, indicating that axonal segregation can occur in the absence of peripherally driven activity (Chiaia et aI., 1992; Henderson et aI., 1992; Schlaggar et aI., 1993). Blocking axonal transport, however, does disrupt formation of the vibrissae-related patterns in the central nervous system suggesting that axon-axon recognition may be mediated by a chemical signal that is anterogradely transported along axons and released from their terminals (Chiaia et aI., 1993). In addition, developing thalamocortical axons may be constrained to barrel hollows by extracellular matrix molecules until synaptic stabilization occurs (Steindler 167 et aI., 1989; 1990). Several features of the immature auditory thalamocortical patches labeled in the present study indicate exuberant growth during early development and subsequent axonal remodeling during postnatal maturaton. For example, the radial extent of the terminal fields of the thalamocortical axons, and the projection to layer I is far greater in neonates than in adults. In addition, axons that course tangentially in layer I contribute numerous descending branches to neonatal patches, whereas layer I axons in adults are relatively unbranched and extend preferentially within layer I. The exuberant growth of dendrites on neurons that are cortical targets of the thalamic afferent may be related to the exuberant growth of the immature axons (McMullen and Glaser, 1988; McMullen et al. , 1988a;b). The significant axonal remodeling during postnatal development may be activity dependent. At the early postnatal ages examined in the present study, two striking features of the individual axons that were reconstructed are significant to the organization of the adult thalamocortical projection from MGV to AI. First, individual axons contributed numerous branches to more than one patch, suggesting that there may be a class of adult thalamocortical axons that have a divergent pattern of projection. Second, individual axons projected both to the midcortical layers and to layer I, suggesting that previous retrograde tracing studies of adult auditory thalamocortical connectivity may have incorrectly concluded that axons terminating in layer I and layers III/IV originate from different populations ofMGB neurons (Mitani, 1987; Hashikawa et aI., 1991). Ongoing developmental studies are attempting to determine whether these features represent a 168 transient growth phase of individual axons, or indicate the existence of multiple classes of thalamocortical axons arising from the MGV. 169 REFERENCES Adams, J.C. (1981) Heavy metal intensification of DAB-based HRP reaction product. J. Histochem. Cytochem. 29:775. Agmon, A., L.T. Yang, D.K. O'Dowd, and E.G. Jones (1993) Organized growth of thalamocortical axons from the deep tier of terminations into layer IV of mouse barrel cortex. J. Neurosci. 13:5365-5382. Agmon, A., L.T. Yang, E.G. Jones, and D.K. O'Dowd (1995) Topological precision in the thalamic projection to neonatal mouse barrel cortex. J. Neurosci. 15:549-561. Ahlsen, G., K. Grant, and S. Lindstrom (1982) Monosynaptic excitation of principal cells in the lateral geniculate nucleus by corticofugal fibers. Brain Res. 234:454-458. Aitkin, L. (1990) The Auditory Cortex. Structural and Functional Basis of Auditory Perception, Chapman and Hall, London, 146pp. Andersen, R.A., P.L. Knight, and M.M. Merzenich (1980a) The thalamocortical and corticothalamic connections of AI, All and the anterior auditory field (AFF) in the cat. Evidence for two largely segregated systems of connections. J. Compo Neurol. 194:663- 701. Andersen, R.A., G.L. Roth, L.M. Aitkin, and M.M. Merzenich (1980b) The efferent projections of the central nucleus and the pericentral nucleus of the inferior colliculus in the cat. J. Compo Neurol. 194:649-662. Andressen, C., I. Bliimcke, and M.R. Celio (1993) Calcium binding proteins: selective markers of nerve cells. Cell Tissue Res. 271:181-208. Angelucci, A., F. Clasca, and M. Sur (1993) Multiple cortical fields in the ferret defined by their architectonics and thalamocortical connections. Soc. Neurosci. Abst. 19: 1427. Angevine, J.B., Jr., and R.L. Sidman (1961) Autoradiographic study of cell migration during histogenesis of cerebral cortex in mouse. Nature. 192:766-768. Anggard, L. (1965) An electrophysiological study of the development of cochlear functions in the rabbit. Acta. Otolaryngol. (Stockh) Suppl. 203: 1-64. 170 Antonini, A., and M.P. Stryker (1993a) Development of individual geniculocortical arbors in cat striate cortex and effects of binocular impulse blockade. J. Neurosci. 13:3549-3573. Antonini, A., and M.P. Stryker (1993b) Rapid remodeling of axonal arbors in the visual cortex. Sciene 260:1819-1821. Arai, R, L. Winsky, M. Arai and D.M. Jacobowitz (1991) Immunohistochemical localization of calretinin in the rat hindbrain. J. Compo Neurol. 310:21-44.' Arnault, P. and M. Roger (1990) The ventral temporal cortex in the rat. Connections of secondary auditory areas Te2 and Te3. J. Compo Neurol. 302:110-123. Austin, C., and C.L. Cepko (1990) Cellular migration patterns in the developing mouse cerebral cortex. Development. 110:713. Baimbridge, K.O., M.R. Celio and J.H. Rogers (1992) Calcium-binding proteins in the nervous system. Trends Neurosci. 15:303-308. Bayer, S.A., and J. Altman (1991) Neocortical Development. New York: Raven Press. Benevento, C.A. and J.H. Fallon (1975) The ascending projections of the superior colliculus in the rhesus monkey (Macaca mu[atta). J. Compo Neurol. 160:339-362. Bennett-Clarke, C.A., N.L. Chiaia, M.F. Jacquin, and RW. Rhoades (1992) Parvalbumin and calbindin immunocytochemistry reveal functionally distinct cell groups and vibrissa-related patterns in the trigeminal brainstem complex of the adult rat. J. Compo Neurol. 320:323-338. BIiimcke, I., P.R Hof, J.H. Morrison, and M.R Celio (1990) Distribution of parvalbumin immunoreactivity in the visual cortex of old world monkeys and humans. J. Compo Neurol. 301:417-432. Bliimcke, I., P.R Hof, J.H. Morrison and M.R. Celio (1991) Parvalbumin in the monkey striate cortex: a quantitative immunoelectron-microscopy study. Brain Res. 554:237-243. 171 Bliimcke, I., E. Weruaga, S. Kasas, A.E. Hendrickson, and M.R. Celio (1994) Discrete reduction patterns of parvalbumin and calbindin D-28k immunoreactivity in the dorsal lateral geniculate nucleus and the striate cortex of adult macaque monkeys after monocular enucleation. Vis. Neurosci. 11:1-11. Bolz, J., M. Gotz, M. Hubener, and N. Novak (1993) Reconstruction cortical connections in a dish. TINS. 16:310-316. Brandner, S. and H. Redies (1990) The projection from medial geniculate field AI in the cat: Organization in the isofrequency dimension. J. Neurosci. 10:50-61. Braun, K., H. Scheich, M. Schachner, and C.W. Heizmann (1985a) Distribution of parvalbumin, cytochrome oxidase activity and [14C]2-deoxyglucose uptake in the brain of the zebra finch. I. Auditory and vocal motor systems. Cell Tissue Res. 240: 101-115. Braun, K., H. Scheich, M. Schachner, and C.W. Heizmann (1985b) Distribution of parvalbumin, cytochrome oxidase activity and [14C]2-deoxyglucose uptake in the brain of the zebra finch. II. Visual system. Cell Tissue Res. 240:101-115. Braun, K., H. Scheich, C.W. Heizmann and W. Hunziker (1991) Parvalbumin and calbindin D-28k immunoreactivity as developmental markers of auditory and vocal motor nuclei of the zebra finch. Neuroscience. 40:853-869. Brugge, J.F. and R.A. Reale (1985) Auditory Cortex. In A. Peters and E.G Jones (Eds) Cerebral Cortex, Vol 4, Plenum Press, New York, pp. 229-271. Bullier, J. and G.H. Henry (1979) Laminar distribution of first-order neurons and afferent terminals in cat striate cortex. J. Neurophysiol. 42:1271-1282. Burton, H. and E.G. Jones (1976) The posterior thalamic region and its cortical projection in new world and old world monkeys. J. Compo Neurol. 168:249-302. Caballero-Bieda, M., B. Fernandez, and L. Puelles (1991a) Acetylcholinesterase and NADH-diaphorase chemoarchitectonic subdivisions in the rabbit medial geniculate body. J. Chem. Neuroanat. 4:271-280. Caballero-Bieda, M., B. Fernandez, and L.Puelles (1991b) Comparative mapping of acetylcholinesterase and reduced nicotinamide adenine dinucleotide diaphorase in the rabbit dorsal thalamus. Acta. Anat. 140:224-235 172 Caird, D., H. Scheich, and R. Klinke (1991) Functional organization of auditory cortical fields in the Mongolian gerbil (Meriones unguiculatus): Binaural2-deoxyglucose patterns. J. Compo Physiol. (A) 168:13-26. Calford, M.B. (1983) The parcellation of the medial geniculate body of the cat defmed by the auditory response properties of single units. J. Neurosci. 3:2350-2364. Calford, M.B. and L.M. Aitkin (1983) Ascending projections to the medial geniculate body of the cat: evidence for mUltiple, parallel auditory pathways through the thalamus. J. Neurosci. 3:2365-2380. Campbell, M.J., and D.N. Pandya (1994) Medial geniculate body chemoarchitectonic subdivisions. Soc. Neurosci. Abstr. 20:977. Casseday, J.H., LT. Diamond, and J.K. Harting (1976) Auditory pathways to the cortex in Tupaia glis. J. Compo Neurol. 166:303-340. Catalano, S.M., R.T. Robertson, and H.P. Killackey (1991) Early ingrowth of thalamocortical afferents to the neocortex of the prenatal rat. Proc. Nat. Acad. Sci. 88:2999-3003. Celio, M.R. (1986) Parvalbumin in most gamma-aminobutyric acid-containing neurons of the rat cerebral cortex. Science. 231 :995-996. Celio, M.R. (1990) Calbindin D-28k and parvalbumin in the rat nervous system. Neuroscience. 35:375-475. Celio, M.R., and C.W. Heizmann (1981) Calcium-binding protein parvalbumin as a neuronal marker. Nature. 243:300-302. Celio, M.R., L. Scharer, J.H. Morrison, A.W. Norman, and F.E. Bloom (1986) Calbindin immunoreactivity alternates with cytochrome C oxidase-rich zones in some layers of the primate visual cortex. Nature. 323:715-717. Celio, M.R., W. Baier, L. Scharer, P.A. de Viragh, and C.H. Gerday (1988) Monoclonal antibodies directed against the calcium-binding protein parvalbumin. Cell Calcium. 9:81-86. 173 Celio, M.R, W. Baier, L. Scharer, H.J. Gregersen, P.A. de Viragh, and A.W. Norman (1990) Monoclonal antibodies directed against the calcium-binding protein calbindin D- 28k. Cell Calcium. 11 :599-602. Cellerino, A., R. Siciliano, L. Domenici, and L. Maffei (1992) Parvalbumin immunoreactivity: A reliable marker for the effects of monocular deprivation in the rat visual cortex. Neuroscience. 51:749-753. Chard, P.S., D. Bleakman, S. Christakos, C.S. Fullmer and RJ. Miller (1993) Calcium buffering properties of calbindin D28k and parvalbumin in rat sensory neurones. J. Physiol. 472:341-357. Chiaia, N.L., S.E. Fish, W.R. Bauer, C.A. Bennett-Clarke, and RW. Rhoades (1992) Postnatal blockade of cortical activity by tetrodotoxin does not disrupt the formation of vibrissa-related patterns in the rat's somatosensory cortex. Dev. Brain Res. 66:244-250. Chiaia, N.L., R.S. Crissman, C.A. Bennet-Clarke, and RW. Rhoades (1993) Axonal transport blockade mimics effects of neonatal infra orbital nerve transection upon vibrissa-related patterns. Soc. Neurosci. Abstr. 19:1107. Clarey, J.C., P. Barone, and T.J. Imig (1992) Physiology of thalamus and cortex. In A.N. Popper and RR. Fay (Eds) The Mammalian Auditory Pathway: Neurophysiology, Springer-Verlag, New York, pp 232-334. Clerici, W.J., A.J. McDonald, R Thompson, and J.R Coleman (1990) Anatomy of the rat medial geniculate body: II. Dendritic morphology. J. Compo Neurol. 297:32-54. Clerici, W.J., and J.R. Coleman (1990) Anatomy of the rat medial geniculate body: I. Cytoarchitecture, myeloarchitecture, and neocortical connectivity. J. Compo Neurol. 297:14-31. Coleman, P.D., and A.H. Rieser (1968) Environmental effects on cortical dendritic fields. I. Rearing in the dark. J. Anat. 102:363-374. Colwell, S.A. and M.M. Merzenich (1975) Organization of thalamocortical and corticothalamic projections to and from physiologically defined loci within primary auditory cortex in cat. Anat. Rec. 181:336. 174 Demeulemeester, H., L. Arckens, F. Vandesande, G.A. Orban, C.W. Heizmann and R Pochet (1991) Calcium binding proteins as molecular markers for cat geniculate neurons. Exp. Brain Res. 83:513-520. De Carlos, J.A., B.L. Schlaggar, and D.D.M. O'Leary (1991) Development of acetylcholinesterase-positive thalamocortical afferents to embryonic rat neocortex. Soc. Neurosci. Abstr. 17:764. DeFelipe, J., S.H.C. Hendry and E.G. Jones (1989) Visualization of chandelier cell axons by parvalbumin immunoreactivity in monkey cerebral cortex. Proc. Natl. Acad. Sci., U.S.A. 86:2093-2097. DeFelipe, J. and E.G. Jones (1991) Parvalbumin immunoreactivity reveals layer IV of monkey cerebral cortex as a mosaic of microzones of thalamic afferent terminations. Brain Res. 562:39-47. de Venecia, R.K. and N.T. McMullen (1991) Prenatal and early postnatal development of auditory thalamocortical axons. Soc. Neurosci. Abstr. 17:451. de Venecia, R.K. and N.T. McMullen (1992) Postnatal development of thalamocortical afferent "patches" in auditory neocortex. Soc. Neurosci. Abstr. 18:779. de Venecia, R.K. and N.T. McMullen (1994) Single thalamocortical axons diverge to multiple patches in neonatal auditory cortex. Dev. Brain Res. 81:135-142. de Venecia, RK., C.B. Smelser, S.D. Lossmann and N.T. McMullen (1995a) Complementary expression of parvalbumin and calbindin D-28k delineates subdivisions of the rabbit medial geniculate body. Submitted to Journal of Comparative Neurology. de Venecia, RK., C.B. Smelser and N.T. McMullen (1995b) Parvalbumin is expressed in a reciprocal pathway linking the medial geniculate body and auditory cortex. Soc. Neurosci. Abstr. 21. Diamond, I.T. (1983) Parallel pathways in the auditory, visual and somatosensory systems. In G. Macchi et al. (Ed): Somatosensory Integration in the Thalamus. Amsterdam: Elsevier, 251-272. 175 Diamond, I.T., D. Fitzpatrick, and D. Schmechel (1993) Calcium binding proteins distinguish large and small cells of the ventral posterior and lateral geniculate nuclei of the prosimian galago and the tree shrew (Tupaia belangeri). Proc. Natl. Acad. Sci., USA. 90:1425-1429. Erzurumlu, R.S. and S. Jhaveri (1990) Thalamic axons confer a blueprint of the sensory periphery onto the developing rat somatosensory cortex. Dev. Brain Res. 56:229-234. Fernandez, V., and H. Bravo (1974) Autoradiographic study of development of cerebral cortex in the rabbit. Brain, Behav., Evol. 9:317-332. Ferster, D. and S. LeVay (1978) The axonal arborizations of lateral geniculate neurons in the striate cortex of the cat. J. Compo Neurol. 182:923-944. Fitzpatrick, K. and T J. Imig (1978) Projections of auditory cortex upon the thalamus and midbrain in the owl monkey. J. Compo Neurol. 177:537-556. Florence, S.L., and V.A. Casagrande (1990) Development of geniculocortical axon arbors in a primate. Visual Neurosci. 5:291-309. Foss, I, and G. Flottorp (1974) A comparative study of the development of hearing and vision in various species commonly used in experiments. Acta. Otolaryngol. (Stockh.) 77:202-214. Freund, T.F., Z. Magloczky, I. Soltesz, and P. Somogyi (1986) Synaptic connections, axonal and dendritic patterns of neurons immunoreactive for cholecystokinenin the visual cortex of the cat. Neuroscience. 19:1133-1159. Friauf, E. (1994) Distribution of calcium-binding protein calbindin-D28k in the auditory system of adult and developing rats. J. Compo Neurol. 349:193-211. Frost, D.O. and V.S. Caviness (1980) Radial organization of thalamic projections to the neocortex in the mouse. J. Compo Neurol. 194:369-393. Geeraedts, L.M.G. (1978) An experimental neuro-anatomical study of the primary optic pathways and the descending supra-optic pathways in the rabbit. PhD. Thesis. Catholic University of Nijmegen. 176 Gerfen, C.R. and P.E. Sawchenko (1984) An anterograde neuroanatomical tracing method that shows the detailed morphology of neurons, their axons and terminals: Immunohistochemical localization of an axonally transported plant lectin. Phaseolus vulgaris Leucoagglutin (PHA-L). Brain Res. 290:219-238. Ghosh, A. and C.J. Shatz (1992) Pathfmding and target selection by developing geniculocortical axons. J. Neurosci. 12:39-55. Gilbert, C.D. and J.P. Kelly (1975) The projections of cells in different layers of the cat's visual cortex. J. Compo Neurol. 163:81-106. Gilbert, C.D. (1977) Laminar differences in receptive field properties of cells in cat primary visual cortex. J. Physiol.(London) 268:391-421. Gilbert, C.D. and T.N. Wiesel (1979) Morphology and intracortical projections of functionally characterised neurones in the cat visual cortex. Nature 280: 120-125. Gilbert, C.D. and T.N. Wiesel (1983) Clustered intrinsic connections in cat visual cortex. J. Neurosci. 3:1116-1133. Giolli, R.A. and M.D. Guthrie (1971) Organization of subcortical projections of visual areas I and II in the rabbit. An experimental degeneration study. J. Compo Neurol. 142:351-376. Glaser, E.M., H. Van der Loos, and M.Gissler (1979) Tangential orientation and spatial order in dendrites of cat auditory cortex: A computer microscope study of Golgi impregnated material. Exp. Brain Res. 36:411-431. Glaser, E.M. and N.T. McMullen (1980) Tonotopic organization and dendrite orientation in primary auditory cortex of the rabbit. Soc. Neurosci. Abstr. 6:557. Glaser, E.M., M. Tagamets, N.T. McMullen, and H. Van der Loos (1983) The image combining microscope: An interactive instrument for morphometry of the nervous system. J. Neurosci. Meth. 8: 17-32. Glaser, E.M., and N.T. McMullen (1984) The fan-in projection method for analyzing dendrite and axon systems. J .. Neurosci. Meth. 12:37-42. 177 Goodman, C.S. and C.J. Shatz (1993) Developmental mechanisms that generate precise patterns of neuronal connectivity. Neuron (Suppl.) 10:77-98. Graham, J. (1977) An autoradiographic study of the efferent connections of the superior colliculus in the cat. J. Compo Neurol. 173:629-654. Hallman, L.E., B.R Scholfield, and C.S. Lin (1988) Dendritic morphology and axon collaterals of corticotectal, corticopontine, and callosal neurons in layer V of primary visual cortex of the hooded rat. J. Compo Neurol. 272:149-160. Harris, RM., and T.A. Woolsey (1981) Dendritic plasticity in mouse barrel cortex following postnatal vibrissa follicle damage. J. Compo Neurol. 196:357-376. Harvey, A.R (1978) Characteristics of corticothalamic neurons in area 17 of the cat. Neurosci. Let. 7: 177-181. Hashikawa, T., E. Rausell, M. Molinari, and E.G. Jones (1991) Parvalbumin-and calbindin-containing neurons in the monkey medial geniculate complex: Differential distribution and cortical lamina specific projections. Brain Res. 544:335-341. Hashikawa, T., H. Ojima, and E.G. Jones (1992) Differential cortical projections from subnuclei of monkey medial geniculate complex revealed by anterograde transport of PHA-L. Soc. Neurosci. Abstr. 18:1038. Hashikawa, T., M. Molinari, E. Rausell, M.E. Dell ' Anna, M.G. Leggio, and E.G. Jones (1994) The macaque monkey auditory thalamocortical system defmed by calcium binding protein immunoreactivity. Soc. Neurosci. Abst. 20:978. Heath, C.G. and E.G. Jones (1971) An experimental study of the ascending connections from the posterior group of thalamic nuclei in the cat. J. Compo Neurol. 141:397-426. Heizmann, C.W. (1984) Parvalbumin, an intracellular calcium-binding protein; distribution, properties and possible roles in mammalian cells. Experientia.40:910-921. Henderson, T.A., T.A. Woolsey, and M.F. Jacquin (1992) Infraorbital nerve blockade does not disrupt central trigeminal pattern formation in the rat. Dev. Brain Res. 66: 146- 152. 178 Hendrickson, A.E., J.F.M. van Brederode, K.A. Mulligan, and M.R. Celio (1991) Development of calcium-binding proteins parvalbumin and calbindin in monkey striate cortex. J. Compo Neurol. 307:626-646. Hendry, S.H.C. and E.G. Jones (1983a) The organization of pyramidal and non pyramidal cell dendrites in relation to thalamic afferent terminations in the monkey somatic sensory cortex. J. Neurocytol. 12:227-298 Hendry, S.H.C. and E.G. Jones (1983b) Thalamic inputs to identified commissural neurons in the monkey somatic sensory cortex. J. Neurocytol. 12:299-316. Hendry, S.H.C., E.G. Jones, P.C. Emson, D.E.M. Lawson, C.W. Heizman and P. Streit (1989) Two classes of cortical GABA neurons defined by differential calcium binding protein immunoreactivities. Exp. Brain Res. 76:467-472. Hendry, S.H.C. and E.G. Jones (1991) GABA neuronal subpopulations in cat primary auditory cortex: Co-localization with calcium binding proteins. Brain Res. 543:45-55. Hersch, S.M. and E.L. White (1981) Thalamocortical synapses with corticothalamic projection neurons in mouse SmI Cortex: electron microscopic demonstration of a monosynaptic feedback loop. Neurosci. Let. 24:207-210. Hersch, S.M. and E.L. White (1981) Thalamocortical synapses involving identified neurons in mouse somatosensory cortex: a terminal degeneration and Golgi/EM study. J. Comp Neurol. 195:253-263. Heumann, D., and T. Rabinowicz (1982) Postnatal development of the visual cortex of the mouse after enucleation at birth. Exp. Brain Res. 46:99-106. Holstege, G. and Collewijn (1982) The efferent connections of the nucleus of the optic tract and the superior colliculus in the rabbit. J. Compo Neurol. 209:139-175. Hubel, D.H., and T.N. Wiesel (1969) Anatomical demonstration of columns in the monkey striate cortex. Nature. 221:747-750. Hubel, D.H. and T.N. Wiesel (1972) Laminar and columnar distribution of geniculocortical fibers in the macaque monkey. J. Compo Neurol. 146:421-450. Hubel, D.H. and T.N. Wiesel (1977) Functional architecture of macaque monkey visual cortex. Proc. Roy. Soc. Lond. 198:1-59. 179 Humphrey, A.L., M. Sur, D.J. Uhlrich, and S.M. Sherman (1985) Projection patterns of individual X-and Y-cell axon from the lateral geniculate nucleus to cortical area 17 in the cat. J. Compo Neurol. 233:159-189. Hutchins, J.B., and V.A. Casagrande (1990) Development of the lateral geniculate nucleus: interactions between retinal afferent, cytoarchitectonic, and glial cell process lamination in ferrets and tree shrews. J. Compo Neurol. 298:113-128. Imig, T.J. and H.O. Adrian (1977) Binaural columns in the primary field (AI) of cat auditory cortex. Brain Res. 138:241-257. Imig, T.J. and J.F. Brugge (1978) Sources and terminations of callosal axons related to Binaural and frequency maps in primary auditory cortex of the cat. J. Compo Neurol. 182:637-660. Imig, T.J. and R.A. Reale (1980) Patterns of cortico-cortico connections related to tonotopic maps in cat auditory cortex. J. Compo Neurol. 192:293-332. Imig, T.J. and R.A. Reale (1981) Ipsilateral corticocortical projections to binaural columns in cat primary auditory cortex. J. Compo Neurol. 203:1-14. Imig, T.J. and A. Morel (1983) Organization of the thalamocortical auditory system in the cat. Ann. Rev. Neurosci. 6:95-120. Imig, T.J. and A. Morel (1984) Topographic and cytoarchitectonic organization of thalamic neurons related to their targets in low-, middle-, and high-frequency representations in cat auditory cortex. J. Compo Neurol. 227:511-539. Imig, T.J. and A. Morel (1985) Tonotopic organization in ventral nucleus of medial geniculate body in the cat. J. Neurophysiol. 53:309-340. Imig, T.J. and A. Morel (1988) Organization of the cat's auditory thalamus. In G.H. Edelman, W.E. Gall and W.M. Cowan (Eds) Auditory Function: Neurobiological Basis of Hearing, Wiley, New York, pp. 457-484. Jensen, K.F. and H.P. Killackey (1987) Terminal arbors ofaxons projecting to the somatosensory cortex of the adult rat: I. The normal morphology of specific thalamocortical afferents. J. Neurosci. 7:3529-3543. 180 Jacobson, S. and J.Q. Trojanowski (1974) The cells of origin of the corpus callosum in the rat, cat, and rhesus monkey. Brain Res. 74:149-155. Jacobson, S. and J.Q. Trojanowski (1975) Corticothalamic neurons and thalamocortical terminal fields: an investigation in rat using horseradish peroxidase and autoradiography. Brain Res. 85:385-401. Jiang, X., R.R. Johnson and A. Burkhalter (1993) Visualization of dendritic morphology of cortical projection neurons by retrograde axonal tracing. J. Neurosci. Meth. 50:45- 60. Jones, E.G. (1975) Lamination and differential distribution of thalamic afferents within the sensory-motor cortex of the squirrel monkey. J. Compo Neurol. 160:167-204. Jones, E.G. (1984) Laminar distribution of cortical efferent cells. In A. Peters and E.G. Jones (Eds.), Cerebral Cortex, Vol. 1. New York: Plenum, pp. 521-553. Jones, E.G. (1985) The Thalamus. New York: Plenum Press. Jones, E.G. and H. Burton (1976) Areal differences in the laminar distribution of thalamic afferents in cortical fields of the insular, parietal and temporal regions of primates. J. Compo Neurol. 168:197-248. Jones, E.G. and S.P. Wise (1979) Size, laminar and columnar distribution of efferent cells in the sensory-motor cortex of monkeys. J. Compo Neurol. 175:391-438. Jones, E.G., and S.H.C. Hendry (1984) Basket Cells. In A. Peters and E.G. Jones (Eds.) Cerebral Cortex. Volume 1. Cellular Components of the Cerebral Cortex. New York: Plenum Press, pp309-336. Jones, E.G., and S.H.C. Hendry (1989) Differential calcium binding protein immunoreactivity distinguishes classes of relay neurons in monkey thalamic nuclei. Eur. J. Neurosci. 1:222-246. Kageyama, G.H., M. Beilstein, L. Hoang, S. Catalano, J. Yu, and R.T. Robertson Geniculocortical axon ingrowth and morphogenesis of cortical neurons in rat auditory and visual cortices. Soc. Neurosci. Abst. 18:778. 181 Kato, N., S. Kawaguchi, and H. Miyata (1984) Geniculocortical projection to layer I of area 17 in kittens: Orthograde and retrograde HRP studies. J. Compo Neurol. 225:441- 447. Katz, L.C. (1987) Local circuitry of identified projection neurons in cat visual cortex brain slices. J. Neurosci. 7: 1223-1249. Kawaguchi, Y., H. Katsumaru, T. Kosaka, C.W. Heizmann and K. Hama (1987) Fast spiking cells in rat hippocampus (CA 1 regions) contain the calcium binding protein parvalbumin. Brain Res. 416:369-374. Kawaguchi, Y. and K. Kubota (1993) Correlation of physiological subgroupings of nonpyramidal cells with parvalbumin- and calbindin D28k-immunoreactive neurons in layer V of rat frontal cortex. J. Neurophysiol. 70:387-396. Kelley, P.E., RD. Frisina, M.L. Zettel and J.P. Walton (1992) Differential calbindin like immunoreactivity in the brain stem auditory system of the chinchilla. J. Compo Neurol. 319: 196-212. Kelly J.P., and D. Wong (1981) Laminar connections of the cat's auditory cortex. Brain Res. 212:1-15. Kelly, J.B. and S.L. Sally (1988) Organization of auditory cortex in the albino rat: Binaural response properties. J. Neurophysiol. 9:1756-1769. Killackey, H.P., G. Belford, R Ryugo, and D.K. Ryugo (1976) Anomalous organization of thalamocortical projections consequent to vibrissae removal in the newborn rat and mouse. Brain Res. 104:309-315. King, M.A., P.M. Lewis, B.E. Hunter, and O.W. Walker (1989) Biocytin: a versatile anterograde neuroanatomical tract-tracing alternative. Brain Res. 497:361-367. Kisvarday, Z.F., K.A.C. Martin, M.J. Friedlander, and P. Somogyi (1987) Evidence for interlaminar inhibitory circuits in the striate cortex of the cat. J. Compo Neurol. 260:1-19. Klyvania, M.P. and G.A. Obrastova (1968) The influence of sound stimulus intensity on the time and amplitude characteristics of the cerebral cortex of the rabbit. In: L. Jilek and S. Trojan (Eds.) Ontogenesis of the Brain, Charles University Press, Prague, pp. 395-400. 182 Konig, N. and R. Marty (1974) On functions and structure of deep layers of immature auditory cortex. J. Physiol. (paris) 68:145-155. Kudo, M., and K. Niimi (1978) Ascending projections of the inferior colliculus onto the medial geniculate body in the cat studied by anterograde and retrograde tracing techniques. Brain Res. 155:113-117. Kudo, M., and K. Niimi (1980) Ascending projections of the inferior colliculus in the cat: an autoradiographic study. J. Compo Neurol. 191:545-556. Lachica, E.A. and V.A. Casagrande (1992) Direct W-like geniculate projections to cytochrome oxidase (CO) blobs in primate visual cortex: axon morphology. J. Compo Neurol. 319:141-158. Landry, P., J. Villemure, and M. Deschenes (1982) Geometry and orientation of thalamocortical arborizations in the cat somatosensory cortex as revealed by computer reconstruction. Brain Res. 237:222-226. LeDoux, J.E., D.A. Ruggiero, and D.J. Reis (1985) Projections to the subcortical forebrain from anatomically defmed regions of the medial geniculate body in the rat. J. Compo Neurol. 242:182-213. LeDoux, J.E., D.A. Ruggiero, R. Forest, S. Stornetta, and D.J. Reis (1987) Topographic organization of convergent projections to the thalamus from the inferior colliculus and spinal cord in the rat. J. Compo Neurol. 264: 123-146. LeVay, S. and C.D. Gilbert (1976) Laminar patterns of geniculocortical projection in the cat. Brain Res. 113:1-19. LeVay, S., M.P. Stryker, and C.J. Shatz (1978) Ocular dominance columns and their development in lamina IV of the cat's visual cortex: a quantitative study. J. Compo Neurol. 179:223-244. LeVay, S., T.N. Wiesel, and D.H. Hubel (1980) The development of ocular dominance columns in normal and visually deprived monkeys. J. Compo Neurol. 191:1-51. Lewis, D.A., and J.S. Lund (1990) Heterogeneity of chandelier neurons in monkey neocortex: corticotropin-releasing factor and parvalbumin immunoreactive populations. J. Compo Neurol. 293:599-615. 183 Livingstone, M.S., and D.H. Hubel (1984) Anatomy and physiology of a color system in the primate visual cortex. J. Neurosci. 4:309-356. Livingstone, M.S., and D.H. Hubel (1987) Psychophysical evidence for separate channels for the perception of form, color, movement and depth. J. Neurosci. 7:3416- 3468. Lorente de No, R. (1922) La corteza cerebral del raton (primera contribucion-Ia corteza acustica). Trab. Labor. Invest. BioI. 20:41-78. Love, J.A. and J.W. Scott (1969) Some response characteristics of cells of the magnocellular division of the medial geniculate body of the cat. Can. J. Physiol. Pharmacol. 47:881-888. Lund, J.S. (1973) Organization of neurons in the visual cortex, area 17, of the monkey (Macaca mulatta). J. Compo Neurol. 147:455-496. Lund, J.S. and R.G. Boothe (1975) Interlaminar connections and pyramidal neuron organization in the visual cortex, area 17, of the Macaque monkey. J. Compo Neurol. 159:305-334. Lund, J.S., R.D. Lund, A.E. Hendrikson, A.H. Bunt and A.F. Fuchs (1975) The origin of the efferent pathways from the visual cortex, area 17, of the macaque monkey as shown by retrograde transport of horseradish peroxidase. J. Compo Neurol. 164:287- 303. Lund, R.D., and M.J. Mustari (1977) Development of the geniculocortical pathway in rats. J. Compo Neurol. 173:289-306. Lund, J.S., G.H. Henry, C.L. MacQueen and A.R. Harvey (1979) Anatomical organization of the primary visual cortex (area 17) of the cat. A comparison with area 17 of the Macaque monkey. J. Compo Neurol. 184:599-618. Lund, J.S., and D.A. Lewis (1993) Local circuit neurons of developing and mature macaque prefrontal cortex: Golgi and immunocytochemical characteristics. J. Compo Neurol. 328:282-312. Luskin, M.B., A.L. Perlman, and J.R. Sanes (1988) Cell lineage in the cerebral cortex of the mouse studied in vivo and in vitro with a recombinant retrovirus. Neuron. 1 :635- 647. 184 Majorossy, K. and M. Rethelyi (1968) Synaptic architecture in the medial geniculate body (ventral division). Exp. Brain Res. 6:306-323. Martin, K.A.C. (1984) Neuronal circuits in cat striate cortex. In E.G. Jones and A. Peters: Cerebral Cortex Volume 2, Functional Properties of Cortical Cells. Plenum Press, New York. pp 241-284. McConnell, S.K. (1985) Migration and differentiation of cerebral cortical neurons after transplantation into the brains of ferrets. Science. 229: 1268-1271. McConnell, S.K. (1988) Fates of visual cortical neurons in the ferret after isochronic and heterochronic transplantation. J. Neurosci. 8:945-974. McGuire, B.A., J.P. Hornung, C.D. Gilbert, and T.N. Wiesel (1984) Patterns of synaptic input to layer 4 of the cat striate cortex. J. Neurosci. 4:3021-3033. McMullen, N. T. and E.M. Glaser (1982a) Tonotopic organization of the rabbit auditory cortex. Exp. Neurol. 75:208-220. McMullen, N.T. and E.M. Glaser (1982b) Morphology and laminar distribution of nonpyramidal neurons in the auditory cortex of the rabbit. J. Comp. Neurol. 208: 186- 204. McMullen, N.T., E.M. Glaser, and M. Tagamets (1984) Morphometry of spine-free nonpyramidal neurons in rabbit auditory cortex. J. Compo Neurol. 222:383-395. McMullen, N.T. and E.M. Glaser (1985) The organization of thalamocortical afferents in rabbit auditory cortex. Anat. Rec. 211:119A. McMullen, N.T. and E.M. Glaser (1986) Terminal arborization of afferents to auditory cortex: a comparison of thalamocortical and callosal axons using the PHA-L method. Soc. Neurosci. Abstr. 12:1275. McMullen, N. T., and A. M. Prescott (1987) Light microscopic evidence for thalamocortical input to GABA-ergic neurons in auditory neocortex: Combination of PHA-L and target neuron immunocytochemistry. Soc. Neurosci. Abstr. 13:326. McMullen, N. T. and E. M. Glaser (1988) Auditory cortical responses to neonatal deafening: Pyramidal neuron spine loss without changes in growth or orientation. Exp. Brain Res. 79: 175-200. 185 McMullen, N.T., B. Goldberger, and E.M. Glaser (1988) The postnatal development of lamina IIIIIV nonpyramidal neurons in rabbit auditory cortex: Quantitative and spatial analyses of Golgi-impregnated material. J. Compo Neurol. 278:139-155. McMullen, N.T., B. Goldberger, C.M. Suter, and E.M. Glaser (1988) Neonatal deafening alters nonpyramidal dendrite orientation in auditory cortex: a computer microscope study in the rabbit. J. Compo Neurol. 267:92-106. McMullen, N.T. and RK. de Venecia (1993) Thalamocortical patches in auditory neocortex. Brain Res. 620:317-322. McMullen, N.T., C.B. Smelser, and RK. de Venecia (1994a) A quantitative analysis of parvalbumin neurons in rabbit auditory cortex. J. Compo Neurol. 349:493-511. McMullen, N.T., R.K. de Venecia, C.B. Smelser, and S.D. Lossman (1994b) Calcium binding protein expression delineates subdivisions of rabbit medial geniculate complex. Soc. Neurosci. Abst. 20:977. McMullen, N.T., C.B. Smelser and F.L.Rice (1994c) Parvalbumin expression reveals a vibrissa-related pattern in rabbit SI cortex. Brain Research. 660:225-231. Merzenich, M.M., S.A. Colwell, and RA. Andersen (1982) Thalamocortical and corticothalamic connections in the auditory system of the cat. In C.N. Woolsey (Ed) Cortical Sensory Organization Vol.lII, Humana, Clifton, NJ, pp. 43-57. Merzenich, M.M., W.M. Jenkins, and J.C. Middlebrooks (1984) Observations and hypotheses on special organizational features of the central auditory nervous system, In G.W. Edelman, W.E. Gall and W.M. Cowan (Eds) Dynamic Aspects of Neocortical Function, Wiley, New York, pp 397-424. Mesulam, M.M. and D.N. Pandya (1973) The projections of the medial geniculate complex within the sylvian fissure of the rhesus monkey. Brain Res. 60:315-333. Metin, C., and D.O. Frost (1989) Visual responses of neurons in somatosensory cortex of hamsters with experimentally induced retinal projections to somatosensory thalamus. Proc. Natl. Acad. Sci. USA. 86:357-361. Middlebrooks, J.C., RW. Dykes, and M.M. Merzenich (1980) Binaural response specific bands in primary auditory cortex (AI) of the cat: topographical organization orthogonal to isofrequency contours. Brain Res. 181:31-48. 186 Middlebrooks, J.C. and J.M. Zook (1983) Intrinsic organization of the cat's medial geniculate body identified by projections to binaural response-specific bands in the primary auditory cortex. J. Neurosci. 3:203-224. Mitani, A., and M. Shimokouchi (1985) Neuronal connections in the primary auditory cortex: An electrophysiological study in the cat. J. Compo Neurol. 235:417-429. Mitani, A., M. Shimokouchi, K. Itoh, S. Nomura, M. Kudo, and N. Mizuno (1985) Morphology and laminar organization of electrophysiologically identified neurons in the primary auditory cortex in the cat. J. Compo Neurol. 235:430-447. Montero, V.M. (1983) Ultrastructural identification of axon terminals from the thalamic reticular nucleus in the medial geniculate body in the rat: A EM autoradiographic study. Exp. Brain Res. 51:338-342. Moore, R.Y. and J.M. Goldberg (1966) Projections of the inferior colliculus in the monkey. Exp. Neurol. 14:429-438. Morel A., and T.J. Imig (1987) Thalamic projections to A, AI, P and VP in the cat auditory cortex. J. Compo Neurol. 265:119-144. Morel, A. and J.H. Kaas (1992) Subdivisions and connections of auditory cortex in owl monkeys. J. Compo Neurol. 318:27-63. Morest, D.K. (1964) The neuronal architecture of the medial geniculate body of the cat. J. Anat. (Lond.) 98:611-630. Morest, D.K. (1965a) The laminar structure of the medial geniculate body of the cat. J. Anat. (Lond.) 99:143-160. Morest, D.K. (1965b) The lateral tegmental system of the midbrain and the medial geniculate body: study with Golgi and Nauta methods in cat. J. Anat. 99:611-634. Morest, D.K. (1975) Synaptic relationship of Golgi type II cells in the medial geniculate body of the cat. J. Compo Neurol. 162:157-194. Morest, D.K. and J.A. Winer (1986) The comparative anatomy of neurons: Homologous neurons in the medial geniculate body of the opossum and the cat. Anat. Embryol. Cell BioI. 97: 1-96. 187 Morino-Wannier, P., S.C. Fujita, and E.G. Jones (1992) GABAergic neuronal populations in monkey primary auditory cortex dermed by co-localized calcium binding proteins and surface antigens. Exp. Brain Res. 88:422-432. Naegele, J.R., S. Jhaveri, and G.E. Schneider (1988) Sharpening of topographical projections and maturation of geniculocortical axon arbors in the hamster. J. Compo Neurol. 277:593-607. Niimi, K. and F. Naito (1974) Cortical projections of the medial geniculate body in the cat. Exp. Brain Res. 19:326-342. O'Leary, D.D.M., and B.B. Stanfield (1989) Selective elimination ofaxons extended by developing cortical neurons is dependent on regional locale: experiments utilizing fetal cortical transplants. J. Neurosci. 9:2230-2246. O'Leary, D.D.M. and S.E. Koester (1993) Development of projection neuron types, axon pathways, and patterned connections of the mammalian cerebral cortex. Neuron. 10:991-1006. Oliver, D.L. and W.C. Hall (1975) Subdivisions of the medial geniculate body in the tree shrew (Tupaia g?is). Brain Res. 86:217-227. Oliver, D.L. and W.C. Hall (1978a) The medial geniculate body of the tree shrew, Tupaia glis. I: Cytoarchitecture and midbrain connections. J. Compo Neurol. 182:423- 458. Oliver, D.L. and W.C. Hall (1978b) The medial geniculate body of the tree shrew, Tupaia glis. II: Connections with the neocortex. 1. Compo Neurol. 182:459-494. Oliver, D.L. (1982) A Golgi study of the medial geniculate body of the tree shrew (Tupaia glis). J. Compo Neurol. 209:1-16. Oliver, D.L. (1984) Dorsal cochlear nucleus projections to the inferior colliculus in the cat: a light and electron microscope study. J. Compo Neurol. 224:155-172. Oliver, D.L. (1987) Projections to the inferior colliculus from the anteroventral cochlear nucleus in the cat: possible substrates for binaural interaction. J. Compo Neurol. 264:24-46. 188 Oliver, D.L. and A. Shneidennan (1991) The anatomy of the inferior colliculus: a cellular basis for integration of monaural and binaural infonnation. In R.A. Altschuler, R.P. Bobbin, B.M. Clopton and D.W. Hoffman (Eds) Neurobiology of Hearing: The Central Auditory System. Raven Press, Ltd., New York, pp. 293-333. Paxinos, G. and C. Watson (1986) The Rat Brain in Stereotaxic Coordinates. Academic Press, New York. Peters, A. and I. Saldanha (1976) The projection of the lateral geniculate nucleus to area 17 of the rat cerebral cortex. III. Layer VI. Brain Res. 105:533-537. Peters, A., C.C. Proskauer, M.L. Feldman, and L. Kimerer (1979) The projection of the lateral geniculate nucleus to area 17 of the rat cerebral cortex. V. Degenerating axon tenninals synapsing with Golgi impregnated neurons. I. Neurocytol. 8:331-357. Peters, A., C.C. Proskauer, and C.E. Ribak (1982) Chandelier cells in rat visual cortex. I. Compo Neurol. 206:397-416. Phillips, D.P., R.A. Reale, and I.F. Brugge (1991) Stimulus processing in the auditory cortex, In R.A. Altschuler et aI, (Eds) Neurobiology of Hearing: The Central Auditory System, Raven, New York, pp. 335-365. Puelles, L., M.P. Sanchez, R. Spreafico, and A. Fairen (1992) Prenatal development of calbindin immunoreactivity in the dorsal thalamus of the rat. Neuroscience 46:135-147. Rakic, P. (1972) Mode of cell migration to the superficial layers of monkey neocortex. I. Compo Neurol. 145:61-84. Rakic, P. (1974) Neurons in rhesus monkey visual cortex: systematic relation between time of origin and eventual disposition. Science. 183:425-426. Rakic, P. (1976) Prenatal genesis of connections subserving ocular dominance in the rhesus monkey. Nature 261:467-471. Rakic, P. (1977) Prenatal development of the visual system in the rhesus monkey. Phil. Trans. R. Soc. London (BioI.) 278:245-260. Rakic, P. (1988) Specification of cerebral cortical areas. Science. 241:170-176. 189 Ramon y Cajal, S. (1966) Studies on the Diencephalon. Translated by E. Ramon Moliner. Springfield, IL: Charles C. Thomas, Publisher. Rausell, E. and E.G. Jones (1991a) Chemically distinct compartments of the thalamic VPM nucleus in monkeys relay principal and spinal trigeminal pathways to different layers of the somatosensory cortex. J. Neurosci. 11:226-237. Rausell, E. and E.G. Jones (1991b) Histochemical and immunocytochemical compartments of the thalamic VPM nucleus in monkeys and their relationship to the representational map. J. Neurosci. 11:210-225. Rausell, E., C.S. Bae, A. Vinuela, G.W. Huntley, and E.G. Jones (1992) Calbindin and parvalbumin cells in monkey VPL tha1amic nucleus: Distribution, laminar cortical projections and relations to spinothalamic terminations. J. Neurosci. 12:4088-4111. Rausell, E., C.G. Cusick, E. Taub, and E.G. Jones (1992) Chronic deafferentation in monkeys differentially affects nociceptive and nonnociceptive pathways distinguished by specific calcium-binding proteins and down-regulates gamma-aminobutyric acid type A receptors at thalamic levels. Proc. Natl. Acad. Sci., U.S.A. 89:2571-2575. Ravizza R.J., R.B. Straw and P.D. Long (1976) Laminar origin of efferent projections from auditory cortex in the Golden Syrian Hamster. Brain Res. 114:497-500. Redies, H., S. Brandner, and O.D. Creutzfeldt (1989) Anatomy of the auditory thalamocortical system of the guinea pig. J. Compo Neurol. 282:489-511. Robson, J. A. and W. C. Hall (1975) Connections of layer VI in striate cortex of the grey squirrel (Sciurus carolinensis). Brain Res. 93: 133-130. Rockel, A.J., R.W. Hiorns, and T.P.S. Powell (1980) The basic uniformity in structure of the neocortex. Brain. 103:221-244. Rodrigues-Dagaeff, C., G. Simm, Y. De Ribaupierre, A. Villa, F. De Ribaupierre and E.M. Rouiller (1989) Functional organization of the ventral division of the medial geniculate body of the cat: Evidence for a rostro-caudal gradient of response properties and cortical projections. Hearing Res. 39:103-126. Roger, M. and P. Arnault (1989) Anatomical study of the connections of the primary auditory area in the rat. J. Compo Neurol. 287:339-356. 190 Romanski, L.M. and J.E. LeDoux (1993) Organization of rodent auditory cortex: anterograde transport of PHA-L from MGv to temporal neocortex. Cerebral Cortex 3:499-514. Rose, J.E. (1942) The thalamus of the sheep: Cellular and fibrous structure and comparison with pig, rabbit, and cat. J. Compo Neurol. 77:469-523. Rose, J.E. (1949) The cellular structure of the auditory region of the cat. J. Compo Neurol. 91:409-440. Rouiller, E.M. and F. de Ribaupierre (1985) Origins of afferents to physiologically defined regions of the medial geniculate body of the cat: ventral and dorsal divisions. Hear. Res. 19:97-114. Rouiller, E.M., E. Colomb, M. Capt and F. de Ribaupierre (1985) Projections of the reticular complex of the thalamus onto physiologically characterized regions of the medial geniculate body. Neurosci. Lett. 53:227-232. Rouiller, E.M., C. Rodrigues-Dagaeff, G. Simm, Y. De Ribaupierre, A. Villa and F. De Ribaupierre (1989) Functional organization of the medial division of the medial geniculate body of the cat: Tonotopic organization, spatial distribution of response properties and cortical connections. Hearing Res. 39:127-142. Rouiller, E.M. and E. Welker (1991) Morphology of corticothalamic terminals arising from the auditory cortex of the rat: A Phaseolus vulgaris-Ieucoagglutinin (PHA-L) tracing study. Hearing Res. 56:179-190. Rubel, E. W. (1978) Ontogeny of structure and function in the vertebrate auditory system. In: M. Jacobson (Ed): Handbook of Sensory Physiology, Development of Sensory Systems, Springer-Verlag, New York, pp. 135-237. Rubel, E.W. (1984) Ontogeny of auditory system function. Annu. Rev. Physiol. 46:213- 229. Ryugo, D.K., R. Ryugo, and H.P. Killackey (1975) Changes in pyramidal cell density consequent to vibrissae removal in the newborn rat. Brain Res. 96:82-87. Ryugo, D.K. and N.W. Weinberger (1976) Corticofugal modulation of the medical geniculate body. Exp. Neurol. 51:377-391. 191 Schlaggar, B.L., and D.D.M. O'Leary (1991a) Potential of visual cortex to develop an array of functional units unique to somatosensory cortex. Science. 252:1556-1560. Schlaggar, B.L., and D.D.M. O'Leary (1991b) A periphery-related pattern is evident in rat somatosensory cortex at birth. Soc. Neurosci. Abstr. 17: 1127. Schlaggar, B.L., K. Fox, and D.D.M. O'Leary (1993) Postsynaptic control of plasticity in developing somatosensory cortex. Nature. 364:623-626. Scheel, M. (1988) Topographic organization of the auditory thalamocortical system in the albino rat. Anat. Embryol. 179:181-190. Schreiner, C.E., I.R. Mendelson and M.L. Sutter (1992) Functional topography of cat primary auditory cortex: representation of tone intensity. Exp. Brain Res. 92: 105-122. Schroeder, D.M. and I.A. lane (1971) Projections of dorsal column nuclei and spinal cord to brain stem and thalamus in the tree shrew, Tupaia glis. I. Compo Neurol. 142:309-350. Senft, S.L. and T.A. Woolsey (1991) Growth of thalamic afferents into mouse barrel cortex. Cerebral Cortex. 1:308-355. Sequier, I.M., W. Hunziker, C. Andressen, and M.R. Celio (1990) Calbindin D-28k protein and mRNA localization in the rat brain. Eur. I. Neurosci. 2:1118-1126. Seto-Oshima, A., P.C. Emson, M.W. Berchtold, and C.W. Heizmann (1989) Localization of parvalbumin mRNA in rat brain by in situ hybridization histochemistry. Exp. Brain Res. 75:653-658. Shatz, C.I., and M.P. Stryker (1978) Ocular dominance in layer IV of the cat's visual cortex and the effects of monocular deprivation. I. Physiol. 281:267-283. Shatz, C.I., and M.B. Luskin (1986) Relationship between the geniculocortical afferents and their cortical target cells during development of the cat's primary visual cortex. I. Neurosci. 6:3655-3668. Shatz, C.I. and M.P. Stryker (1988) Prenatal tetrodotoxin infusion blocks segregation of retinogeniculate afferents. Science 242:87-89. 192 Shek, J.W., G.Y. Wen, and H.M. Wisniewski (1986) Atlas of the Rabbit Brain and Spinal Cord. Karger, Basel, 139 pp. Shneidennan, A., D.L. Oliver, and C.K. Henkel (1988) The connections of the dorsal nucleus of the lateral lemniscus. An inhibitory parallel pathway in the ascending auditory system? J. Compo Neurol. 276:188-208. Smelser, C.B., and N. T. McMullen (1992) Parvalbumin immunocytochemistry delineates primary auditory neocortex. Soc. Neurosci. Abstr. 18:1038. Smelser, C.B., and N.T. McMullen (1993) A stereological analysis of parvalbumin containing neurons in auditory neocortex. Soc. Neurosci. Abstr. 19: 1426. Somogyi, P. and I. Soltesz (1986) Immunogold demonstration of GABA in synaptic tenninals of intracellularly recorded, horseradish peroxidase-filled basket cells and clutch cells in the cat's visual cortex. Neurosci. 19: 1051-1-65. Sousa-Pinto, A. (1973) Cortical projections of the medial geniculate body of the cat. Adv. Anat. Embryol. Cell. BioI. 48: 1-42. Spatz, W.B., R.-B. Illing, and D.M.V. Weisenhorn (1994) Distribution of cytochrome oxidase and parvalbumin in the primary visual cortex of the adult and neonate monkey, Callithrixjacchus. J. Compo Neurol. 339:519-534. Stanfield, B.B., and D.D.M. O'leary (1985) Fetal occipital cortical neurons transplanted to the rostral cortex can extend and maintain a pyramidal tract axon. Nature. 313:135- 137. Steindler, D.A., N.G.F. Cooper, A. Faissner, and M. Schachner (1989) Boundaries defined by adhesion molecules during development of the cerebral cortex: the J1/tenascin glycoprotein in the mouse somatosensory cortical barrel field. Dev. BioI. 131:243-60. Steindler, D.A., T.F. O'Brien, E. Laywell, K. Harrington, A. Faissner, and M Schachner (1990) Boundaries during nonnal and abnonnal brain development: in vivo and in vitro studies of glia and glucoconjugates. Exp. NeuroI. 109:35-56. Stichel, C.C., W. Singer, C.W. Heizmann, and A.W. Nonnan (1987) Immunohistochemical localization of calcium-binding proteins, parvalbumin and calbindin-D 28k, in the adult and developing visual cortex of cats: A light and electron microscopic study. J. Compo NeuroI. 262:563-577. 193 Stryker, M.P., and W.A. Harris (1986) Binocular impulse blockade prevents the formation of ocular dominance columns in cat visual cortex. J. Neurosci. 6:2117-2133. Sur, M., P.E. Garraghty, and A.W. Roe (1988) Experimentally induced visual projections into auditory thalamus and cortex. Science. 242:1437-1441. Tadov, E. C. and R. Y. Moore (1966) The tecto-thalamic connections in the brain of the rabbit. J. Compo Neurol. 126:403-422. Ts'o, D.Y., and C.D. Gilbert (1988) The organization of chromatic and spatial interactions in the primate striate cortex. J. Neurosci. 8: 1712-1727. Usrey, W.M., E.C. Muly, and D. Fitzpatrick (1992) Lateral geniculate projections to the superficial layers of visual cortex in the tree shrew. J. Compo Neurol. 319:159-171. Van Brederode, J.F.M., K.A. Mulligan and A.E. Hendrickson (1990) Calcium-binding proteins as markers for subpopulations of GABAergic neurons in monkey striate cortex. J. Compo Neurol. 298:1-22. Vaughan, D.W. and S. Foundas (1982) Synaptic proliferation in the auditory cortex of the young adult rat following callosal lesions. J. Neurocytol. 11:29-51. Vaughan, D.W. (1983) Thalamic and callosal connections of the rat auditory cortex. Brain Res. 260:181-189. Valverde, F. (1968) Structural changes in area striata of the mouse after enucleation. Exp. Brain Res. 5:274-292. Valverde, F. (1976) Aspects of cortical organization related to the geometry of neurons with intracortical axons. J. Neurocytol. 5:509-529. Van Brederode, J.F.M., M.K. Helliesen, and A.E. Hendrickson (1991) Distribution of the calcium-binding proteins parvalbumin and calbindin-D28k in the sensorimotor cortex of the rat. Neuroscience. 44:157-171. Van der Loos, H., and T.A. Woolsey (1973) Somatosensory cortex: structural alterations following ear injury to sense organs. Science. 179:395-398. 194 Vrensen, G., D. DeGroot, and J. Nunes-Cardozo (1977) Postnatal development of neurons and synapses in the visual and motor cortex of rabbits: A quantitative light and electron microscopic study. Brain Res. Bull. 2:405-416. Wallace, M.N., L.M. Kitzes, and E.G. Jones (1991) Chemoarchitectonic organization of the cat primary auditory cortex. Exp. Brain Res. 86:518-526. Walsh, C., and C.L. Cepko (1992) Widespread dispersion of neuronal clones across functional regions of the cerebral cortex. Science. 255:434-440. Watanbe, T., K. Yanagisawa, J. Kanzaki and Y. Katsuki (1966) Cortical efferent flow influencing unit responses of the medical geniculate body to sound stimulation. Exp. Brain Res. 2:302-317. Wenstrup, J.J. (1994) The auditory thalamus in bats. In Springer Handbook of Auditory Research, Vol. 11, Hearing by Bats. (in press). Wenstrup, J.J., D.T. Larue, and J.A. Winer (1994) Projections of physiologically defined subdivisions of the inferior colliculus in the mustached bat: Targets in the medial geniculate body and extrathalamic nuclei. J. Compo Neurol. 346:207-236. . Wepsic, J.G. (1966) Multimodal sensory activation of cells in the magnocellular medial geniculate nucleus. Exp. Neurol. 15:299-318. White, E.L. (1978) Identified neurons in mouse SmI cortex which are postsynaptic to thalamocortical axon terminals: A combined Golgi-electron microscopic and degeneration study. J. Compo Neurol. 181:627-662. White, E.L., and M.P. Rock (1981) A comparison of thalamocortical and other synaptic inputs to dendrites of two non-spiny neurons in a single barrel of mouse SmI cortex. J. Compo Neurol. 195:265-277. White, E.L. (1989) Cortical Circuits: Synaptic Organization of the Cerebral Cortex. Structure Function and Theory. Boston: Birkhauser. Williams, S.M., P.S. Goldman-Rakic and C. Leranth (1992) The synaptology of the parvalbumin-immunoreactive neurons in the primate prefrontal cortex. J. Compo Neurol. 320: 353-369. Winer, J.A. (1984) The human medial geniculate body. Hear. Res. 15:225-247. 195 Winer, I.A. (1984) Anatomy of layer IV of cat primary auditory cortex (AI). I. Compo Neurol. 224:535-567. Winer, I.A. (1985) The medial geniculate body of the cat. Adv. Anat. Embryol. Cell. BioI. 86:1-98. Winer, I.A. (1986) Neurons accumulating [3H] gamma-aminobutyric acid (GABA) in supragranular layers of cat primary auditory cortex (AI). Neuroscience. 19:771-792. Winer, I.A., LT. Diamond, and D. Raczkowski (1977) Subdivisions of the auditory cortex of the cat: the retrograde transport of horseradish peroxidase to the medial geniculate body and posterior thalamic nuclei. I. Compo Neurol. 176:387-418. Winer, I.A. and D.K. Morest (1983a) The medial division of the medial geniculate body of the cat: Implications for thalamic organization. I. Neurosci. 3:2629-2651. Winer, I.A. and D.K. Morest (1983b) The neuronal architecture of the dorsal division of the medial geniculate body of the cat: A study with the rapid Golgi method. I. Compo Neurol. 224:344-370. Winer, I.A. and D.T. Larue (1987) Patterns of reciprocity in auditory thalamocortical and corticothalamic connections: study with horseradish peroxidase and autoradiographic methods in the rat medial geniculate body. I. Compo Neurol. 257:282-315. Winer, I.A., D.K. Morest, and LT. Diamond (1988) A cytoarchitectonic atlas of the medial geniculate body of the opossum, Didelphys virginiana, with a comment on the posterior intralaminar nuclei of the thalamus. I. Compo Neurol. 274:422-448. Winer, I.A. and D.T. Larue (1989) Populations of GABAergic neurons and axons in layer I of rat auditory cortex. Neuroscience 33:499-515. Winer, I.A. (1991) Anatomy of the medial geniculate body. In R.A. Altschuler, R.P. Bobbin, B.M. Clopton and D.W. Hoffman (Eels) Neurobiology of Hearing: The Central Auditory System. Raven Press, Ltd., New York, pp. 293-333. Winer, I.A. (1992) The functional architecture of the medial geniculate body and the primary auditory cortex. In D.B. Webster, A.N. Popper and R.R. Fay (Eels) The Mammalian Auditory Pathway: Neuroanatomy, Springer Verlag, New York, pp. 222- 409. 196 Winer, J.A. and J.J. Wenstrup (1994a) Cytoarchitecture of the medial geniculate body in the mustached bat (Pteronotus parnellz). J. Compo Neurol. 346:161-182. Winer, J.A. and J.J. Wenstrup (1994b) The neurons of the medial geniculate body in the mustached bat (Pteronotus parnellz). J. Compo Neurol. 346:183-206. Wise, S.P. and E.G. Jones (1977) Cells of origin and terminal distribution of descending projections of the rat somatic sensory cortex. J. Compo Neurol. 175:129-158. Wong, D. and J.P. Kelly (1981) Differentially projecting cells in individual layers of the auditory cortex: a double-labeling study. Brain Res. 230:362-366. Woolsey, T.A., and H. Van der Loos (1970) The structural organization of lamina IV in the somatosensory region (SI) of mouse cerebral cortex: the description of a cortical field composed of discrete cytoarchitectonic units. Brain Res. 17:205-242. Zettel, M.L., C.E. Carr, and W.E. O'Neill (1991) Calbindin-like immunoreactivity in the central auditory system of the mustached bat, Pteronotus parnelli. J. Compo Neurol. 313: 1-16.