- Email: [email protected]
Ascending efferent projections of the gustatory parabrachial nuclei in the rabbit
Brain Research, 259 (1983) 1-9 Elsevier Biomedical Press
1
Research Reports
Ascending Efferent Projections of the Gustatory Parabrachial Nuclei in the Rabbit CHRISTINE H. BLOCK and J. S. SCHWARTZBAUM Department of Psychology, University of Rochester, Rochester, N Y 14627 (U.S.A.) (Accepted June 29th, 1982) Key words: rabbit - - parabrachial nuclei - - gustation - - amygdala - - substantia innominata - - thalamus
Neuroanatomical and electrophysiological techniques were used to study efferent projections of the gustatory region of the parabrachial nuclei (PBN) in rabbits. Autoradiographic analysis of [SH]leucinemicroinjected into electrophysiologicallydefined gustatory areas of the PBN revealed strong projections to the ventromedial taste component of the ventrobasal complex of the thalamus and also evidence of projections to the lateral hypothalamus, substantia innominata and nucleus centralis amygdala (ACE). Electrical stimulation in the region of ACE was shown to activate antidromically neurons in the PBN responsive to sapid stimuli, confirming gustatory projections to the ventral forebrain. An electrophysiologicalmapping of units in the amygdala and substantia innominata responsive to electrical stimulation of the gustatory part of the PBN agreed with the autoradiographic data. A proportion of these units were also responsive to gustatory stimuli. However, latencies of evoked response to stimulation of the PBN were unusually long. Outputs from the gustatory PBN to the ventral forebrain may include other modes of polysynaptically integrated information that are largely controlled by interneuronal networks within the PBN. INTRODUCTION The parabrachial nuclei (PBN) in the pontine region of the brain stem participate in the transmission of gustatory information to the forebrain. The PBN in rats, which receive input from the gustatory part of the nucleus of the tractus solitarius (NTS) la, were shown through autoradiographic analysis 10 to project to the ventromedial taste nucleus of the thalamus (VMT) and to a terminal field in the ventral forebrain that included the lateral hypothalamus, substantia innominata and nucleus centralis amygdala (ACE). Further electrophysiological tests (refs. 9, 10) confirmed a gustatory contribution to some of these ventral forebrain projections. Electrical stimulation of both VMT and A C E were shown to activate antidromically neurons in the PBN sensitive to sapid stimuli. In some cases, the same units responded to the two sites of stimulation suggesting closely interlocked bifurcating gustatory projections to the thalamus and ventral forebrain. More recent anatomical studies, using anterograde and retrograde labeling techniques, have verified this divergent pattern of projections from the 0006-8993/83/0000-0000/$03.00 © 1983 Elsevier Biomedical Press
PBNT,8,15,z4,26, but they have also raised questions about the ftmctional nature of the projections to the ventral forebrain. Little evidence was found of double-labeling of PBN cells in rats by retrograde tracers applied to ACE and VMT 24. The projections to these divergent structures also appeared to originate to some degree from different ftmctional parts of the PBN z4. While projections to VMT arose from the medial part of the PBN, which is linked to the gustatory segment of NTS, projections to ACE seemed to originate largely from the more lateral part of the PBN which receives primary input from the visceral segment of NTS 7,11. Thus, the PBN might convey mainly visceral rather than gustatory information to A C E and the ventral forebrain. The present study was undertaken to analyze specifically the gustatory projections of the PBN in the rabbit. Previous work had demonstrated a gustatory zone in the PBN of this species3, 21,e6. A 3-fold approach was followed. First, anterograde labeling was utilized to trace the ascending projections from the electrophysiologically defined gustatory region of the PBN. Second, electrophysiological tests were conducted to determine whether or not gusta-
tory neurons in the PBN could be antidromically activated by stimulation of ACE as seen in the ratg, 1°. Third, an electrophysiological mapping was made of the distribution of cells in the amygdala and substantia innominata that were: (a) responsive to stimulation of the PBN; and (b) responsive to gustatory stimulation. MATERIALS AND METHODS
Subject and general test procedures Twenty-three male, New Zealand rabbits weighing 1.8-2.7 kg were prepared surgically for later electrophysiological and/or neuroanatomical study. The skull was positioned with bregma 3.5 mm above lambda to provide vertical access to the PBN. Details of the surgical procedure are described elsewhere 3. Experimental tests were initiated after at least 1 week post-operative recovery and, for the electrophysiology, were repeated at weekly intervals. The animal was re-anesthetized with a 4.5 ~ mixture of halothane and oxygen administered through a face mask. Following a supplementary injection of sodium pentabarbital ( ~ 20 mg i.v.) to ensure relaxation of jaw and lingual muscles, the trachea was intubated. The animal was then positioned on a heating pad and its head was painlessly immobilized for stereotactic approach to the brain 3. Anesthesia was maintained with a 1.5-2.0 ~ halothane mixture at a flow rate of 0.4-1.0 liter/min. Electrophysiological methods are described in other reportsZ, 19. Taste methodology is also detailed elsewhere 3. Four basic types of taste stimuli were used in addition to a 0.01 M NaCI rinse solution: 0.75 M NaC1, 0.03 M HCI, 0.5 M sucrose, and 0.01 M QHCI. Neuroanatomy An autoradiographic analysis was made of the distribution of L-[4,5-ZH(N)]leucine (spec. act. 56.5 Ci/mmol; New England Nuclear) microinjected into the gustatory region of the PBN in 11 rabbits. Electrophysiological tests were first carried out to locate the gustatory part of the PBN 3. A 1.0/tl Hamilton syringe (23-gauge injection needle) filled with [ZH]leucine and calibrated stereotactically was then positioned in the PBN taste area, A volume of 0.25-0.35 #1 of [3H]leucine, concentrated to
i0/~Ci/#l in 0.9 ~ NaC1, was injected over a 30-min period. The injection needle was left in place for an additional 15 min. All injections were unilateral. The animals were sacrificed 3-6 days following injection 2. They were anesthetized with sodium pentabarbital and perfused intracardially with 0.9 NaCI, followed by l 0 ~ formalin in 0.9 ~ NaCl. The brains were embedded in paraffin (Paraplast) and cut on a microtome at I0 #m thickness. Three consecutive sections were saved at 80/~m intervals and were mounted on glass slides. Autoradiographic procedure was applied to 2 of each 3 sections. The slides were coated with Kodak NTB-2 emulsion and stored in a light-tight box containing a dessicant at 0 °C for 4 weeks. The slides were then processed in Kodak D-19 developer at 14 °C and Ektaflo fixative at 14 °C. They were stained lightly with cresyl violet and analyzed microscopically with dark-field and bright-field condensers.
Antidromic activation of PBN gustatory units A pair of stimulation leads was implanted chronically in ACE in two preparations. The electrodes were constructed from 0.13 mm tungsten wires with the cross-cut tip left tminsulated. The tips were positioned at the same depth, about 1 mm apart along the A-P plane. To aid in the placement of the leads, recording tests were first carried out with flash stimuli to locate the dorsolateral edge of the optic tract at a mid-A-P level of the amygdala. The stimulation leads were then aimed for the dorsal part of the ACE at 1.5 mm lateral and 1.5 mm above the dorsolateral boundary of the optic tract. Following localization of the PBN gustatory area, conventional criteria were used to evaluate antidromic activation of PBN units: (a) relatively short, invariant latency of evoked response; (b) ability of the unit to follow high frequency pulse trains; and (c) susceptibility of the evoked response to collision from spontaneous discharges. These criteria, though not infallible, are generally reliable 5. Symmetrical biphasic pulses (each 0.25 ms duration, about 8.0 V amplitude) to the ACE were delivered through stimulus isolation units (Model PC-1 W-P Instruments) at a frequency of 1/3 s. Pulse trains consisted of 3 pulses delivered at a frequency of 100 Hz. In the collision tests, a spike-amplitude discriminator with a delay circuit was used to control the electrical
stimulation. The interval between a detected spontaneous discharge and the output pulse to the stimulators was set at about half the latency value of the evoked response. The tests for gustatory responsivity have been described previously a.
Ventral forebrain units responsive to stimulation of the P B N Following localization of the PBN gustatory area, concentric stimulation leads were positioned in the gustatory PBN. The leads consisted of a low-impedance etched tungsten wire cemented within a 25gauge thin-wall, stainless-steel tubing that contained a plastic Kapton tubing insert. The tubing was insulated with Epoxylite to within about 1 mm of the base. The tip of the tungsten wire was spaced 3-4 mm below the tubing to minimize damage to the PBN area. Cathodal pulses or pulse trains were applied to the deeper electrode (0.25 ms pulse duration, 150 #A constant current). Polarity of stimulation was sometimes reversed to check on the effective site of stimulation. The stimulation leads were not chronically implanted in the PBN because of damage produced by movement of the brain when the animal was unrestrained between test sessions. The electrodes were repositioned in the PBN at about the same stereotactic coordinates at the start of each test session. In most instances, recordings
made through the deep electrode confirmed the positioning within the gustatory region of the PBN. A systematic exploration was made of the amygdala and substantia innominata in 10 animals to map the distribution of traits responsive to either pulse or pulse-train stimulation of the gustatory PBN. Responsive units were then tested for gustatory sensitivity. In some instances, units unresponsive to PBN stimulation were also checked for gustatory sensitivity. At the conclusion of one or more electrode penetrations during a test session, a marker lesion was placed above the recording sites by passing 20 #A cathodal DC through the electrode for 30 s. Stimulation sites were marked with currents of 50 #A for 30 s. Histological procedures are described elsewhere a. RESULTS
Autoradiographic analyses Fig. 1 illustrates the PBN injection sites for the [3H]leucine in 9 of the 11 preparations; histology was inadequate for 2 of the preparations. These results are superimposed upon a histological reconstruction of recording sites of gustatory units in rabbit PBN derived from earlier work 3. The gustatory region of the PBN in the rabbit encompasses much of the medial nucleus lying below the brachi-
': .J'--
/
]
i
Fig. l. Injection sites of [aH]leucine (indicated by large circled areas) superimposed upon electrophysiological mapping of the gustatory area within the PBNa. The small filled circles indicate the location of gustatory units displaying 'excitatory' responses; the small open circles refer to gustatory units displaying 'inhibitory' responses. Two of the injection sites in the second most rostral section overlapped one another.
E
F
G
Fig. 2. Diagrammatic representation of the ascending distribution of labeled fibers resulting from an injection of [aH]leucinein the medial part of the PBN (A). Stippling indicates labeled fibers in the central tegmental tract (B and C), ventromedial part of the ventrobasal thalamic complex and region dorsal to the substantia nigra (D), lateral hypothalamus (E and F), and substantia innominata and nucleus centralis amygdala (F and G). um conjunctivum and extends into tbe medial part of the lateral nucleus overlying the brachium. Gustatory cells were not found in the lateral half of the lateral nucleus nor in the Kolliker-Fuse nucleus along the lateral edge of the brachium. Seven of the microinjections of amino acid were located in the rostral half of the PBN gustatory area (3 most rostral sections, Fig. 1). They included medial parts of the medial and lateral nuclei and, in two cases, extended into the mesencephalic nucleus of the trigeminal nerve. The ascending projections traced from these relatively rostral PBN injection sites were generally consistent witll one another and had no relation to the involvement of the mesencephalic nucleus. Fig. 2 illustrates diagrammatically the projections traced in one of these preparations. Fibers from rostromedial cells in the PBN (Fig. 2A) congregate in the area of the central tegmental tract, ventrolateral to the central gray (Fig. 2B). Some of the fibers apparently cross the midline shortly after leaving the PBN because there is also a smaller contralateral component 10. At the mesencephalic level of the posterior commissure (Fig. 2C), the pathway begins to diverge. A major contingent of fibers
ascends dorsally to the caudal ventromedial component of the ventrobasal thalamic nucleus (Fig. 2D). A dense concentration of labeled fibers was invariably found within this subnucleus ipsilateral to the injection site. There is a sparser projection to the contralateral thalamic structure. A second contingent of fibers in the mesencephalon courses ventrally above the substantia nigra and subthalamic nucleus (Fig. 2D). These fibers can be traced into the lateral hypothalamus where some appear to terminate (Fig. 2E and F). In 3 of the 7 preparations with mediorostral injections of the PBN, labeled fibers in the lateral hypothalamus could be followed dorsolaterally above the optic tract into the substantia innominata and dorsal part of the nucleus centralis amygdala (Fig. 2F and G). This distribution of fibers to the ventral forebrain was also predominately ipsilateral. Fig. 3 shows some representative photomicrographs of the anatomical material to illustrate pathways and distribution of labeled fibers for a rostromedial injection site in the PBN. In two preparations, the microinjection of amino acid was located in the mediocaudal part of the gustatory PBN (2 most caudal sections, Fig. 1). These
injection sites only labeled cells that projected to the ventromedial component of the ventrobasal thalamus, predominately on the ipsilateral side. In contrast to the mediorostral PBN injections, no fibers could be traced into the ventral forebrain. These results suggest that virtually all parts of the electrophysiologically defined gustatory region of the PBN project heavily to the ventromedial subnucleus of the thalamus involved in gustationl, 4.
Projections to the ventral forebrain, particularly the substantia innominata and amygdala, may have a more delimited origin within the gustatory PBN or may simply involve fewer elements. Antidromic activation of PBN gustatory units Stimulation leads were histologically confirmed in the ACE in one preparation, and in the ACE and substantia innominata in the second preparation.
J
B
C
7,
Fig. 3. Dark-field photomicrographs of labeled fibers resulting from an injection of [aH]leucine in the medial part of the PBN. The apparent label in the area medial to vb (Panel C) is a condensor artifact. Magnification 50 x .
1 NaCl
HCI *
S u c r o se
OHCl
2
41'
3 4"
•
•
•
Fig. 4. Example of antidromic activation of a gustatory neuron in the PBN by electrical stimulation of the nucleus centralis amygdala (ACE). Peristimulus histograms on the left illustrate the response pattern of the unit to the 4 taste stimuli (time bin = 256 msec; arrow indicates onset of taste stimulation for 6.0 s, followed by 10.0 s interval and 10.0 s of 0.01 M NaCI rinse solution). Oscilloscope tracings on the right illustrate: (l) unit response to repeated single pulse stimulation of ACE; (2) collision blockade of evoked unit response; and (3) evoked unit response to each pulse of a high frequency pulse train. Time scale ~ 2 ms for (1) and (2) and 10 ms for (3). Filled circles denote stimulus artifact.
Electrophysiological analysis of the PBN in these animals confirmed the presence of units that: (a) responded antidromically to stimulation of these ventral forebrain structures; and (b) showed gustatory sensitivity. Seven units met these criteria. Fig. 4 illustrates the findings for one of the units. As shown in Panel I, single pulse stimulation in the ACE evoked responses in a small cluster of units including a unit with a relatively large (upward) positive component. The latency of response for this unit
was about 4 ms and was invariant with repeated stimulation. For the 7 units, latencies averaged about 5 ms. In testing for collision, the pulse stimulation was delayed about 2 ms after spontaneous discharge of the positive-going unit. The resultant blockade of the evoked response (Panel 2) satisfied the collision criterion; response of the two small negative-going units persisted. The unit was also capable of following relatively high-frequency (100 Hz) pulse trains (Panel 3). Tests with gustatory
stimuli (Fig. 4 PSTHs) revealed a responsivity of the unit to NaCI, sucrose and HCI. These findings, which confirm observations in the rat a, demonstrate that at least some gustatory neurons in the PBN project directly to the ventral forebrain.
symbol in Fig. 5) showed gustatory sensitivity of which 6 also responded to stimulation of the PBN. The gustatory units, which constituted about 14% of the sample, were all located within the substantia innominata and ACE. For two of the gustatory units, only the pulse trains were effective in evoking responses. The pattern of 'excitatory' or 'inhibitory' response evoked by the gustatory stimuli agreed with the pattern of evoked response to the PBN stimulation. The 0.75 M NaC1 and 0.03 M HCI appeared to be the most effective gustatory stimuli. The latencies of evoked response to stimulation of the PBN were, however, unusually long. As shown in Table I, only about 25 ~ of the units had latencies that were less than 20 ms (units symbolized by open circle in Fig. 5). All of these un.its were located in the substantia innominata and dorsal part of the ACE.
Topography of ventral forebrain units responsive to stimulation of the PBN The results from 10 animals are summarized in Fig. 5. A total of 103 units were isolated in the amygdala and substar~.tia innominata that responded to stimulation of tb.e PBN and/or gustatory stimuli. Most of the responsive units were located in the medial and sublenticular parts of the substantia innominata and in the dorsal half of the ACE. A scattering of responsive units was found in deeper amygdaloid nuclei. Eleven units (indicated by X
CIL)
._.
" -
'
,
.<---~..~ o
,
°
,'
'
-
.
:
~
,
F I 0
. . . .
!
.
.
5
Fig. 5. Distribution of units in the substantia innominata and amygdala that were responsive to electrical stimulation of the PBN and/or gustatory stimuli. X indicates units responsive to gustatory stimuli. Open circles denote units with latency of evoked response to stimulation of the PBN less than 20 ms; filled circles denote units with latency of evoked response greater than 20 ms. The sections, 0.5 mm apart on the A - P plane, are modified from the Atlas of Urban and Richard ~a.
8
TABLE I Distribution of latencies of evoked responses
n = 103. Latency range (ms)
Percentage of total
0-10 11-20 21-30 31-40 41-50 51 q-
5.5 % 17.6~ 14.3 26.3 13.2 23.1
Nevertheless, more than half of the units in these structures had latencies in excess of 20 ms. The same wide range of latencies applied to the units responsive to the gustatory stimuli and stimulation of the PBN. Taken at face value, these findings would suggest relatively few monosyn.aptic projections from the gustatory zone of the PBN to neurons in the ventral forebrain. DISCUSSION The present findings provide evidence of divergent gustatory projections from the PBN to the ventromedial taste nucleus of the thalamus and ventral forebrain, in agreement with anatomical and electrophysiological observations in ratsg,a0, a2. The results argue against the notion of a gustatory-visceral dichotomy in the information conveyed by the PBN, respectively, to the thalamus and ventral forebrain 24. But, by the same token, the data raise some questions about the anatomical and functional equivalence of these projections. The autoradiographic data revealed projections to taste thalamus from all injection sites within the electrophysiologically defined gustatory region of the PBN. The projections to the ventral forebrain, particularly amygdala and substantia innominata, were less consistent across injection sites. This would suggest a less uniform distribution of medially located PBN neurons and/or fewer neurons in this region that project to the ventral forebrain compared to taste thalamus. Retrograde labeling of PBN cells following tracer injections into ACE and taste thalamus in rats z4 in line with other findings s, 14 suggests such a disparity of projections for the medial gustatory area of the PBN.
The topographical distribution of units in the ventral forebrain that were responsive to stimulation of the gustatory region of the PBN agreed with the autoradiographic data. Responsive neurons were concentrated in the substantia innominata and ACE. However, the electrophysiology revealed few apparent monosynaptically driven cells and implicated only a small proportion of the output from medial PBN to ventral forebrain in gustatory functions. Although these findings may reflect in part thalamocortical and diencephalic pathways to the ventral forebrain activated by the PBN stimulation 6, 13,16,17,z2 we suspect that most of the transmission lag arises within PBN and relates to polysynaptic integration of sensory input within PBN. Electrophysiological unit analysis of gustatory PBN in the awake behaving rabbiOS, 20 has revealed evidence of hedonic integration of gustatory input within PBN. This integrated input may be utilized in the control of orolingual motor activity since other types of elements show relationships to fluid input and motor output. Some of these elements may also transmit information to the ventral forebrain. They can be antidromically activated by stimulation of the ventral forebrain 20. The gustatory part of PBN may, in effect, provide the ventral forebrain with a variety of polysynaptically integrated information pertaining to fluid input and orolingual movement and may not simply relay gustatory information. Whether this expanded role also applies to output directed at taste thalamus is tmclear. The transmission of gustatory messages through PBN to taste cortex is rapid. Electrical stimulation of peripheral taste nerves in rabbits evokes responses in gustatory cortex with average latencies of less than 15 ms 25. ACKNOWLEDGEMENT This research was supported by National Institutes of Health Grant N I M H 14594-10. Abbreviations used in figures: ACE bc ci ct fr FR GP
nucleuscentralis arnygdala brachium conjunctivum internal capsule central tegmental area fasciculusretroflexus reticular formation globus pallidus
HYL IC LL MB MesV MG ml nat MV
lateral hypothalamus inferior colliculus lateral lemniscus mammillary bodies mesencephalic trigeminal nucleus magnocellular nucleus, basal forebrain medial lemniscus mammillothalamic tract motor trigeminal nucleus
REFERENCES 1 Benjamin, R. M., Some thalamic and cortical mechanisms of taste. In Y. Zotterman (Ed.), Proceedings of the First International Symposium on Olfaction and Taste, Pergamon Press, Oxford, 1963, pp. 309-330. 2 Cowan, W. M., Gottlieb, D. I., I-Iendrickson, A. E., Price, J. L. and Woolsey, T. A., The autoradiographic demonstration of axonal connections in the central nervous system, Brain Research, 37 (1972) 21-51. 3 DiLorenzo, P. M. and Schwartzbaum, J. S., Coding of gustatory information in the pontine parabrachial nuclei of the rabbit: magnitude of neural response, Brain Research, in press. 4 Emmers, R., Separate relays for tactile, pressure, thermal and gustatory modalities in the cat thalamus, Proc. Soc. exp. BioL Med., 121 (1966) 527-532. 5 Fuller, J. H. and Schlag, J. D.~ Determination of antidromic excitation by the collision test: problems of interpretation, Brain Research, 112 (1976) 283-298. 6 Krieger, M. S., Conrad, M. C. A. and Pfaff, D. W., An autoradiographic study of the efferent connections of the ventromedial nucleus of the hypothalamus, J. comp. Neurol., 183 (1979) 785-816. 7 Koh, E. T. and Ricardo, J. A., Afferents and efferents of the parabrachial region in the rat: evidence for parallel ascending gustatory versus visceroceptive systems arising from the nucleus of the solitary tract, Anat. Rec., 190 (1978) 449. 8 Nomura, S., Mizuno, N., Itoh, K., Matsuda, K., Sugimoto, T. and Nakamura, Y., Localization of parabrachial nucleus neurons projecting to the thalamus or the amygdala in the cat using horseradish peroxidase, Exp. Neurol., 64 (1979) 375-385. 9 Norgren, R., Gustatory afferents to the ventral forebrain, Brain Research, 81 (1974) 285-295. 10 Norgren, R., Taste pathways to hypothalamus and amygdala, J. comp. Neurol., 156 (1976) 17-30. 11 Norgren, R., Projections of the nucleus of the solitary tract in the rat, Neuroscience, 3 (1978) 207-218. 12 Norgren, R. and Leonard, C. M., Ascending central gustatory pathways, J. comp. NeuroL, 150 (1973) 217-238. 13 Norgren, R. and Wolf, G , Projections of the thalamic gustatory and lingual areas in the rat, Brain Research, 92 (1975) 123-129.
PC PU R si sn so SV to vb
posterior commissure putamen red nucleus substantia innominata substantia nigra superior olivary nucleus sensory trigeminal nucleus optic tract ventrobasal complex
14 Otterson, O. P. and Ben-Ari, Y., Pontine and mesencephalic afferents to the central nucleus of the amygdala of the rat, Neurosci. Lett., 8 (1978) 329-334. 15 Saper, C. B. and Lowey, A. D., Efferent connections of the parabrachial nucleus in the rat, Brain Research, 197 (1980) 291-317. 16 Saper, C. B., Swanson, L. W. and Cowan, W. M., The efferent connections of the ventromedial nucleus of the rat, J. eomp. Neurol., 169 (1976) 409442. 17 Saper, C. B., Swanson, L. W. and Cowan, W. M., An autoradiographic study of the efferent connections of the lateral hypothalamic area in the rat, J. comp. Neurol., 183 (1979) 689-706. 18 Schwartzbaum, J. S., Electrophysiology of taste-mediated functions in the pontine parabrachial nuclei of the behaving rabbit, in preparation. 19 Schwartzbaum, J. S. and DiLorenzo, P. M., Gustatory functions of the nucleus tractus solitarius in the rabbit, Brain Res. Bull., 8 (1982) 285-292. 20 Schwartzbaum, J. S. and Block, C. H., Interrelations between parabrachial pons and ventral forebrain of rabbits in taste-mediated functions. In Y. Ben-Ari (Ed.), The Amygdaloid Complex, Elsevier, Amsterdam, 1981, pp. 367-382. 21 Tsui, H. C. J., The localization of the pontine taste area in the rabbit by multiple unit recording, unpublished Senior Thesis, University of Rochester, 1979. 22 Turner, B. H.,, The cortical sequence and terminal distribution of sensory related afferents to the amygdaloid complex of the rat and monkey. In Y. Ben-Ari (Ed.), The Amygdaloid Complex, Elsevier, Amsterdam, 1981, pp. 51-62. 23 Urban, I. and Richard, P., .4 Stereotaxic atlas of the New Zealand Rabbit's Brain, C. C. Thomas, Springfield, 1972. 24 Voshart, K. and Van der Kooy, D., The organization of the efferent projections of the parabrachial nucleus to the forebrain in the rat: a retrograde fluorescent doublelabeling study, Brain Research, 212 (1981) 271-286. 25 Yamamoto, T. and Kawamura, Y., Cortical responses to electrical and gustatory stimuli in the rabbit, Brain Research, 94 (1975) 447-463. 26 Yamamoto, T., Matsuo, R. and Kawamura, Y., The pontine taste area in the rabbit, Neurosci. Lett., 16 (1980) 5-9.
1
Research Reports
Ascending Efferent Projections of the Gustatory Parabrachial Nuclei in the Rabbit CHRISTINE H. BLOCK and J. S. SCHWARTZBAUM Department of Psychology, University of Rochester, Rochester, N Y 14627 (U.S.A.) (Accepted June 29th, 1982) Key words: rabbit - - parabrachial nuclei - - gustation - - amygdala - - substantia innominata - - thalamus
Neuroanatomical and electrophysiological techniques were used to study efferent projections of the gustatory region of the parabrachial nuclei (PBN) in rabbits. Autoradiographic analysis of [SH]leucinemicroinjected into electrophysiologicallydefined gustatory areas of the PBN revealed strong projections to the ventromedial taste component of the ventrobasal complex of the thalamus and also evidence of projections to the lateral hypothalamus, substantia innominata and nucleus centralis amygdala (ACE). Electrical stimulation in the region of ACE was shown to activate antidromically neurons in the PBN responsive to sapid stimuli, confirming gustatory projections to the ventral forebrain. An electrophysiologicalmapping of units in the amygdala and substantia innominata responsive to electrical stimulation of the gustatory part of the PBN agreed with the autoradiographic data. A proportion of these units were also responsive to gustatory stimuli. However, latencies of evoked response to stimulation of the PBN were unusually long. Outputs from the gustatory PBN to the ventral forebrain may include other modes of polysynaptically integrated information that are largely controlled by interneuronal networks within the PBN. INTRODUCTION The parabrachial nuclei (PBN) in the pontine region of the brain stem participate in the transmission of gustatory information to the forebrain. The PBN in rats, which receive input from the gustatory part of the nucleus of the tractus solitarius (NTS) la, were shown through autoradiographic analysis 10 to project to the ventromedial taste nucleus of the thalamus (VMT) and to a terminal field in the ventral forebrain that included the lateral hypothalamus, substantia innominata and nucleus centralis amygdala (ACE). Further electrophysiological tests (refs. 9, 10) confirmed a gustatory contribution to some of these ventral forebrain projections. Electrical stimulation of both VMT and A C E were shown to activate antidromically neurons in the PBN sensitive to sapid stimuli. In some cases, the same units responded to the two sites of stimulation suggesting closely interlocked bifurcating gustatory projections to the thalamus and ventral forebrain. More recent anatomical studies, using anterograde and retrograde labeling techniques, have verified this divergent pattern of projections from the 0006-8993/83/0000-0000/$03.00 © 1983 Elsevier Biomedical Press
PBNT,8,15,z4,26, but they have also raised questions about the ftmctional nature of the projections to the ventral forebrain. Little evidence was found of double-labeling of PBN cells in rats by retrograde tracers applied to ACE and VMT 24. The projections to these divergent structures also appeared to originate to some degree from different ftmctional parts of the PBN z4. While projections to VMT arose from the medial part of the PBN, which is linked to the gustatory segment of NTS, projections to ACE seemed to originate largely from the more lateral part of the PBN which receives primary input from the visceral segment of NTS 7,11. Thus, the PBN might convey mainly visceral rather than gustatory information to A C E and the ventral forebrain. The present study was undertaken to analyze specifically the gustatory projections of the PBN in the rabbit. Previous work had demonstrated a gustatory zone in the PBN of this species3, 21,e6. A 3-fold approach was followed. First, anterograde labeling was utilized to trace the ascending projections from the electrophysiologically defined gustatory region of the PBN. Second, electrophysiological tests were conducted to determine whether or not gusta-
tory neurons in the PBN could be antidromically activated by stimulation of ACE as seen in the ratg, 1°. Third, an electrophysiological mapping was made of the distribution of cells in the amygdala and substantia innominata that were: (a) responsive to stimulation of the PBN; and (b) responsive to gustatory stimulation. MATERIALS AND METHODS
Subject and general test procedures Twenty-three male, New Zealand rabbits weighing 1.8-2.7 kg were prepared surgically for later electrophysiological and/or neuroanatomical study. The skull was positioned with bregma 3.5 mm above lambda to provide vertical access to the PBN. Details of the surgical procedure are described elsewhere 3. Experimental tests were initiated after at least 1 week post-operative recovery and, for the electrophysiology, were repeated at weekly intervals. The animal was re-anesthetized with a 4.5 ~ mixture of halothane and oxygen administered through a face mask. Following a supplementary injection of sodium pentabarbital ( ~ 20 mg i.v.) to ensure relaxation of jaw and lingual muscles, the trachea was intubated. The animal was then positioned on a heating pad and its head was painlessly immobilized for stereotactic approach to the brain 3. Anesthesia was maintained with a 1.5-2.0 ~ halothane mixture at a flow rate of 0.4-1.0 liter/min. Electrophysiological methods are described in other reportsZ, 19. Taste methodology is also detailed elsewhere 3. Four basic types of taste stimuli were used in addition to a 0.01 M NaCI rinse solution: 0.75 M NaC1, 0.03 M HCI, 0.5 M sucrose, and 0.01 M QHCI. Neuroanatomy An autoradiographic analysis was made of the distribution of L-[4,5-ZH(N)]leucine (spec. act. 56.5 Ci/mmol; New England Nuclear) microinjected into the gustatory region of the PBN in 11 rabbits. Electrophysiological tests were first carried out to locate the gustatory part of the PBN 3. A 1.0/tl Hamilton syringe (23-gauge injection needle) filled with [ZH]leucine and calibrated stereotactically was then positioned in the PBN taste area, A volume of 0.25-0.35 #1 of [3H]leucine, concentrated to
i0/~Ci/#l in 0.9 ~ NaC1, was injected over a 30-min period. The injection needle was left in place for an additional 15 min. All injections were unilateral. The animals were sacrificed 3-6 days following injection 2. They were anesthetized with sodium pentabarbital and perfused intracardially with 0.9 NaCI, followed by l 0 ~ formalin in 0.9 ~ NaCl. The brains were embedded in paraffin (Paraplast) and cut on a microtome at I0 #m thickness. Three consecutive sections were saved at 80/~m intervals and were mounted on glass slides. Autoradiographic procedure was applied to 2 of each 3 sections. The slides were coated with Kodak NTB-2 emulsion and stored in a light-tight box containing a dessicant at 0 °C for 4 weeks. The slides were then processed in Kodak D-19 developer at 14 °C and Ektaflo fixative at 14 °C. They were stained lightly with cresyl violet and analyzed microscopically with dark-field and bright-field condensers.
Antidromic activation of PBN gustatory units A pair of stimulation leads was implanted chronically in ACE in two preparations. The electrodes were constructed from 0.13 mm tungsten wires with the cross-cut tip left tminsulated. The tips were positioned at the same depth, about 1 mm apart along the A-P plane. To aid in the placement of the leads, recording tests were first carried out with flash stimuli to locate the dorsolateral edge of the optic tract at a mid-A-P level of the amygdala. The stimulation leads were then aimed for the dorsal part of the ACE at 1.5 mm lateral and 1.5 mm above the dorsolateral boundary of the optic tract. Following localization of the PBN gustatory area, conventional criteria were used to evaluate antidromic activation of PBN units: (a) relatively short, invariant latency of evoked response; (b) ability of the unit to follow high frequency pulse trains; and (c) susceptibility of the evoked response to collision from spontaneous discharges. These criteria, though not infallible, are generally reliable 5. Symmetrical biphasic pulses (each 0.25 ms duration, about 8.0 V amplitude) to the ACE were delivered through stimulus isolation units (Model PC-1 W-P Instruments) at a frequency of 1/3 s. Pulse trains consisted of 3 pulses delivered at a frequency of 100 Hz. In the collision tests, a spike-amplitude discriminator with a delay circuit was used to control the electrical
stimulation. The interval between a detected spontaneous discharge and the output pulse to the stimulators was set at about half the latency value of the evoked response. The tests for gustatory responsivity have been described previously a.
Ventral forebrain units responsive to stimulation of the P B N Following localization of the PBN gustatory area, concentric stimulation leads were positioned in the gustatory PBN. The leads consisted of a low-impedance etched tungsten wire cemented within a 25gauge thin-wall, stainless-steel tubing that contained a plastic Kapton tubing insert. The tubing was insulated with Epoxylite to within about 1 mm of the base. The tip of the tungsten wire was spaced 3-4 mm below the tubing to minimize damage to the PBN area. Cathodal pulses or pulse trains were applied to the deeper electrode (0.25 ms pulse duration, 150 #A constant current). Polarity of stimulation was sometimes reversed to check on the effective site of stimulation. The stimulation leads were not chronically implanted in the PBN because of damage produced by movement of the brain when the animal was unrestrained between test sessions. The electrodes were repositioned in the PBN at about the same stereotactic coordinates at the start of each test session. In most instances, recordings
made through the deep electrode confirmed the positioning within the gustatory region of the PBN. A systematic exploration was made of the amygdala and substantia innominata in 10 animals to map the distribution of traits responsive to either pulse or pulse-train stimulation of the gustatory PBN. Responsive units were then tested for gustatory sensitivity. In some instances, units unresponsive to PBN stimulation were also checked for gustatory sensitivity. At the conclusion of one or more electrode penetrations during a test session, a marker lesion was placed above the recording sites by passing 20 #A cathodal DC through the electrode for 30 s. Stimulation sites were marked with currents of 50 #A for 30 s. Histological procedures are described elsewhere a. RESULTS
Autoradiographic analyses Fig. 1 illustrates the PBN injection sites for the [3H]leucine in 9 of the 11 preparations; histology was inadequate for 2 of the preparations. These results are superimposed upon a histological reconstruction of recording sites of gustatory units in rabbit PBN derived from earlier work 3. The gustatory region of the PBN in the rabbit encompasses much of the medial nucleus lying below the brachi-
': .J'--
/
]
i
Fig. l. Injection sites of [aH]leucine (indicated by large circled areas) superimposed upon electrophysiological mapping of the gustatory area within the PBNa. The small filled circles indicate the location of gustatory units displaying 'excitatory' responses; the small open circles refer to gustatory units displaying 'inhibitory' responses. Two of the injection sites in the second most rostral section overlapped one another.
E
F
G
Fig. 2. Diagrammatic representation of the ascending distribution of labeled fibers resulting from an injection of [aH]leucinein the medial part of the PBN (A). Stippling indicates labeled fibers in the central tegmental tract (B and C), ventromedial part of the ventrobasal thalamic complex and region dorsal to the substantia nigra (D), lateral hypothalamus (E and F), and substantia innominata and nucleus centralis amygdala (F and G). um conjunctivum and extends into tbe medial part of the lateral nucleus overlying the brachium. Gustatory cells were not found in the lateral half of the lateral nucleus nor in the Kolliker-Fuse nucleus along the lateral edge of the brachium. Seven of the microinjections of amino acid were located in the rostral half of the PBN gustatory area (3 most rostral sections, Fig. 1). They included medial parts of the medial and lateral nuclei and, in two cases, extended into the mesencephalic nucleus of the trigeminal nerve. The ascending projections traced from these relatively rostral PBN injection sites were generally consistent witll one another and had no relation to the involvement of the mesencephalic nucleus. Fig. 2 illustrates diagrammatically the projections traced in one of these preparations. Fibers from rostromedial cells in the PBN (Fig. 2A) congregate in the area of the central tegmental tract, ventrolateral to the central gray (Fig. 2B). Some of the fibers apparently cross the midline shortly after leaving the PBN because there is also a smaller contralateral component 10. At the mesencephalic level of the posterior commissure (Fig. 2C), the pathway begins to diverge. A major contingent of fibers
ascends dorsally to the caudal ventromedial component of the ventrobasal thalamic nucleus (Fig. 2D). A dense concentration of labeled fibers was invariably found within this subnucleus ipsilateral to the injection site. There is a sparser projection to the contralateral thalamic structure. A second contingent of fibers in the mesencephalon courses ventrally above the substantia nigra and subthalamic nucleus (Fig. 2D). These fibers can be traced into the lateral hypothalamus where some appear to terminate (Fig. 2E and F). In 3 of the 7 preparations with mediorostral injections of the PBN, labeled fibers in the lateral hypothalamus could be followed dorsolaterally above the optic tract into the substantia innominata and dorsal part of the nucleus centralis amygdala (Fig. 2F and G). This distribution of fibers to the ventral forebrain was also predominately ipsilateral. Fig. 3 shows some representative photomicrographs of the anatomical material to illustrate pathways and distribution of labeled fibers for a rostromedial injection site in the PBN. In two preparations, the microinjection of amino acid was located in the mediocaudal part of the gustatory PBN (2 most caudal sections, Fig. 1). These
injection sites only labeled cells that projected to the ventromedial component of the ventrobasal thalamus, predominately on the ipsilateral side. In contrast to the mediorostral PBN injections, no fibers could be traced into the ventral forebrain. These results suggest that virtually all parts of the electrophysiologically defined gustatory region of the PBN project heavily to the ventromedial subnucleus of the thalamus involved in gustationl, 4.
Projections to the ventral forebrain, particularly the substantia innominata and amygdala, may have a more delimited origin within the gustatory PBN or may simply involve fewer elements. Antidromic activation of PBN gustatory units Stimulation leads were histologically confirmed in the ACE in one preparation, and in the ACE and substantia innominata in the second preparation.
J
B
C
7,
Fig. 3. Dark-field photomicrographs of labeled fibers resulting from an injection of [aH]leucine in the medial part of the PBN. The apparent label in the area medial to vb (Panel C) is a condensor artifact. Magnification 50 x .
1 NaCl
HCI *
S u c r o se
OHCl
2
41'
3 4"
•
•
•
Fig. 4. Example of antidromic activation of a gustatory neuron in the PBN by electrical stimulation of the nucleus centralis amygdala (ACE). Peristimulus histograms on the left illustrate the response pattern of the unit to the 4 taste stimuli (time bin = 256 msec; arrow indicates onset of taste stimulation for 6.0 s, followed by 10.0 s interval and 10.0 s of 0.01 M NaCI rinse solution). Oscilloscope tracings on the right illustrate: (l) unit response to repeated single pulse stimulation of ACE; (2) collision blockade of evoked unit response; and (3) evoked unit response to each pulse of a high frequency pulse train. Time scale ~ 2 ms for (1) and (2) and 10 ms for (3). Filled circles denote stimulus artifact.
Electrophysiological analysis of the PBN in these animals confirmed the presence of units that: (a) responded antidromically to stimulation of these ventral forebrain structures; and (b) showed gustatory sensitivity. Seven units met these criteria. Fig. 4 illustrates the findings for one of the units. As shown in Panel I, single pulse stimulation in the ACE evoked responses in a small cluster of units including a unit with a relatively large (upward) positive component. The latency of response for this unit
was about 4 ms and was invariant with repeated stimulation. For the 7 units, latencies averaged about 5 ms. In testing for collision, the pulse stimulation was delayed about 2 ms after spontaneous discharge of the positive-going unit. The resultant blockade of the evoked response (Panel 2) satisfied the collision criterion; response of the two small negative-going units persisted. The unit was also capable of following relatively high-frequency (100 Hz) pulse trains (Panel 3). Tests with gustatory
stimuli (Fig. 4 PSTHs) revealed a responsivity of the unit to NaCI, sucrose and HCI. These findings, which confirm observations in the rat a, demonstrate that at least some gustatory neurons in the PBN project directly to the ventral forebrain.
symbol in Fig. 5) showed gustatory sensitivity of which 6 also responded to stimulation of the PBN. The gustatory units, which constituted about 14% of the sample, were all located within the substantia innominata and ACE. For two of the gustatory units, only the pulse trains were effective in evoking responses. The pattern of 'excitatory' or 'inhibitory' response evoked by the gustatory stimuli agreed with the pattern of evoked response to the PBN stimulation. The 0.75 M NaC1 and 0.03 M HCI appeared to be the most effective gustatory stimuli. The latencies of evoked response to stimulation of the PBN were, however, unusually long. As shown in Table I, only about 25 ~ of the units had latencies that were less than 20 ms (units symbolized by open circle in Fig. 5). All of these un.its were located in the substantia innominata and dorsal part of the ACE.
Topography of ventral forebrain units responsive to stimulation of the PBN The results from 10 animals are summarized in Fig. 5. A total of 103 units were isolated in the amygdala and substar~.tia innominata that responded to stimulation of tb.e PBN and/or gustatory stimuli. Most of the responsive units were located in the medial and sublenticular parts of the substantia innominata and in the dorsal half of the ACE. A scattering of responsive units was found in deeper amygdaloid nuclei. Eleven units (indicated by X
CIL)
._.
" -
'
,
.<---~..~ o
,
°
,'
'
-
.
:
~
,
F I 0
. . . .
!
.
.
5
Fig. 5. Distribution of units in the substantia innominata and amygdala that were responsive to electrical stimulation of the PBN and/or gustatory stimuli. X indicates units responsive to gustatory stimuli. Open circles denote units with latency of evoked response to stimulation of the PBN less than 20 ms; filled circles denote units with latency of evoked response greater than 20 ms. The sections, 0.5 mm apart on the A - P plane, are modified from the Atlas of Urban and Richard ~a.
8
TABLE I Distribution of latencies of evoked responses
n = 103. Latency range (ms)
Percentage of total
0-10 11-20 21-30 31-40 41-50 51 q-
5.5 % 17.6~ 14.3 26.3 13.2 23.1
Nevertheless, more than half of the units in these structures had latencies in excess of 20 ms. The same wide range of latencies applied to the units responsive to the gustatory stimuli and stimulation of the PBN. Taken at face value, these findings would suggest relatively few monosyn.aptic projections from the gustatory zone of the PBN to neurons in the ventral forebrain. DISCUSSION The present findings provide evidence of divergent gustatory projections from the PBN to the ventromedial taste nucleus of the thalamus and ventral forebrain, in agreement with anatomical and electrophysiological observations in ratsg,a0, a2. The results argue against the notion of a gustatory-visceral dichotomy in the information conveyed by the PBN, respectively, to the thalamus and ventral forebrain 24. But, by the same token, the data raise some questions about the anatomical and functional equivalence of these projections. The autoradiographic data revealed projections to taste thalamus from all injection sites within the electrophysiologically defined gustatory region of the PBN. The projections to the ventral forebrain, particularly amygdala and substantia innominata, were less consistent across injection sites. This would suggest a less uniform distribution of medially located PBN neurons and/or fewer neurons in this region that project to the ventral forebrain compared to taste thalamus. Retrograde labeling of PBN cells following tracer injections into ACE and taste thalamus in rats z4 in line with other findings s, 14 suggests such a disparity of projections for the medial gustatory area of the PBN.
The topographical distribution of units in the ventral forebrain that were responsive to stimulation of the gustatory region of the PBN agreed with the autoradiographic data. Responsive neurons were concentrated in the substantia innominata and ACE. However, the electrophysiology revealed few apparent monosynaptically driven cells and implicated only a small proportion of the output from medial PBN to ventral forebrain in gustatory functions. Although these findings may reflect in part thalamocortical and diencephalic pathways to the ventral forebrain activated by the PBN stimulation 6, 13,16,17,z2 we suspect that most of the transmission lag arises within PBN and relates to polysynaptic integration of sensory input within PBN. Electrophysiological unit analysis of gustatory PBN in the awake behaving rabbiOS, 20 has revealed evidence of hedonic integration of gustatory input within PBN. This integrated input may be utilized in the control of orolingual motor activity since other types of elements show relationships to fluid input and motor output. Some of these elements may also transmit information to the ventral forebrain. They can be antidromically activated by stimulation of the ventral forebrain 20. The gustatory part of PBN may, in effect, provide the ventral forebrain with a variety of polysynaptically integrated information pertaining to fluid input and orolingual movement and may not simply relay gustatory information. Whether this expanded role also applies to output directed at taste thalamus is tmclear. The transmission of gustatory messages through PBN to taste cortex is rapid. Electrical stimulation of peripheral taste nerves in rabbits evokes responses in gustatory cortex with average latencies of less than 15 ms 25. ACKNOWLEDGEMENT This research was supported by National Institutes of Health Grant N I M H 14594-10. Abbreviations used in figures: ACE bc ci ct fr FR GP
nucleuscentralis arnygdala brachium conjunctivum internal capsule central tegmental area fasciculusretroflexus reticular formation globus pallidus
HYL IC LL MB MesV MG ml nat MV
lateral hypothalamus inferior colliculus lateral lemniscus mammillary bodies mesencephalic trigeminal nucleus magnocellular nucleus, basal forebrain medial lemniscus mammillothalamic tract motor trigeminal nucleus
REFERENCES 1 Benjamin, R. M., Some thalamic and cortical mechanisms of taste. In Y. Zotterman (Ed.), Proceedings of the First International Symposium on Olfaction and Taste, Pergamon Press, Oxford, 1963, pp. 309-330. 2 Cowan, W. M., Gottlieb, D. I., I-Iendrickson, A. E., Price, J. L. and Woolsey, T. A., The autoradiographic demonstration of axonal connections in the central nervous system, Brain Research, 37 (1972) 21-51. 3 DiLorenzo, P. M. and Schwartzbaum, J. S., Coding of gustatory information in the pontine parabrachial nuclei of the rabbit: magnitude of neural response, Brain Research, in press. 4 Emmers, R., Separate relays for tactile, pressure, thermal and gustatory modalities in the cat thalamus, Proc. Soc. exp. BioL Med., 121 (1966) 527-532. 5 Fuller, J. H. and Schlag, J. D.~ Determination of antidromic excitation by the collision test: problems of interpretation, Brain Research, 112 (1976) 283-298. 6 Krieger, M. S., Conrad, M. C. A. and Pfaff, D. W., An autoradiographic study of the efferent connections of the ventromedial nucleus of the hypothalamus, J. comp. Neurol., 183 (1979) 785-816. 7 Koh, E. T. and Ricardo, J. A., Afferents and efferents of the parabrachial region in the rat: evidence for parallel ascending gustatory versus visceroceptive systems arising from the nucleus of the solitary tract, Anat. Rec., 190 (1978) 449. 8 Nomura, S., Mizuno, N., Itoh, K., Matsuda, K., Sugimoto, T. and Nakamura, Y., Localization of parabrachial nucleus neurons projecting to the thalamus or the amygdala in the cat using horseradish peroxidase, Exp. Neurol., 64 (1979) 375-385. 9 Norgren, R., Gustatory afferents to the ventral forebrain, Brain Research, 81 (1974) 285-295. 10 Norgren, R., Taste pathways to hypothalamus and amygdala, J. comp. Neurol., 156 (1976) 17-30. 11 Norgren, R., Projections of the nucleus of the solitary tract in the rat, Neuroscience, 3 (1978) 207-218. 12 Norgren, R. and Leonard, C. M., Ascending central gustatory pathways, J. comp. NeuroL, 150 (1973) 217-238. 13 Norgren, R. and Wolf, G , Projections of the thalamic gustatory and lingual areas in the rat, Brain Research, 92 (1975) 123-129.
PC PU R si sn so SV to vb
posterior commissure putamen red nucleus substantia innominata substantia nigra superior olivary nucleus sensory trigeminal nucleus optic tract ventrobasal complex
14 Otterson, O. P. and Ben-Ari, Y., Pontine and mesencephalic afferents to the central nucleus of the amygdala of the rat, Neurosci. Lett., 8 (1978) 329-334. 15 Saper, C. B. and Lowey, A. D., Efferent connections of the parabrachial nucleus in the rat, Brain Research, 197 (1980) 291-317. 16 Saper, C. B., Swanson, L. W. and Cowan, W. M., The efferent connections of the ventromedial nucleus of the rat, J. eomp. Neurol., 169 (1976) 409442. 17 Saper, C. B., Swanson, L. W. and Cowan, W. M., An autoradiographic study of the efferent connections of the lateral hypothalamic area in the rat, J. comp. Neurol., 183 (1979) 689-706. 18 Schwartzbaum, J. S., Electrophysiology of taste-mediated functions in the pontine parabrachial nuclei of the behaving rabbit, in preparation. 19 Schwartzbaum, J. S. and DiLorenzo, P. M., Gustatory functions of the nucleus tractus solitarius in the rabbit, Brain Res. Bull., 8 (1982) 285-292. 20 Schwartzbaum, J. S. and Block, C. H., Interrelations between parabrachial pons and ventral forebrain of rabbits in taste-mediated functions. In Y. Ben-Ari (Ed.), The Amygdaloid Complex, Elsevier, Amsterdam, 1981, pp. 367-382. 21 Tsui, H. C. J., The localization of the pontine taste area in the rabbit by multiple unit recording, unpublished Senior Thesis, University of Rochester, 1979. 22 Turner, B. H.,, The cortical sequence and terminal distribution of sensory related afferents to the amygdaloid complex of the rat and monkey. In Y. Ben-Ari (Ed.), The Amygdaloid Complex, Elsevier, Amsterdam, 1981, pp. 51-62. 23 Urban, I. and Richard, P., .4 Stereotaxic atlas of the New Zealand Rabbit's Brain, C. C. Thomas, Springfield, 1972. 24 Voshart, K. and Van der Kooy, D., The organization of the efferent projections of the parabrachial nucleus to the forebrain in the rat: a retrograde fluorescent doublelabeling study, Brain Research, 212 (1981) 271-286. 25 Yamamoto, T. and Kawamura, Y., Cortical responses to electrical and gustatory stimuli in the rabbit, Brain Research, 94 (1975) 447-463. 26 Yamamoto, T., Matsuo, R. and Kawamura, Y., The pontine taste area in the rabbit, Neurosci. Lett., 16 (1980) 5-9.