- Email: [email protected]
Frontal cortex projections to the amygdaloid central nucleus in the rabbit
0306-4522/85 53.00 + 0.00 Pergamon Press Ltd G 1985 IBRO
Neuroscience Vol. IS. No. 2, pp. 327-346. 1985 Printed in Great Britain
FRONTAL
CORTEX PROJECTIONS TO THE AMYGDALOID CENTRAL NUCLEUS IN THE RABBIT
B. S. KAPP*, J. S. ScHwABERt and P. A. DRISCOLL *Department of Psychology, University of Vermont, Burlington, VT 05405, U.S.A.; and tDupont Central Research and Development, E. I. duPont de Nemours and Company, Wilmington, DE 19036, U.S.A. Abstract-Evidence has recently been presented which demonstrates that the amygdaloid central nucleus projects directly upon cardiovascular/autonomic regulatory nuclei of the dorsal medulla and that in the rabbit this nucleus may influence cardiovascular activity during emotional states. The present study is one of a series of investigations designed to provide information on the innervation of the central nucleus in the rabbit and describes the topography and origin of frontal cortex projections to the nucleus based upon retrograde and anterograde axonal transport techniques. Injections of horseradish peroxidase or the fluorescent dyes, Bisbenzimide or Nuclear Yellow, into the central nucleus resulted in abundant numbers of retrogradely labeled neurons in three regions of the frontal cortex: the insular cortex on the lateral surface and areas 25 and 32 on the medial surface of the hemisphere. The majority of labeled neurons in the insular cortex were located in layer V of the dorsal and posterior agranular insular regions, although labeled neurons were observed in layer V of the granular insular cortex as well as in layers II and III of the posterior agranular insular cortex. Labeled neurons in areas 25 and 32 were located throughout all layers and the total number of these neurons was substantially less than that observed in the insular cortex. Autoradiographic experiments in which amino acids were injected into the insular cortex resulted in a dense pattern of transported label within the central nucleus that extended rostrally into the sublenticular substantia innominata and lateral component of the bed nucleus of the stria terminalis. Label was also observed in the cortical, lateral, basolateral and basomedial amygdaloid nuclei. In contrast to the projections from the insular cortex, amino acid injections into areas 25 and 32 resulted in only relatively light labeling within the most rostra1 region of the central nucleus; otherwise the nucleus was partially encapsulated and virtually devoid of label. These results suggest that the insular cortex possesses the potential to directly influence the central nucleus projection to cardiovascular/autonomic regulatory nuclei of the dorsal medulla and thus, together with the amygdaloid central nucleus, appears to be an important component of a forebrain system involved in cardiovascular/autonomic regulation.
INTRODUCTION Recent evidence suggests that the amygdaloid complex influences autonomic, and in particular cardio-
vascular, regu1ation7.9.“.‘8~“,s4and that the amygdaloid central nucleus and its efferent projection system may provide an anatomical substrate for this influence.‘7.2’.3“.4’.49 For example, our anatomical experiments in the rabbit,4R.49as well as those in the rat.” cat” and monkey,” have demonstrated that the central nucleus projects directly upon cardiovascular regulatory nuclei of the dorsal medulla; the nucleus of the solitary tract, the dorsal motor nucleus of the vagus nerve and the nucleus ambiguus. In the rabbit the cells of origin of these descending projections are primarily located throughout the entire rostralcaudal length of the large, media1 component of the central nucleus and extend rostrally from the nucleus in an uninterrupted continuous region through the sublenticular substantia innominata and into the lateral component of the bed nucleus of the stria terminalis4’ Additional research in the rabbit has demonstrated that (1) electrical stimulation of the central nucleus
*To whom all correspondence
should be addressed. 327
elicits profound heart rate deceleration and depressor2.27 responses as well as activating cardiovascular regulatory interneurons within the nucleus of the solitary tract;” (2) lesions and pharmacological manipulations within the region of the central nucleus attenuate conditioned heart rate deceleration during Pavlovian fear conditioning;‘2.‘3.‘5 and (3) significant increases in central nucleus multiple unit activity emerge in response to a conditioned stimulus over the course of Pavlovian fear conditioning at a time when conditioned heart rate decelerative responses emerge to that stimulus.’ In this context, the present investigation, that of delineating the projections from the frontal cortex to the central nucleus in the rabbit, is one of a series of experiments designed to gain further insight into the function of the central nucleus as one component of a larger forebrain system regulating cardiovascular and autonomic activity during emotional states. The existence of projections from the frontal cortex to the amygdala has been known for some time,i9.)j as has physiological evidence demonstrating autonomic responses to electrical stimulation of the medial and lateral aspects of the frontal cortex, including the lateral and medial orbital, insular and cingulate cortices.‘5.20.24.53”More recent anatomical findings have demonstrated that various regions of
B. S. Kapp ei lil
328 the frontal cortex. including
areas 25 and 32 on the medial surface as well as the insular cortex on the lateral surface of the hemisphere, project to the region of the central nucleus.3.‘J.37.“3.‘sHowever, the extent to which these projections innervate the region of origin of the central nucleus projection to medullary cardioregulatory nuclei, as well as the very nucleus insular cortex-central existence of projections in all species studied, is equivocal. For example, while Saper4’ reported a well-defined projection from the insular cortex to the amygdaloid central nucteus in the rat, Beckstead’ reported that projections from the insular region surrounded, but did not appear to enter, the central nucleus in the rat. Furthermore, Reep and Winans@ did not find evidence for a projection from the insular region to the amygdaloid central nucteus in the hamster. Given the data in rabbit implicating a centrai nucleus function in cardiovascular regulation. the present investigation was performed to provide information on frontal cortex projections to the central nucleus in this species using retrograde and anterograde axonal transport techniques. In particular. the topo~aphy and specific origin of frontal cortex projections to the central nucleus were determined with respect to the medial component of the central nucleus and to the closely related sublenticular substantia innominata and lateral component of the bed nucieus of the stria terminalis, since it is from this continuum that a prominent project&n to medul~ar~ cardiovascular and autonomic regulatory nuclei arises. EXPERIMENTAL PROCEDURES Retrograde transport experiments
Thirty-three pure-bred New Zealand rabbits weighing from 2.0-2.8 kg were pretreated with chlorpromazine hydrochloride (20 mg in 0.8 cc saline i.v.) and anesthetized with pentobarbitai (Nembutal; 30-75 mg, i.v.). They were placed in a Kopf stereotaxic instrument fitted with a rabbit head-holder. Unilateral injections of either horseradish peroxidase fn = ii; Sigma VI; Z-507;; IO-50 ni) or the fluorescent dyes Bisbenzimide (n = 19: Hoechst 33258; 5-IO?;; W60ni) or Nuclear Yefiow (n = 3; Hoechst S 769121: 5%; 40ni) were aimed at the central nucleus using the following co-ordinates: 0.0-0.5 mm anterior to bregma, 5.7 mm lateral to the midline and 12.0-12.5 mm ventral to dura. The head was adjusted such that lambda was I.5 mm below bregma. Injections were delivered via a 26 gauge needle or a glass micropipette (4&6Opm tip diameter) cemented to a 26 gauge needle attached to a 1.Opl Hamilton syringe. All injections were made over a 20-40 min period. The animals were sacrificed with injections of T-61 Euthanasia solution (Hoechst) from 6 h to 3 days following the injection. The horseradish peroxidase-injected animals were perfused and 40pm frozen sections were taken throughout the rostrai-caudai extent of the frontal cortex. The sections were processed using the tetramethyl~n~dine procedure as described previously in the rabbit” and munterstained with Nuclear Red. Sections were viewed under both bright- and dark-field microscopy. Animals receiving Bisbenzimide or Nuclear Yellow were perfused with 1000mi of physiological saline followed by 2000mi of a phosphate-buffered loo/, fonnaiin solution (pH 7.4). The brains were postfixed for I8 h in a 20?/,
sucrose-buffered formalm soiut~on and altern~~r J\ ;I iE frozen sections were collected in drstrllrd water The> WT!. immediately mounted onto uncoated shdes. allowed to stir dry for several hours. and then viewed and photographed using an Olympus BHA photomicroscope equipped for epifluorescence. A 360 nm excitation waveiength (filter I.&1) was used for both Bisbenzimide and Nuclear Yetlou The additional. alternate sections were collected in lo”, formalin and stained with thionin to facilitate the localization of labeled neurons. Line drawings of serial sections were made with the use of a drawing tube (camera lucida) and Bausch and Lomb microprojector. The locations of labeled neurons were plotted with reference to various landmarks and with the aid of the thionin-sIained secttons. as well as additional nissl-stained brains prepared as atlases. Anterograde transport r.yerrmenrs
Based on the results of our retrograde transport expertments. stereotaxic injections of radioactively labeled amino acids were made into the agranular and granular regions of the insular cortex (n = 17) and areas 25 and 32 of the medial prefrontal cortex (11= i6).R Co-ordinates for the agranular insular injections varied from 3.0-5.5 mm anterior to bregma. 4.2-5.7 mm lateral to the midline and 5.5-8.5 mm ventral to dura. Those for areas 25 and 32 were 5.5-8.5 mm anterior to bregrna. 0.7 mm lateral to the midline and 6.0-8.5 mm ventral to dura. Injection volumes varied from 30 to 300111 and were comprised of equal parts of ~3H]proline and [‘Hjieucine f i .OmCi!mi; Amershamf in a concentration of 50 ~Cci;~i. Ail injections were delivered via a glass micropipette (4&6Opm tip diameter) over a 3@-40min period. Survival times ranged from I to 7 days. The animals were then sacrificed and perfused with 1000 ml of saline followed by 2000ml of a IO‘?,;) buffered formalin solution (pH 7.4). Frozen sections were taken at ?5pm. mounted on chrom alum-subbed slides. defatted and coated with Kodak NTB-2 emulsion and exposed from 6 to i? weeks at 4 C. The autoradiographs were then developed with Kodak D-IV developer and stained with thionin. Sections were examined under bright- and dark-field microscopy and the distribution of silver grains within the amygdala was photographed and reconstructed with the atd of additional nissl-stained sections from separate brains prcpared atlases. The te~Ino~ogy and su~ivisi~~ns used to describe the amygdaloid nuciei in this study follows that previously used by us in the rabbit.” The terminoiogy used to describe the insular cortex adheres to that used by RoseJ’ to describe this region in the rabbit. while areas 25 and 32 refer to the frontal corttcal regtons on the medial wall of the hemispheres as described in the rabbit by Brodmann.’ RESLLTS Retrograde
transport
experiments
All three retrograde tracers produced similar patterns of retrograde labeling. Bisbenzimide proved the most useful of the three. producing both the most clearly definable, restricted injection sites and the greatest number of retrogradely labeled neurons. optimal Bisbenzimide neuronal labeling with minimal glial involvement was achieved with 48-72 h survival times. Abundant numbers of retrogradely labeled neurons were located in three frontai corticaf regions: the insular cortex on the lateral surface of the hemisphere and areas 25 and 32 on the medial surface (Fig. I). Labeled neurons were located primarily ipsilateral to the injection site. although a few, faintly labeled neurons were observed on the contralateral side.
Frontal cortex projections to amygdaloid central nucleus particularly within the insular cortex. Of the three frontal cortical regions. the insular cortex contained by far the greatest numbers of labeled neurons while area 32 contained the least. A few labeled neurons were also observed to be scattered within the medial precentral cortex and the dorsal division of the anterior cingulate cortex. The insular cortex in the rabbit, as defined by Rose,@ overlies the claustrum on the lateral surface of the hemisphere and lies dorsal to and within the fundus of the rhinal sulcus. It is comprised of a dorsal granular cortex, containing a well-developed layer IV (GI, Figs 1 and 2) and a ventrally adjacent agranular region (AI, Figs 1 and 2). The agranular region, in turn, has been further divided into three regions: a dorsal agranular insular region (ail), a ventral agranular insular region (ai2) and a posterior agranular insular region (ai3).42 These areas were readily distinguished in the present study. A representative case in which 30 nl of a 10% Bisbenzimide suspension was injected into the rostra1 portion of the central nucleus is illustrated in Fig. 2. The intensely fluorescent injection site was primarily confined to the central nucleus, with some slight encroachment upon the substantia innominata, the ventral globus palhdus and the ventromedial putamen. Within the more rostra1 insula the majority of Bisbenzimide-labeled neurons were located in layer V of the dorsal agranular insula (Figs 2b,c). A few labeled neurons were also observed in layer V of the ventral agranular insula and in layer V of the granular insula lying dorsal to the dorsal agranular insula. More posteriorly, numerous labeled neurons were observed in layer V of the posterior agranular insula and this field of labeled neurons extended dorsally into layer V of the granular insular region (Fig. 2e). In addition, faintly labeled neurons were observed in layers II and III of-the posterior agranular insula although the total number of these neurons was far fewer than those located in layer V (Fig. 2e). On the medial surface of the hemisphere labeled neurons were observed scattered throughout all layers of area 25 while fewer numbers were observed throughout the layers of area 32 (Figs 2b,c). The total numbers and density of labeled neurons in these two areas were substantially less than the numbers observed in the insular cortex.
Anterograde
transport experiments
The insular cortex. The injection sites of tritiated amino acids varied in placement within the dorsal-ventral and rostral-caudal extents of the insular cortex. However, all those injections in which the effective injection site (determined by a homogeneous blackening of the majority of cells due to the density of silver grains) covered at least a portion of layer V of the agranular and granular insular regions resulted in a distinct pattern of transported label within the central nucleus, with no differences in the topography
329
of label within the nucleus as a function of differences in the location of injection sites along the rostral-caudal extent of the insular cortex. _ The injection site from a representative case is shown in Fig. 3. The effective injection site covers the ventral portion of the dorsal agranular insular region (ail), the ventral agranular insular region (ai2) and the lateral edge of the claustrum. Much of the granular insular region was located dorsal to the effective injection site. Labeled fibers destined for the amygdala appeared to course within the internal and external capsules. At the rostra1 border of the amygdala, labeled fibers were observed to exit the internal capsule and course in a ventral and ventrolateral direction to innervate the sublenticular substantia innominata and the amygdala (Fig. 4). Within the amygdala, the central nucleus, including both its medial and lateral subdivisions,49 was dense in silver grain concentration along its entire rostral-caudal extent (Figs 4-7). The central nucleus was the most densely labeled nucleus of the deep nuclei of the amygdaia. An intermediate fiber-rich region of the nucleus appeared somewhat less densely labeled at some levels (Figs 5 and 6). While at more rostra1 levels labeled fibers appeared to stream toward the nucleus from the internal capsule, at more caudal levels a broad band of labeled fibers was observed to be oriented in a dorsolateral to ventromediai direction extending from the external capsule through the ventral putamen where it appeared to enter the central nucleus on its dorsolateral border (Fig. 7). In addition, at caudal levels a diffuse network of labeled fibers, possibly originating from the ventral-most extent of the external capsule, was observed to traverse the anterior pole of the anterior basolateral nucleus and to extend into the ventral border of the nucleus. In addition to the central nucleus, a moderate amount of transported label was also distributed within the lateral component of the bed nucleus of the stria terminalis (Fig. 8). Label was also distributed throughout the rostral
330
8. S. Kapp et al.
Abbreviations used in jigures AC :I ail ai BI Bm BST Ce c”L”
anterior cortical nucleus of the amygdala anterior commissure agranular insular cortex dorsal agranular insular cortex ventral agranular insular cortex basolateral nucleus of the amygdala basomedial nucleus of the amygdala bed nucleus of the stria terminalis central nucleus of the amygdala external capsule claustrum
fx GI GP La Me 01 P PLC PMC PP SI
fomix granular insular cortex globus pallidus lateral nucleus of the amygdala medial nucleus of the amygdala optic tract putamen posterolateral cortical nucleus of the amygdala posteromedial cortical nucleus of the amygdala prepiriform cortex substantia innominata
Fig. 1 (A), Photomicrograph of frontal section of rabbit brain showing the insular and medial prefrontal cortical regions where retrogradely labeled neurons were observed following injections of retrograde tracers into the central nucleus of the amygdala. (B) Higher power photomicrograph of frontal section of insular cortex in the rabbit. Large arrowheads delineate innermost border of Layer V, the layer within which the vast majority of retrogradely labeled neurons were located. Sections are 25 pm and stained with thinonin. Fig. 2. Line drawings of the amygdaloid injection site and of representative frontal sections through the insular cortex from case Bb27 in which Bisbenzimide was injected into the central nucleus of the amygdala. The greatest extent of the injection site is illustrated in (A). The darkened region depicts the intensely fluorescent central area, while the lined region depicts the less intensely fluorescent “halo” region surrounding the central area. The distribution of labeled neurons in the insular and medial prefrontal cortex is illustrated on sections from rostra1 (B) to caudal (E) levels. Each dot represents a single-labeled neuron from that section alone. Figs. 4-7. Bright-(A) and dark-field (B) photomicrographs of representative frontal sections taken at four rostra1 (Fig. 4) to caudal (Fig. 7) levels through the amygdaloid complex of case I-4. Note the highly specific, markedly dense distribution of silver grains along the entire rostral-caudal extent of the central nucleus. Rectangles in Figs 4 and 5 represent areas illustrated in Figs 9 and 10, respectively. Fig. 9. Bright-(A) and dark-field (B) photomicrographs of frontal sections showing the distinct laminar field of silver grains in layer IB of the plexiform layer of the amygdaloid anterior cortical nucleus. Area shown is outlined by rectangle in Fig. 4. Fig. 10. Bright-(A) and dark-field (B) photomicrographs of frontal sections showing the laminar field of silver grains in layer IB of the plexiform layer of the amygdaloid posterolateral cortical nucleus. Area shown is outlined by rectangle in Fig. 5. Fig. 12. Dark-field photomicrographs of frontal sections taken at two levels through the central nucleus of the amygdala of case FM-IO. (A) is more rostra]. Note that silver grains encapsulate the nucleus and that the nucleus itself is relatively devoid of label, as is particularly noteworthy in (B).
B Fig. 1 331
:
GI
;-,‘;. ::‘: ..,
I
.’
32
Al
.. . ”
.:..
-
25 )
B
C
Fig. 2. 332
32
Fig. 3. Photomicrograph
of the largest extent of the injection site (60 nl; 4 day survival; 8 week exposure time) within the insular cortex of case I-4.
333
Fig. 4. 334
Fig. 5. 335
Fig. 6 316
Fig. 7. 337
Fig. 8. Dark-field photomicrograph of frontal section showing the distribution of silver grains within the lateral part of the bed nucleus of the stria terminalis for case l-4.
338
Fig. 9. 339
Fig. 10.
340
Fig. Il. Pha ltomicrograph
of the largest extent of the injection site (100 nl; 7 day survival; time) within the medial prefrontal cortex of case FM-IO.
341
8 week exposure
Fig.
12.
Frontal cortex projections to amygdaloid central nucleus piriform nucleus and layers IB and III of the prepirifo~ cortex were heavily labeled throu~out the entire rostraI
343
appeared destined to innervate nuclei lying ventral to the central nucleus. Although the central nucleus appeared devoid of label along most of its rostral-caudal extent, other nuclei of the amygdala were innervated. At the rostra1 pole of the amygdala label was diffusely distributed within layers II and III of the anterior cortical nucieus, while at more posterior levels label was observed within and surrounding the basolateral nucleus, which was the most heavily labeled region within the amygdala. Lighter amounts of label were present in the basomedial nucleus. The medial, lateral, posteroIatera1 cortical and ~steromed~al cortical nuclei were not labeled nor was significant label present within the prepiriform cortex. DISCUSSION
Insular cortex-central nuclei projection The results of the present study demonstrate that the central nucleus of the amygdala in the rabbit is the recipient of a prominent projection from the insular cortex. The projection heavily innervates the medial component of the central nucleus, the adjacent sublenticular substantia innominata and the lateral component of the bed nucleus of the stria terminalis. These three areas have in turn been observed to form a continuous region in the rabbit from which arises a prominent projection to autonomic regulatory nuclei of the dorsomedial medullaJv This observation has suggested the anatomical unity of this continuum as a single entity,4g a concept supported by the observations that these three structures possess common cytoarchitecture,~.‘6,22 staining properties,‘0”6 neurochemical innervatiotP’ and afferentation from the nucleus of the solitary tract.” The results of the present study demonstrating a projection from the insular cortex to the entire extent of this continuous region are consistent with this notion. Based upon differential thalamic a;ferentation, Guldin and Markowitsch” have recently suggested the heterogeneous nature of the insular cortex, and its division into prefrontal insular, gustatory insular and associative insular regions in the rat. The existence of regional differentiation of the insular cortex in the rabbit based upon thalamocortical afferentation has yet to be determined. Hence the origin of the insular-central nucleus projection in the context of the divisions proposed by Guldin and Markowitschi4 is not known and awaits further research. However, our results demonstrating that the insular projection to the central nucleus arises primarily from layer V of the agranular and granular insular regions along the entire rostral-caudal extent of these regions suggest that the projection may originate from all of the regions delineated by Guldin and Markowitsch.14 The results of the present study are consistent with recent reports describing an insular-amygdala projection in the rat,37”5.57hamster,@ caP3 and mon-
344
B. S. Kapp et al
key.” Inconsistencies exist. however, among these reports concerning the extent to which this projection innervates the central nucleus. For example, the results of the present study are similar to the results of Saper”’ demonstrating the presence of anterogradely transported label within the rat amygdaloid central nucleus, lateral nucleus, basolateral nucleus and lateral component of the bed nucleus of the stria terminalis following horseradish peroxidase injections into the insular cortex. The results of the present study and those of Saper4’ differ, however, from those of Beckstead who reported that tritiated amino acid injections into the insular cortex of the rat produced label mainly in the lateral. basolateral and basomedial nuclei, while labeled fibers surrounded but did not appear to enter the central nucleus. Results similar to those of Beckstead’ have recently been reported by Reep and Winan?’ who found that amino acid injections into the anterior agranular insular regions of the hamster did not result in label within the central nucleus. Several factors may account for the absence of label within the central nucleus as reported by Beckstead’ and Reep and Winans.40 First, as suggested by Saper,” most of the injection sites reported by Beckstead were located in the lateral prefrontal cortex just anterior to the insula and barely encroached upon the anterior insula. Second, the injection sites of Reep and Winans”” were located within the anterior insula. Since our retrograde transport experiments of the present study demonstrated fewer numbers of ‘labeled neurons within the rostral-most portions of the insula (see Fig. 1) and since SaperJ! observed that horseradish peroxidase injections placed more caudally within the insula resulted in a more dense pattern of label within the central nucleus, the relative absence of label within the central nucleus as reported by Reep and Winans” may well be a function of the anterior placement of their injection sites. In any event. the results of the present study clearly demonstrate that in the rabbit the central nucleus receives a significant projection originating along the entire rostralxaudal extent of the insular cortex. The results of the present study also demonstrate a more extensive insular projection to the amygdala than previously reported in studies using anterograde tracing techniques in the rat and hamster.‘~J0~“5 The transported label observed in the present study in the anterior cortical, posteromedial cortical and posterolateral cortical nuclei was not described in these studies. Otterson, however, reported that injections of horseradish peroxidase into the anterior and posterior cortical nuclei resulted in retrogradely labeled neurons in the agranular insular cortex. These findings, as well as the observations by Mufson et al.34 in monkey demonstrating that tritiated amino acids injected into the anterior insula produced widespread labeling in the amygdala, including the cortical nucleus, are consistent with the results of the present study. Hence it appears that the insular cortex
proJection to the amygdala IS widespread. inner\uttng all of the amygdaloid nuclei with the apparent exception of the medtal nucleus and with a particularI> heavy innervation cf the central nucleus.
The results of the present study demonstrate that the projections from the medial prefrontal cortex to the region of the central nucleus differ markedly from the projections from the insular cortex. With the exception of the presence of the diffuse distribution of silver grains overlying a broad area of the anterior amygdala including the most anterior region of the central nucleus, the central nucleus was free of label following injections of amino acids into the medtal prefrontal cortex. Given the presence of labeled fibers and terminals surrounding the central nucleus. it is probable that much of the retrograde label observed in the medial frontal cortex following injections of retrograde tracers into the central nucleus was a function of uptake and transport of label from either injured fibers of passage and:or from terminals surrounding the nucleus. Our observations in the rabbit are similar to those of Beckstead’ who also reported that injections of amino acids into area 32 produced label in the lateral, basolateral and basomedial nuclei while labeled fibers surrounded. but did not enter the central nucleus in the rat. More recently Russchen” has reported a similar result in the cat. An injection of [‘Hlleucine covering area 25 and the caudal part of are3 32 produced label overlymg the basolateral nucleus and a “capsule” surrounding the medial component of the central nucleus. although some label was reported to overlie the medial component.
The functional significance of the Insular projection to the central nucleus is at present a matter for speculation. Our recent research, however. has suggested that the central nucleus may function m vagal modulation of heart rate in the rabbit to the presentation of conditioned fear-arousing stimuli, quite possibly via direct projections from the central nucleus to the nucleus of the solitary tract, the dorsal motor nucleus of the vagus nerve and the nucleus ambiguus.‘6 The latter two regions contain vagal preganglionic cardioinhibitory neurons in the rabbit?3_j- and stimulation of the central nucleus directly influences neurons of the nucleus of the solitary tract which receive baroreceptor input.” The insular cortex projection to the site of origin of the central nucleus descending pathway to medullary autonomic regulatory nuclei endows the insular cortex with the potential to exert a significant influence on the outflow of the central nucleus in its functional role in cardiovascular and autonomic regulation. Indeed recent experiments in both the anesthetized and awake rabbit38 (C. G. Markgraf and B. S Kapp
Frontal cortex projections to ~ygdaloid unpublished observations) have demonstrated that stimulation at sites within &he insular cortex elicits profound bradycardia and depressor responses, a response pattern identical to that observed upon stimulation of the central nucleus in the rabbit.227 The extent to which the responses from st~muiation of the insula are mediated via the central nucleus~orsa~ medulla projection and/or via direct projections from the insula to the dorsal medulla24.s0 in the rabbit awaits further research. Nevertheless, the results of the present experiment, taken together with recent findings of connections of the insu!ar
central nucleus
345
cortex and of the amygdaloid central nucleus with other areas implicated in autonomic function, including the parabrachial complex,6~29~~~M-46~s2~s6 the nucleus of the solitary tract, the dorsal motor nucleus of the vagus nerve and the nucleus ambiguuszi~2s~33~39~4s~49~~~~’ and the laterai h~othalamus4.29.*.4s suggest the inclusion of the insular cortex and amygdaloid central nucleus in a forebrain system concerned with cardiovascular/autonomic regulation. Acknowledgements-This
work was supported by USPHS
Grant NSI 607.
REFERENCES 1. Applegate C. D., Frysinger R. C., Kapp B. S. and Gallagher M. (1982) Multiple unit activity recorded from amyg~aia central nucleus during Pavlovian heart rate conditioning in rabbit. Brain Res. 238, 457-462. M. D. and McNall C. L. (1983) Autonomic and somatomotor effects of 2. Applegate C. D., Kapp B. S., Unde~~ amygdala central n. stimulation in awake rabbits. Physioi. Behau. 31, 353-360. 3. Beckstead R. M. (1979) An autoradiographic examination of corticocortical and subcortical projections of the mediodorsal-projection (prefrontal) cortex in the rat. J. camp. Neurol. 184, 43-62. 4. Berk M. L. and Finkelstein J. A. (1982) Efferent connections of the lateral hypothalamic area of the rat: an autoradiographic investigation. Brain Res. Bull. 8, 51 I-526. 5. Berkelbach van der Sprenkel H. (1926) Stria terminalis and amygdala in the brain of the opossum (Didelphis rirginianu). J. romp. Neuroi. 42, 211-254.
6. Block C. H. and Schwartzbaum J. S. (1983) Ascending efferent projections of the gustatory parabrachial nuclei in the rabbit. Brain Res. 259, 1-9.
1. Bonvallet M. and Gary Bobo E. (1972) Changes in phrenic activity and heart rate elicited by localized stimulation of amygdala and adjacent structures. Eiectroenceph. clin. Neurophysiol. 32, I-16.
8. Brodmann K. (1909) Verg~ei~hende ~kat~a(~onsiehre der Grosshirnrinde in ihren Prinzipien dargesteltr a$ Grund des Zellenbaues. J. A. Barth. Leipzig.
9. Cechetto D. and Calaresu F. R. (1983) Single units in the amygdala of the cat responding to activation of baroreceptors and chemoreceptors.
Neurosci. Absrr. 9, 612.
10. DeOlmos J. S. (1972) The amygdaloid projection field in the rat as studied with the cupric-silver method. In 7&r Neurobiolonv fed. Eleftheriou 3. E.), up. .,. of” rhe Amxedula . ._ . . 145-204. Plenum Press. New York. Il. Faiers A. A., Calaresu F. R. and Mogenson G. J. (1975) Pathway mediating hypotension elicited by stimulation of the amvadala in the rat. Am. J. Phvsiof. 22%. 1358-1364. M., Kapp B. S.. Frysingkr R. C. and Rapp P. R. (1980) /?-Adrenergic manipulation in amygdala central n. alters rabbit heart rate conditioning. Pharmac. Biuchem. Behao. 12, 419-426. 13. Gallagher M., Kapp B. S.. McNall C. L. and Pascoe 3. P. (1981) Effects of opiate administration into the amygdala on Pavlovian heart rate conditioning in the rabbit. Pharmac. Biochent. B&XL 14, 496-505. 14. Guldin W. 0. and Markowitsch H. J. (1983) Cortical and thalamic tierent connections of the insular and adjacent cortex of the rat. J. camp. Neural. 215, 135-153. 15. Hall R. E. and Comish K. (1977) Role of the orbital cortex in cardiac dysfunction in unanesthetized rhesus monkey.
12. Gallag&
Expl Neural. 56, 289-297.
16. Heimer L. (1978) The olfactory cortex and the ventral striatum. In Limb& mechanisms (eds Livingston K. and Hornykiewics 0.). pp. 95-187. Plenum Press, New York.
17. Higgins G. A. and Schwaber J. S. (1983) Somatostatinergic projections from the central nucleus of the amygdala to the vagal nuclei. Peptides 4, l-6.
18. Hilton S. M. and Zbrozyna A. W. (1963) Amydaloid region for defense reactions and its efferent pathway to the brain stem. J. Physiof., Land. 165. 16U-I 73.
19. Hirata Y. (1965) Subcortical projections from the orbital surface of the cat brain. Acta med. stand. 13, 123-142. 20. Hoffman B. L. and Rasmussen T. (1953) Stimulation studies of insular cortex of Macuea mulatta. J. Neur(~ph,vsio~.16, 343-35 I. 21. Hopkins D. A. and Holstege G. (1978) Amygdaioid projections to the mesencephalon, pons and medulla oblongata in the cat. Expr Brain Res. 32, 529-547.
22. Johnston J. 3. (1923) Further contributions to the study of the evolution of the forebrain. J. camp. Neurof. 35.337-481. 23. Jordan D.. Khalid M. E. M., Schneiderman N. and Spyer K. M. (1982) The location and properties of preganglionic neurones in the rabbit. Ppiigers Arch. ges. physiol. 395, 244-250. autonomic and el~tr~ortico~aphic responses to electrical stimu~tion of “rhinencephalic” and other structures in primates. cat and dog. A study of responses from the limbic. subcdllosal. orbito-insular, piriform and temporal cortex, hippocampus-fornix and amygdala. Acta physiol. scund. 24, Suppl. 83, l-285. 25. Kapp B. S.. Frysinger R. C., Gallagher M. and Haselton J. (1979) Amygdala central nucleus lesions: effects on heart rate conditionin in the rabbit. Physiol. Behm. 23, 1109-1117. 26. Kapp B. S., Gallagher M., Applegate C. D. and Frysinger R. C. (1982) The amygdala central nucleus: contributions to conditioned cardiovascular responding during aversive Pavlovian conditioning in the rabbit. in Condjrion~n~: Representation of Inr&ed Neural &zctiok (ed. woody, C. D.), pp. 577-596. Plenum Press, New York. .’ 27. Kapp B. S., Gallagher M., Underwood M. D., McNall C. L. and Whitehom D. (1982) Cardiovascular responses elicited by electrical stimulation of the amygdala central nucleus in the rabbit. Bruin Res. 234, 251-262. vagal cardiomotor
24. Kaada 3. R. (1951) Somato-motor,
346
B. S. Kapp et 01.
28. Kapp B. S., Schwaber J. S. and Driscoll-Mendes P. A. (1983) The organization of insular cortex projections to the amygdaloid central nucleus and autonomic regulatory nuclei of the dorsal medulla. Sot. Neurosci. Absrr. 9, 113. 29. Krettek J. E. and Price J. L. (1978) Amygdaloid projections to subcortical structures within the basal forebram and brainstem in the rat and cat. J. romp. Neurol. 178, 225-254. 30. Lindvall 0. and Bjdrklund A. (1974) The organization of the ascending catecholamine neuron system in the rat brain as revealed by the glyoxylic acid fluorescence method. Acra physiol. scund. Suppl. 412, l-48. 31. Mogenson G. J. and Calaresu F. R. (1973) Cardiovascular responses to electrical stimulation of the amygdala m the rat. Expl Neural. 39, 166180. 32. Moruzzi P., Schwaber J. S., Spyer K. M. and Turner S. A. (1984) Amygdaloid influences on brainstem neurones m the rabbit. J. Physiol., Lond. 66. 33. Mraovitch S., Ruggiero D. A., Ross C. A. and Reis D. J. (1983) Direct projections to autonomic centers of forebrain and brainstem from a cortical vasopressor area in rat. Neurosci. Absrr. 8, 77. 34. Mufson E. J., Mesulam M.-M. and Pandya D. N. (1981) Insular interconnections with the amygdala in the rhesus monkey. Neuroscience 7, 1231-1248. 35. Naura W. J. H. (1961) Fibre degeneration following lesions of the amygdaloid complex in monkey. J. Anar. 95,5 15-53 I. 36. Novotny G. E. K. (I 977) A direct ventral connection between the bed nucleus of the stria terminalis and the amygdaloid complex in the monkey (Macacu fuscicularis). J. Hirnforsch. 18, 271-284. 37. Otterson 0. P. (1982) Connections of the amygdala of the rat. IV. Corticoamygdaloid and intra-amygdaloid connections as studied with axonal transport of horseradish peroxidase. J. camp. Neural. 205, 30-48. 38. Powell D. A., Hemandez L. L. and Buchanan S. L. (1983) Insular cortex participates in cardiovascular inhibition but is not necessary for Pavlovian conditioned bradycardia in rabbits. Neurosci. Absfr. 9, 106. 39. Price J. L. and Amaral D. G. (1981) An autoradiographic study of the projections of the central nucleus of the monkey amygdala. J. Neurosci. 1, 1242-1259. 40. Reep R. L. and Winans S. S. (1982) Efferent connections of dorsal and ventral agranular insular cortex in the hamster. Mesocricerus auratus. Neuroscience 7, 2609-2635.
41. Ricardo J. A. and Koh E. T. (1978) Anatomical evidence of direct projection from the nucleus of the solitary tract to the hypothalamus, amygdala, and forebrain structures in the rat. Brain Res. 153, l-26. 42. Rose M. (1928) Die Insehinde des Menschen und der Tiere. J. Psvchol. Nexrol., 37, 467-624. 43. Russchen F. T. (1982)Amygdalopetal projections in the cat. 1. Co&al afferent connections. A study with retrograde and anterograde tracting techniques. j. camp. Neural. 206, 159-172. 44. Sawr C. B. and Loewv A. D. f 1980) Efferent connections of the narabrachial nucleus in the rat. Brain Res. 197.291-317. 45. Saper C. B. (1982) Convergence of autonomic and limbic connections in the insular cortex of the rat. J. corn>. Neural. 210, 163-173. 46. Saper C. B. (1982) Reciprocal parabrachial-cortical connections in the rat. Bruin Res. 242, 33-40. 47. Schwaber J. S. and Schneiderman N. (1975) Aortic nerve activated cardiovascular neurons and interneurons. Am. J. Ph_vsio/. 299, 783-790.
48. Schwaber J. S., Kapp B. S. and Higgins G. (1980) The origin and extent of direct amygdala projections to the region of the dorsal motor nucleus of the vagus and the nucleus of the solitary tract. Neurosci. Left. 20, 15-20. 49. Schwaber J. S., Kapp B. S., Higgins G. A. and Rapp P. R. (1982) Amygdaloid and basal forebrain direct connections with the nucleus of the solitary tract and the dorsal motor nucleus. J. Neurosci. 2, 1424-1438. 50. Schwaber J. S., Kapp 8. S., Higgins G. A. and Driscoll-Mendes P. A. (1983) Comparison of the topographic distribution of insular cortical and central amygdaloid inputs to the nucleus solitarius and dorsal motor nucleus in the rabbit. Neurosci. Abstr. 9, 113. 51. Shipley M. T. (1982) Insular cortex projections to the nucleus of the solitary tract and brainstem visceromotor nuclei in the mouse. Brain Res. Bull. 8, 139-148. 52. Shipley M. T. and Sanders M. S. (I 982) Special senses are really special: evidence for a reciprocal, bilateral pathway between insular cortex and nucleus parabrachialis. Brain Res. Bull. 8, 493-501. 53. Smith W. K. (1945) Functional significance of rostra1 cingulate cortex as revealed by its responses to electrical excitation. J. Neurophysiol. 8, 241-253. 54. Stock G.. Schlor K. H., Heidt H. and Buss J. (1978) Psychomotor behaviour and cardiovascular patterns during stimulation of the amygdala. PflGgers Arch. ges. physiol. 376, 177-184. 55. Swanson L. W. and Hartman B. K. (1975) The central adrenergic system. An immunofluorescence study of the location of cell bodies and their efferent connections in the rat utilizing dopamine-beta-hydroxylase as a marker. J. camp. Neural. 163, 467-506. 56. Takeuchi Y., McLean J. H. and Hopkins D. A. (1982) Reciprocal connections between the amygdala and parabrachial nuclei: ultrastructural demonstration bv_ degeneration and axonal transport of horseradish perioxidase in the cat. Brain _ Res. 239, 583-588. 57. Veening J. G. (1978) Cortical aBerents of the amygdaloid complex in the rat: An HRP study. Neurosci. Letr. 8, 19l- 195. 58. Wall P. D. and Davis G. D. (1951) Three cerebral cortical systems affecting autonomic function. J. Neurophysiol. 14. 508-5 17. (Accepted 26 November 1984)
Neuroscience Vol. IS. No. 2, pp. 327-346. 1985 Printed in Great Britain
FRONTAL
CORTEX PROJECTIONS TO THE AMYGDALOID CENTRAL NUCLEUS IN THE RABBIT
B. S. KAPP*, J. S. ScHwABERt and P. A. DRISCOLL *Department of Psychology, University of Vermont, Burlington, VT 05405, U.S.A.; and tDupont Central Research and Development, E. I. duPont de Nemours and Company, Wilmington, DE 19036, U.S.A. Abstract-Evidence has recently been presented which demonstrates that the amygdaloid central nucleus projects directly upon cardiovascular/autonomic regulatory nuclei of the dorsal medulla and that in the rabbit this nucleus may influence cardiovascular activity during emotional states. The present study is one of a series of investigations designed to provide information on the innervation of the central nucleus in the rabbit and describes the topography and origin of frontal cortex projections to the nucleus based upon retrograde and anterograde axonal transport techniques. Injections of horseradish peroxidase or the fluorescent dyes, Bisbenzimide or Nuclear Yellow, into the central nucleus resulted in abundant numbers of retrogradely labeled neurons in three regions of the frontal cortex: the insular cortex on the lateral surface and areas 25 and 32 on the medial surface of the hemisphere. The majority of labeled neurons in the insular cortex were located in layer V of the dorsal and posterior agranular insular regions, although labeled neurons were observed in layer V of the granular insular cortex as well as in layers II and III of the posterior agranular insular cortex. Labeled neurons in areas 25 and 32 were located throughout all layers and the total number of these neurons was substantially less than that observed in the insular cortex. Autoradiographic experiments in which amino acids were injected into the insular cortex resulted in a dense pattern of transported label within the central nucleus that extended rostrally into the sublenticular substantia innominata and lateral component of the bed nucleus of the stria terminalis. Label was also observed in the cortical, lateral, basolateral and basomedial amygdaloid nuclei. In contrast to the projections from the insular cortex, amino acid injections into areas 25 and 32 resulted in only relatively light labeling within the most rostra1 region of the central nucleus; otherwise the nucleus was partially encapsulated and virtually devoid of label. These results suggest that the insular cortex possesses the potential to directly influence the central nucleus projection to cardiovascular/autonomic regulatory nuclei of the dorsal medulla and thus, together with the amygdaloid central nucleus, appears to be an important component of a forebrain system involved in cardiovascular/autonomic regulation.
INTRODUCTION Recent evidence suggests that the amygdaloid complex influences autonomic, and in particular cardio-
vascular, regu1ation7.9.“.‘8~“,s4and that the amygdaloid central nucleus and its efferent projection system may provide an anatomical substrate for this influence.‘7.2’.3“.4’.49 For example, our anatomical experiments in the rabbit,4R.49as well as those in the rat.” cat” and monkey,” have demonstrated that the central nucleus projects directly upon cardiovascular regulatory nuclei of the dorsal medulla; the nucleus of the solitary tract, the dorsal motor nucleus of the vagus nerve and the nucleus ambiguus. In the rabbit the cells of origin of these descending projections are primarily located throughout the entire rostralcaudal length of the large, media1 component of the central nucleus and extend rostrally from the nucleus in an uninterrupted continuous region through the sublenticular substantia innominata and into the lateral component of the bed nucleus of the stria terminalis4’ Additional research in the rabbit has demonstrated that (1) electrical stimulation of the central nucleus
*To whom all correspondence
should be addressed. 327
elicits profound heart rate deceleration and depressor2.27 responses as well as activating cardiovascular regulatory interneurons within the nucleus of the solitary tract;” (2) lesions and pharmacological manipulations within the region of the central nucleus attenuate conditioned heart rate deceleration during Pavlovian fear conditioning;‘2.‘3.‘5 and (3) significant increases in central nucleus multiple unit activity emerge in response to a conditioned stimulus over the course of Pavlovian fear conditioning at a time when conditioned heart rate decelerative responses emerge to that stimulus.’ In this context, the present investigation, that of delineating the projections from the frontal cortex to the central nucleus in the rabbit, is one of a series of experiments designed to gain further insight into the function of the central nucleus as one component of a larger forebrain system regulating cardiovascular and autonomic activity during emotional states. The existence of projections from the frontal cortex to the amygdala has been known for some time,i9.)j as has physiological evidence demonstrating autonomic responses to electrical stimulation of the medial and lateral aspects of the frontal cortex, including the lateral and medial orbital, insular and cingulate cortices.‘5.20.24.53”More recent anatomical findings have demonstrated that various regions of
B. S. Kapp ei lil
328 the frontal cortex. including
areas 25 and 32 on the medial surface as well as the insular cortex on the lateral surface of the hemisphere, project to the region of the central nucleus.3.‘J.37.“3.‘sHowever, the extent to which these projections innervate the region of origin of the central nucleus projection to medullary cardioregulatory nuclei, as well as the very nucleus insular cortex-central existence of projections in all species studied, is equivocal. For example, while Saper4’ reported a well-defined projection from the insular cortex to the amygdaloid central nucteus in the rat, Beckstead’ reported that projections from the insular region surrounded, but did not appear to enter, the central nucleus in the rat. Furthermore, Reep and Winans@ did not find evidence for a projection from the insular region to the amygdaloid central nucteus in the hamster. Given the data in rabbit implicating a centrai nucleus function in cardiovascular regulation. the present investigation was performed to provide information on frontal cortex projections to the central nucleus in this species using retrograde and anterograde axonal transport techniques. In particular. the topo~aphy and specific origin of frontal cortex projections to the central nucleus were determined with respect to the medial component of the central nucleus and to the closely related sublenticular substantia innominata and lateral component of the bed nucieus of the stria terminalis, since it is from this continuum that a prominent project&n to medul~ar~ cardiovascular and autonomic regulatory nuclei arises. EXPERIMENTAL PROCEDURES Retrograde transport experiments
Thirty-three pure-bred New Zealand rabbits weighing from 2.0-2.8 kg were pretreated with chlorpromazine hydrochloride (20 mg in 0.8 cc saline i.v.) and anesthetized with pentobarbitai (Nembutal; 30-75 mg, i.v.). They were placed in a Kopf stereotaxic instrument fitted with a rabbit head-holder. Unilateral injections of either horseradish peroxidase fn = ii; Sigma VI; Z-507;; IO-50 ni) or the fluorescent dyes Bisbenzimide (n = 19: Hoechst 33258; 5-IO?;; W60ni) or Nuclear Yefiow (n = 3; Hoechst S 769121: 5%; 40ni) were aimed at the central nucleus using the following co-ordinates: 0.0-0.5 mm anterior to bregma, 5.7 mm lateral to the midline and 12.0-12.5 mm ventral to dura. The head was adjusted such that lambda was I.5 mm below bregma. Injections were delivered via a 26 gauge needle or a glass micropipette (4&6Opm tip diameter) cemented to a 26 gauge needle attached to a 1.Opl Hamilton syringe. All injections were made over a 20-40 min period. The animals were sacrificed with injections of T-61 Euthanasia solution (Hoechst) from 6 h to 3 days following the injection. The horseradish peroxidase-injected animals were perfused and 40pm frozen sections were taken throughout the rostrai-caudai extent of the frontal cortex. The sections were processed using the tetramethyl~n~dine procedure as described previously in the rabbit” and munterstained with Nuclear Red. Sections were viewed under both bright- and dark-field microscopy. Animals receiving Bisbenzimide or Nuclear Yellow were perfused with 1000mi of physiological saline followed by 2000mi of a phosphate-buffered loo/, fonnaiin solution (pH 7.4). The brains were postfixed for I8 h in a 20?/,
sucrose-buffered formalm soiut~on and altern~~r J\ ;I iE frozen sections were collected in drstrllrd water The> WT!. immediately mounted onto uncoated shdes. allowed to stir dry for several hours. and then viewed and photographed using an Olympus BHA photomicroscope equipped for epifluorescence. A 360 nm excitation waveiength (filter I.&1) was used for both Bisbenzimide and Nuclear Yetlou The additional. alternate sections were collected in lo”, formalin and stained with thionin to facilitate the localization of labeled neurons. Line drawings of serial sections were made with the use of a drawing tube (camera lucida) and Bausch and Lomb microprojector. The locations of labeled neurons were plotted with reference to various landmarks and with the aid of the thionin-sIained secttons. as well as additional nissl-stained brains prepared as atlases. Anterograde transport r.yerrmenrs
Based on the results of our retrograde transport expertments. stereotaxic injections of radioactively labeled amino acids were made into the agranular and granular regions of the insular cortex (n = 17) and areas 25 and 32 of the medial prefrontal cortex (11= i6).R Co-ordinates for the agranular insular injections varied from 3.0-5.5 mm anterior to bregma. 4.2-5.7 mm lateral to the midline and 5.5-8.5 mm ventral to dura. Those for areas 25 and 32 were 5.5-8.5 mm anterior to bregrna. 0.7 mm lateral to the midline and 6.0-8.5 mm ventral to dura. Injection volumes varied from 30 to 300111 and were comprised of equal parts of ~3H]proline and [‘Hjieucine f i .OmCi!mi; Amershamf in a concentration of 50 ~Cci;~i. Ail injections were delivered via a glass micropipette (4&6Opm tip diameter) over a 3@-40min period. Survival times ranged from I to 7 days. The animals were then sacrificed and perfused with 1000 ml of saline followed by 2000ml of a IO‘?,;) buffered formalin solution (pH 7.4). Frozen sections were taken at ?5pm. mounted on chrom alum-subbed slides. defatted and coated with Kodak NTB-2 emulsion and exposed from 6 to i? weeks at 4 C. The autoradiographs were then developed with Kodak D-IV developer and stained with thionin. Sections were examined under bright- and dark-field microscopy and the distribution of silver grains within the amygdala was photographed and reconstructed with the atd of additional nissl-stained sections from separate brains prcpared atlases. The te~Ino~ogy and su~ivisi~~ns used to describe the amygdaloid nuciei in this study follows that previously used by us in the rabbit.” The terminoiogy used to describe the insular cortex adheres to that used by RoseJ’ to describe this region in the rabbit. while areas 25 and 32 refer to the frontal corttcal regtons on the medial wall of the hemispheres as described in the rabbit by Brodmann.’ RESLLTS Retrograde
transport
experiments
All three retrograde tracers produced similar patterns of retrograde labeling. Bisbenzimide proved the most useful of the three. producing both the most clearly definable, restricted injection sites and the greatest number of retrogradely labeled neurons. optimal Bisbenzimide neuronal labeling with minimal glial involvement was achieved with 48-72 h survival times. Abundant numbers of retrogradely labeled neurons were located in three frontai corticaf regions: the insular cortex on the lateral surface of the hemisphere and areas 25 and 32 on the medial surface (Fig. I). Labeled neurons were located primarily ipsilateral to the injection site. although a few, faintly labeled neurons were observed on the contralateral side.
Frontal cortex projections to amygdaloid central nucleus particularly within the insular cortex. Of the three frontal cortical regions. the insular cortex contained by far the greatest numbers of labeled neurons while area 32 contained the least. A few labeled neurons were also observed to be scattered within the medial precentral cortex and the dorsal division of the anterior cingulate cortex. The insular cortex in the rabbit, as defined by Rose,@ overlies the claustrum on the lateral surface of the hemisphere and lies dorsal to and within the fundus of the rhinal sulcus. It is comprised of a dorsal granular cortex, containing a well-developed layer IV (GI, Figs 1 and 2) and a ventrally adjacent agranular region (AI, Figs 1 and 2). The agranular region, in turn, has been further divided into three regions: a dorsal agranular insular region (ail), a ventral agranular insular region (ai2) and a posterior agranular insular region (ai3).42 These areas were readily distinguished in the present study. A representative case in which 30 nl of a 10% Bisbenzimide suspension was injected into the rostra1 portion of the central nucleus is illustrated in Fig. 2. The intensely fluorescent injection site was primarily confined to the central nucleus, with some slight encroachment upon the substantia innominata, the ventral globus palhdus and the ventromedial putamen. Within the more rostra1 insula the majority of Bisbenzimide-labeled neurons were located in layer V of the dorsal agranular insula (Figs 2b,c). A few labeled neurons were also observed in layer V of the ventral agranular insula and in layer V of the granular insula lying dorsal to the dorsal agranular insula. More posteriorly, numerous labeled neurons were observed in layer V of the posterior agranular insula and this field of labeled neurons extended dorsally into layer V of the granular insular region (Fig. 2e). In addition, faintly labeled neurons were observed in layers II and III of-the posterior agranular insula although the total number of these neurons was far fewer than those located in layer V (Fig. 2e). On the medial surface of the hemisphere labeled neurons were observed scattered throughout all layers of area 25 while fewer numbers were observed throughout the layers of area 32 (Figs 2b,c). The total numbers and density of labeled neurons in these two areas were substantially less than the numbers observed in the insular cortex.
Anterograde
transport experiments
The insular cortex. The injection sites of tritiated amino acids varied in placement within the dorsal-ventral and rostral-caudal extents of the insular cortex. However, all those injections in which the effective injection site (determined by a homogeneous blackening of the majority of cells due to the density of silver grains) covered at least a portion of layer V of the agranular and granular insular regions resulted in a distinct pattern of transported label within the central nucleus, with no differences in the topography
329
of label within the nucleus as a function of differences in the location of injection sites along the rostral-caudal extent of the insular cortex. _ The injection site from a representative case is shown in Fig. 3. The effective injection site covers the ventral portion of the dorsal agranular insular region (ail), the ventral agranular insular region (ai2) and the lateral edge of the claustrum. Much of the granular insular region was located dorsal to the effective injection site. Labeled fibers destined for the amygdala appeared to course within the internal and external capsules. At the rostra1 border of the amygdala, labeled fibers were observed to exit the internal capsule and course in a ventral and ventrolateral direction to innervate the sublenticular substantia innominata and the amygdala (Fig. 4). Within the amygdala, the central nucleus, including both its medial and lateral subdivisions,49 was dense in silver grain concentration along its entire rostral-caudal extent (Figs 4-7). The central nucleus was the most densely labeled nucleus of the deep nuclei of the amygdaia. An intermediate fiber-rich region of the nucleus appeared somewhat less densely labeled at some levels (Figs 5 and 6). While at more rostra1 levels labeled fibers appeared to stream toward the nucleus from the internal capsule, at more caudal levels a broad band of labeled fibers was observed to be oriented in a dorsolateral to ventromediai direction extending from the external capsule through the ventral putamen where it appeared to enter the central nucleus on its dorsolateral border (Fig. 7). In addition, at caudal levels a diffuse network of labeled fibers, possibly originating from the ventral-most extent of the external capsule, was observed to traverse the anterior pole of the anterior basolateral nucleus and to extend into the ventral border of the nucleus. In addition to the central nucleus, a moderate amount of transported label was also distributed within the lateral component of the bed nucleus of the stria terminalis (Fig. 8). Label was also distributed throughout the rostral
330
8. S. Kapp et al.
Abbreviations used in jigures AC :I ail ai BI Bm BST Ce c”L”
anterior cortical nucleus of the amygdala anterior commissure agranular insular cortex dorsal agranular insular cortex ventral agranular insular cortex basolateral nucleus of the amygdala basomedial nucleus of the amygdala bed nucleus of the stria terminalis central nucleus of the amygdala external capsule claustrum
fx GI GP La Me 01 P PLC PMC PP SI
fomix granular insular cortex globus pallidus lateral nucleus of the amygdala medial nucleus of the amygdala optic tract putamen posterolateral cortical nucleus of the amygdala posteromedial cortical nucleus of the amygdala prepiriform cortex substantia innominata
Fig. 1 (A), Photomicrograph of frontal section of rabbit brain showing the insular and medial prefrontal cortical regions where retrogradely labeled neurons were observed following injections of retrograde tracers into the central nucleus of the amygdala. (B) Higher power photomicrograph of frontal section of insular cortex in the rabbit. Large arrowheads delineate innermost border of Layer V, the layer within which the vast majority of retrogradely labeled neurons were located. Sections are 25 pm and stained with thinonin. Fig. 2. Line drawings of the amygdaloid injection site and of representative frontal sections through the insular cortex from case Bb27 in which Bisbenzimide was injected into the central nucleus of the amygdala. The greatest extent of the injection site is illustrated in (A). The darkened region depicts the intensely fluorescent central area, while the lined region depicts the less intensely fluorescent “halo” region surrounding the central area. The distribution of labeled neurons in the insular and medial prefrontal cortex is illustrated on sections from rostra1 (B) to caudal (E) levels. Each dot represents a single-labeled neuron from that section alone. Figs. 4-7. Bright-(A) and dark-field (B) photomicrographs of representative frontal sections taken at four rostra1 (Fig. 4) to caudal (Fig. 7) levels through the amygdaloid complex of case I-4. Note the highly specific, markedly dense distribution of silver grains along the entire rostral-caudal extent of the central nucleus. Rectangles in Figs 4 and 5 represent areas illustrated in Figs 9 and 10, respectively. Fig. 9. Bright-(A) and dark-field (B) photomicrographs of frontal sections showing the distinct laminar field of silver grains in layer IB of the plexiform layer of the amygdaloid anterior cortical nucleus. Area shown is outlined by rectangle in Fig. 4. Fig. 10. Bright-(A) and dark-field (B) photomicrographs of frontal sections showing the laminar field of silver grains in layer IB of the plexiform layer of the amygdaloid posterolateral cortical nucleus. Area shown is outlined by rectangle in Fig. 5. Fig. 12. Dark-field photomicrographs of frontal sections taken at two levels through the central nucleus of the amygdala of case FM-IO. (A) is more rostra]. Note that silver grains encapsulate the nucleus and that the nucleus itself is relatively devoid of label, as is particularly noteworthy in (B).
B Fig. 1 331
:
GI
;-,‘;. ::‘: ..,
I
.’
32
Al
.. . ”
.:..
-
25 )
B
C
Fig. 2. 332
32
Fig. 3. Photomicrograph
of the largest extent of the injection site (60 nl; 4 day survival; 8 week exposure time) within the insular cortex of case I-4.
333
Fig. 4. 334
Fig. 5. 335
Fig. 6 316
Fig. 7. 337
Fig. 8. Dark-field photomicrograph of frontal section showing the distribution of silver grains within the lateral part of the bed nucleus of the stria terminalis for case l-4.
338
Fig. 9. 339
Fig. 10.
340
Fig. Il. Pha ltomicrograph
of the largest extent of the injection site (100 nl; 7 day survival; time) within the medial prefrontal cortex of case FM-IO.
341
8 week exposure
Fig.
12.
Frontal cortex projections to amygdaloid central nucleus piriform nucleus and layers IB and III of the prepirifo~ cortex were heavily labeled throu~out the entire rostraI
343
appeared destined to innervate nuclei lying ventral to the central nucleus. Although the central nucleus appeared devoid of label along most of its rostral-caudal extent, other nuclei of the amygdala were innervated. At the rostra1 pole of the amygdala label was diffusely distributed within layers II and III of the anterior cortical nucieus, while at more posterior levels label was observed within and surrounding the basolateral nucleus, which was the most heavily labeled region within the amygdala. Lighter amounts of label were present in the basomedial nucleus. The medial, lateral, posteroIatera1 cortical and ~steromed~al cortical nuclei were not labeled nor was significant label present within the prepiriform cortex. DISCUSSION
Insular cortex-central nuclei projection The results of the present study demonstrate that the central nucleus of the amygdala in the rabbit is the recipient of a prominent projection from the insular cortex. The projection heavily innervates the medial component of the central nucleus, the adjacent sublenticular substantia innominata and the lateral component of the bed nucleus of the stria terminalis. These three areas have in turn been observed to form a continuous region in the rabbit from which arises a prominent projection to autonomic regulatory nuclei of the dorsomedial medullaJv This observation has suggested the anatomical unity of this continuum as a single entity,4g a concept supported by the observations that these three structures possess common cytoarchitecture,~.‘6,22 staining properties,‘0”6 neurochemical innervatiotP’ and afferentation from the nucleus of the solitary tract.” The results of the present study demonstrating a projection from the insular cortex to the entire extent of this continuous region are consistent with this notion. Based upon differential thalamic a;ferentation, Guldin and Markowitsch” have recently suggested the heterogeneous nature of the insular cortex, and its division into prefrontal insular, gustatory insular and associative insular regions in the rat. The existence of regional differentiation of the insular cortex in the rabbit based upon thalamocortical afferentation has yet to be determined. Hence the origin of the insular-central nucleus projection in the context of the divisions proposed by Guldin and Markowitschi4 is not known and awaits further research. However, our results demonstrating that the insular projection to the central nucleus arises primarily from layer V of the agranular and granular insular regions along the entire rostral-caudal extent of these regions suggest that the projection may originate from all of the regions delineated by Guldin and Markowitsch.14 The results of the present study are consistent with recent reports describing an insular-amygdala projection in the rat,37”5.57hamster,@ caP3 and mon-
344
B. S. Kapp et al
key.” Inconsistencies exist. however, among these reports concerning the extent to which this projection innervates the central nucleus. For example, the results of the present study are similar to the results of Saper”’ demonstrating the presence of anterogradely transported label within the rat amygdaloid central nucleus, lateral nucleus, basolateral nucleus and lateral component of the bed nucleus of the stria terminalis following horseradish peroxidase injections into the insular cortex. The results of the present study and those of Saper4’ differ, however, from those of Beckstead who reported that tritiated amino acid injections into the insular cortex of the rat produced label mainly in the lateral. basolateral and basomedial nuclei, while labeled fibers surrounded but did not appear to enter the central nucleus. Results similar to those of Beckstead’ have recently been reported by Reep and Winan?’ who found that amino acid injections into the anterior agranular insular regions of the hamster did not result in label within the central nucleus. Several factors may account for the absence of label within the central nucleus as reported by Beckstead’ and Reep and Winans.40 First, as suggested by Saper,” most of the injection sites reported by Beckstead were located in the lateral prefrontal cortex just anterior to the insula and barely encroached upon the anterior insula. Second, the injection sites of Reep and Winans”” were located within the anterior insula. Since our retrograde transport experiments of the present study demonstrated fewer numbers of ‘labeled neurons within the rostral-most portions of the insula (see Fig. 1) and since SaperJ! observed that horseradish peroxidase injections placed more caudally within the insula resulted in a more dense pattern of label within the central nucleus, the relative absence of label within the central nucleus as reported by Reep and Winans” may well be a function of the anterior placement of their injection sites. In any event. the results of the present study clearly demonstrate that in the rabbit the central nucleus receives a significant projection originating along the entire rostralxaudal extent of the insular cortex. The results of the present study also demonstrate a more extensive insular projection to the amygdala than previously reported in studies using anterograde tracing techniques in the rat and hamster.‘~J0~“5 The transported label observed in the present study in the anterior cortical, posteromedial cortical and posterolateral cortical nuclei was not described in these studies. Otterson, however, reported that injections of horseradish peroxidase into the anterior and posterior cortical nuclei resulted in retrogradely labeled neurons in the agranular insular cortex. These findings, as well as the observations by Mufson et al.34 in monkey demonstrating that tritiated amino acids injected into the anterior insula produced widespread labeling in the amygdala, including the cortical nucleus, are consistent with the results of the present study. Hence it appears that the insular cortex
proJection to the amygdala IS widespread. inner\uttng all of the amygdaloid nuclei with the apparent exception of the medtal nucleus and with a particularI> heavy innervation cf the central nucleus.
The results of the present study demonstrate that the projections from the medial prefrontal cortex to the region of the central nucleus differ markedly from the projections from the insular cortex. With the exception of the presence of the diffuse distribution of silver grains overlying a broad area of the anterior amygdala including the most anterior region of the central nucleus, the central nucleus was free of label following injections of amino acids into the medtal prefrontal cortex. Given the presence of labeled fibers and terminals surrounding the central nucleus. it is probable that much of the retrograde label observed in the medial frontal cortex following injections of retrograde tracers into the central nucleus was a function of uptake and transport of label from either injured fibers of passage and:or from terminals surrounding the nucleus. Our observations in the rabbit are similar to those of Beckstead’ who also reported that injections of amino acids into area 32 produced label in the lateral, basolateral and basomedial nuclei while labeled fibers surrounded. but did not enter the central nucleus in the rat. More recently Russchen” has reported a similar result in the cat. An injection of [‘Hlleucine covering area 25 and the caudal part of are3 32 produced label overlymg the basolateral nucleus and a “capsule” surrounding the medial component of the central nucleus. although some label was reported to overlie the medial component.
The functional significance of the Insular projection to the central nucleus is at present a matter for speculation. Our recent research, however. has suggested that the central nucleus may function m vagal modulation of heart rate in the rabbit to the presentation of conditioned fear-arousing stimuli, quite possibly via direct projections from the central nucleus to the nucleus of the solitary tract, the dorsal motor nucleus of the vagus nerve and the nucleus ambiguus.‘6 The latter two regions contain vagal preganglionic cardioinhibitory neurons in the rabbit?3_j- and stimulation of the central nucleus directly influences neurons of the nucleus of the solitary tract which receive baroreceptor input.” The insular cortex projection to the site of origin of the central nucleus descending pathway to medullary autonomic regulatory nuclei endows the insular cortex with the potential to exert a significant influence on the outflow of the central nucleus in its functional role in cardiovascular and autonomic regulation. Indeed recent experiments in both the anesthetized and awake rabbit38 (C. G. Markgraf and B. S Kapp
Frontal cortex projections to ~ygdaloid unpublished observations) have demonstrated that stimulation at sites within &he insular cortex elicits profound bradycardia and depressor responses, a response pattern identical to that observed upon stimulation of the central nucleus in the rabbit.227 The extent to which the responses from st~muiation of the insula are mediated via the central nucleus~orsa~ medulla projection and/or via direct projections from the insula to the dorsal medulla24.s0 in the rabbit awaits further research. Nevertheless, the results of the present experiment, taken together with recent findings of connections of the insu!ar
central nucleus
345
cortex and of the amygdaloid central nucleus with other areas implicated in autonomic function, including the parabrachial complex,6~29~~~M-46~s2~s6 the nucleus of the solitary tract, the dorsal motor nucleus of the vagus nerve and the nucleus ambiguuszi~2s~33~39~4s~49~~~~’ and the laterai h~othalamus4.29.*.4s suggest the inclusion of the insular cortex and amygdaloid central nucleus in a forebrain system concerned with cardiovascular/autonomic regulation. Acknowledgements-This
work was supported by USPHS
Grant NSI 607.
REFERENCES 1. Applegate C. D., Frysinger R. C., Kapp B. S. and Gallagher M. (1982) Multiple unit activity recorded from amyg~aia central nucleus during Pavlovian heart rate conditioning in rabbit. Brain Res. 238, 457-462. M. D. and McNall C. L. (1983) Autonomic and somatomotor effects of 2. Applegate C. D., Kapp B. S., Unde~~ amygdala central n. stimulation in awake rabbits. Physioi. Behau. 31, 353-360. 3. Beckstead R. M. (1979) An autoradiographic examination of corticocortical and subcortical projections of the mediodorsal-projection (prefrontal) cortex in the rat. J. camp. Neurol. 184, 43-62. 4. Berk M. L. and Finkelstein J. A. (1982) Efferent connections of the lateral hypothalamic area of the rat: an autoradiographic investigation. Brain Res. Bull. 8, 51 I-526. 5. Berkelbach van der Sprenkel H. (1926) Stria terminalis and amygdala in the brain of the opossum (Didelphis rirginianu). J. romp. Neuroi. 42, 211-254.
6. Block C. H. and Schwartzbaum J. S. (1983) Ascending efferent projections of the gustatory parabrachial nuclei in the rabbit. Brain Res. 259, 1-9.
1. Bonvallet M. and Gary Bobo E. (1972) Changes in phrenic activity and heart rate elicited by localized stimulation of amygdala and adjacent structures. Eiectroenceph. clin. Neurophysiol. 32, I-16.
8. Brodmann K. (1909) Verg~ei~hende ~kat~a(~onsiehre der Grosshirnrinde in ihren Prinzipien dargesteltr a$ Grund des Zellenbaues. J. A. Barth. Leipzig.
9. Cechetto D. and Calaresu F. R. (1983) Single units in the amygdala of the cat responding to activation of baroreceptors and chemoreceptors.
Neurosci. Absrr. 9, 612.
10. DeOlmos J. S. (1972) The amygdaloid projection field in the rat as studied with the cupric-silver method. In 7&r Neurobiolonv fed. Eleftheriou 3. E.), up. .,. of” rhe Amxedula . ._ . . 145-204. Plenum Press. New York. Il. Faiers A. A., Calaresu F. R. and Mogenson G. J. (1975) Pathway mediating hypotension elicited by stimulation of the amvadala in the rat. Am. J. Phvsiof. 22%. 1358-1364. M., Kapp B. S.. Frysingkr R. C. and Rapp P. R. (1980) /?-Adrenergic manipulation in amygdala central n. alters rabbit heart rate conditioning. Pharmac. Biuchem. Behao. 12, 419-426. 13. Gallagher M., Kapp B. S.. McNall C. L. and Pascoe 3. P. (1981) Effects of opiate administration into the amygdala on Pavlovian heart rate conditioning in the rabbit. Pharmac. Biochent. B&XL 14, 496-505. 14. Guldin W. 0. and Markowitsch H. J. (1983) Cortical and thalamic tierent connections of the insular and adjacent cortex of the rat. J. camp. Neural. 215, 135-153. 15. Hall R. E. and Comish K. (1977) Role of the orbital cortex in cardiac dysfunction in unanesthetized rhesus monkey.
12. Gallag&
Expl Neural. 56, 289-297.
16. Heimer L. (1978) The olfactory cortex and the ventral striatum. In Limb& mechanisms (eds Livingston K. and Hornykiewics 0.). pp. 95-187. Plenum Press, New York.
17. Higgins G. A. and Schwaber J. S. (1983) Somatostatinergic projections from the central nucleus of the amygdala to the vagal nuclei. Peptides 4, l-6.
18. Hilton S. M. and Zbrozyna A. W. (1963) Amydaloid region for defense reactions and its efferent pathway to the brain stem. J. Physiof., Land. 165. 16U-I 73.
19. Hirata Y. (1965) Subcortical projections from the orbital surface of the cat brain. Acta med. stand. 13, 123-142. 20. Hoffman B. L. and Rasmussen T. (1953) Stimulation studies of insular cortex of Macuea mulatta. J. Neur(~ph,vsio~.16, 343-35 I. 21. Hopkins D. A. and Holstege G. (1978) Amygdaioid projections to the mesencephalon, pons and medulla oblongata in the cat. Expr Brain Res. 32, 529-547.
22. Johnston J. 3. (1923) Further contributions to the study of the evolution of the forebrain. J. camp. Neurof. 35.337-481. 23. Jordan D.. Khalid M. E. M., Schneiderman N. and Spyer K. M. (1982) The location and properties of preganglionic neurones in the rabbit. Ppiigers Arch. ges. physiol. 395, 244-250. autonomic and el~tr~ortico~aphic responses to electrical stimu~tion of “rhinencephalic” and other structures in primates. cat and dog. A study of responses from the limbic. subcdllosal. orbito-insular, piriform and temporal cortex, hippocampus-fornix and amygdala. Acta physiol. scund. 24, Suppl. 83, l-285. 25. Kapp B. S.. Frysinger R. C., Gallagher M. and Haselton J. (1979) Amygdala central nucleus lesions: effects on heart rate conditionin in the rabbit. Physiol. Behm. 23, 1109-1117. 26. Kapp B. S., Gallagher M., Applegate C. D. and Frysinger R. C. (1982) The amygdala central nucleus: contributions to conditioned cardiovascular responding during aversive Pavlovian conditioning in the rabbit. in Condjrion~n~: Representation of Inr&ed Neural &zctiok (ed. woody, C. D.), pp. 577-596. Plenum Press, New York. .’ 27. Kapp B. S., Gallagher M., Underwood M. D., McNall C. L. and Whitehom D. (1982) Cardiovascular responses elicited by electrical stimulation of the amygdala central nucleus in the rabbit. Bruin Res. 234, 251-262. vagal cardiomotor
24. Kaada 3. R. (1951) Somato-motor,
346
B. S. Kapp et 01.
28. Kapp B. S., Schwaber J. S. and Driscoll-Mendes P. A. (1983) The organization of insular cortex projections to the amygdaloid central nucleus and autonomic regulatory nuclei of the dorsal medulla. Sot. Neurosci. Absrr. 9, 113. 29. Krettek J. E. and Price J. L. (1978) Amygdaloid projections to subcortical structures within the basal forebram and brainstem in the rat and cat. J. romp. Neurol. 178, 225-254. 30. Lindvall 0. and Bjdrklund A. (1974) The organization of the ascending catecholamine neuron system in the rat brain as revealed by the glyoxylic acid fluorescence method. Acra physiol. scund. Suppl. 412, l-48. 31. Mogenson G. J. and Calaresu F. R. (1973) Cardiovascular responses to electrical stimulation of the amygdala m the rat. Expl Neural. 39, 166180. 32. Moruzzi P., Schwaber J. S., Spyer K. M. and Turner S. A. (1984) Amygdaloid influences on brainstem neurones m the rabbit. J. Physiol., Lond. 66. 33. Mraovitch S., Ruggiero D. A., Ross C. A. and Reis D. J. (1983) Direct projections to autonomic centers of forebrain and brainstem from a cortical vasopressor area in rat. Neurosci. Absrr. 8, 77. 34. Mufson E. J., Mesulam M.-M. and Pandya D. N. (1981) Insular interconnections with the amygdala in the rhesus monkey. Neuroscience 7, 1231-1248. 35. Naura W. J. H. (1961) Fibre degeneration following lesions of the amygdaloid complex in monkey. J. Anar. 95,5 15-53 I. 36. Novotny G. E. K. (I 977) A direct ventral connection between the bed nucleus of the stria terminalis and the amygdaloid complex in the monkey (Macacu fuscicularis). J. Hirnforsch. 18, 271-284. 37. Otterson 0. P. (1982) Connections of the amygdala of the rat. IV. Corticoamygdaloid and intra-amygdaloid connections as studied with axonal transport of horseradish peroxidase. J. camp. Neural. 205, 30-48. 38. Powell D. A., Hemandez L. L. and Buchanan S. L. (1983) Insular cortex participates in cardiovascular inhibition but is not necessary for Pavlovian conditioned bradycardia in rabbits. Neurosci. Absfr. 9, 106. 39. Price J. L. and Amaral D. G. (1981) An autoradiographic study of the projections of the central nucleus of the monkey amygdala. J. Neurosci. 1, 1242-1259. 40. Reep R. L. and Winans S. S. (1982) Efferent connections of dorsal and ventral agranular insular cortex in the hamster. Mesocricerus auratus. Neuroscience 7, 2609-2635.
41. Ricardo J. A. and Koh E. T. (1978) Anatomical evidence of direct projection from the nucleus of the solitary tract to the hypothalamus, amygdala, and forebrain structures in the rat. Brain Res. 153, l-26. 42. Rose M. (1928) Die Insehinde des Menschen und der Tiere. J. Psvchol. Nexrol., 37, 467-624. 43. Russchen F. T. (1982)Amygdalopetal projections in the cat. 1. Co&al afferent connections. A study with retrograde and anterograde tracting techniques. j. camp. Neural. 206, 159-172. 44. Sawr C. B. and Loewv A. D. f 1980) Efferent connections of the narabrachial nucleus in the rat. Brain Res. 197.291-317. 45. Saper C. B. (1982) Convergence of autonomic and limbic connections in the insular cortex of the rat. J. corn>. Neural. 210, 163-173. 46. Saper C. B. (1982) Reciprocal parabrachial-cortical connections in the rat. Bruin Res. 242, 33-40. 47. Schwaber J. S. and Schneiderman N. (1975) Aortic nerve activated cardiovascular neurons and interneurons. Am. J. Ph_vsio/. 299, 783-790.
48. Schwaber J. S., Kapp B. S. and Higgins G. (1980) The origin and extent of direct amygdala projections to the region of the dorsal motor nucleus of the vagus and the nucleus of the solitary tract. Neurosci. Left. 20, 15-20. 49. Schwaber J. S., Kapp B. S., Higgins G. A. and Rapp P. R. (1982) Amygdaloid and basal forebrain direct connections with the nucleus of the solitary tract and the dorsal motor nucleus. J. Neurosci. 2, 1424-1438. 50. Schwaber J. S., Kapp 8. S., Higgins G. A. and Driscoll-Mendes P. A. (1983) Comparison of the topographic distribution of insular cortical and central amygdaloid inputs to the nucleus solitarius and dorsal motor nucleus in the rabbit. Neurosci. Abstr. 9, 113. 51. Shipley M. T. (1982) Insular cortex projections to the nucleus of the solitary tract and brainstem visceromotor nuclei in the mouse. Brain Res. Bull. 8, 139-148. 52. Shipley M. T. and Sanders M. S. (I 982) Special senses are really special: evidence for a reciprocal, bilateral pathway between insular cortex and nucleus parabrachialis. Brain Res. Bull. 8, 493-501. 53. Smith W. K. (1945) Functional significance of rostra1 cingulate cortex as revealed by its responses to electrical excitation. J. Neurophysiol. 8, 241-253. 54. Stock G.. Schlor K. H., Heidt H. and Buss J. (1978) Psychomotor behaviour and cardiovascular patterns during stimulation of the amygdala. PflGgers Arch. ges. physiol. 376, 177-184. 55. Swanson L. W. and Hartman B. K. (1975) The central adrenergic system. An immunofluorescence study of the location of cell bodies and their efferent connections in the rat utilizing dopamine-beta-hydroxylase as a marker. J. camp. Neural. 163, 467-506. 56. Takeuchi Y., McLean J. H. and Hopkins D. A. (1982) Reciprocal connections between the amygdala and parabrachial nuclei: ultrastructural demonstration bv_ degeneration and axonal transport of horseradish perioxidase in the cat. Brain _ Res. 239, 583-588. 57. Veening J. G. (1978) Cortical aBerents of the amygdaloid complex in the rat: An HRP study. Neurosci. Letr. 8, 19l- 195. 58. Wall P. D. and Davis G. D. (1951) Three cerebral cortical systems affecting autonomic function. J. Neurophysiol. 14. 508-5 17. (Accepted 26 November 1984)