THE JOURNAL OF COMPARATIVE NEUROLOGY 308~42-50 (1991)
Parabrachial Nucleus Projection Towards the Hypothalamic Supraoptic Nucleus: Electrophysiological and Anatomical Observations in the Rat JACK H. JHAMANDAS, KIM H. HARRIS,
AND
TERESA L. KRUKOFF
Departments of Medicine and Neurology (J.H.J., K.H.H.) and Anatomy and Cell Biology (T.L.K.) and the Division of Neuroscience, University of Alberta, Edmonton, Alberta, Canada T6G 2E1
ABSTRACT It has been proposed that the pontine parabrachial nucleus (PBN) participates in the regulation of body fluid balance and the release of vasopressin from the neurohypophysis, although the pathways mediating the latter response are uncertain. This study in the rat, utilizing anatomical and electrophysiological methods, describes a projection from the lateral PBN towards the hypothalamic supraoptic nucleus (SON). Rats received iontophoretic injections of the anterograde tracer Phaseolus vulgaris leucoagglutinin (PHA-L, 2% solution). After 14-17 days, rats were sacrificed and their brains prepared for immunofluorescent visualization of projections to the SON region. PHA-L-labelled terminals were found primarily in perinuclear regions immediately dorsal to the SON. In contrast, injections within the medial PBN and the nearby Kolliker-Fuse nucleus did not reveal labelling in or around the SON, Extracellular recordings from 86 of 118 antidromically identified neurons in anaesthetized rats revealed a set of complex synaptic responses after stimulation in the PBN. Excitatory responses (in 82 of 86 cells) of short ( < 100 msec, 39/82 cells) and long ( > 100 msec, 43/82) duration were observed in both vasopressin- and oxytocin-secreting cells of the SON, while 4/86 cells displayed a depressant response to PBN stimulation. In the adjacent perinuclear zone, 22/39 nonneurosecretory cells responded with an increase in their excitability consequent to an identical stimulus. These data suggest a predominantly facilitatory influence of lateral PBN neurons on SON neurosecretory cells in the rat, and that the PBN-SON projection is an indirect one that utilizes an interneuronal network located in the perinuclear zone adjacent to the SON. Key words: brainstem, Kolliker-Fusenucleus, PHA-L, neurosecretory,vasopressin
The magnocellular neurosecretory neurons within the hypothalamic supraoptic nucleus (SON) release the hormones vasopressin (W) and oxytocin (OXY) from their axon terminals in the neurohypophysis. The amount of hormone released is dependent, in part, upon the unique intrinsic membrane properties of SON neurosecretory cells (Poulin and Wakerley, '82; Bourque, '87), but is also influenced by synaptically generated mechanisms (for review see Renaud, '87). In the rat, the organization of projections to the SON mediating the secretion of VP and OXY has been the focus of several recent anatomical (Sawchenko and Swanson, '82; Oldfield and Silverman, '85; Jhamandas et al., '89; Weiss and Hatton, 1990) and electrophysiological (Day and Renaud, '84; Jhamandas and Renaud, '86; Hatton and Yang, '89) investigations. Of particular interest has been the connectivity of catecholaminecontaining brainstem neurons within the nucleus of the O
1991 WILEY-LISS, INC.
solitary tract (Day and Sibbald, '89; Raby and Renaud, '89) and the caudal ventrolateral medulla (Sawchenko and Swanson, '81, '82; Ciriello and Caverson, '84) to the Vpand OXY-secreting cells in the SON. In contrast, inputs to the hypothalamic SON arising from more rostral levels of the brainstem, including the parabrachial nucleus (PBN) and the locus coeruleus, have been less thoroughly characterized. Anatomical studies that utilize the application of retrograde tracers have identified the PBN as an additional source for the origin of ascending brainstem pathways to the SON (Tribollet et al., '85; Anderson et al., '90). The PBN is a recipient of autonomic-related information (Cechetto and Calaresu, '83; Ward, '89) and is implicated in control of body fluid balance (Ohman and Johnson, '86) as Accepted January 16,1991
PARABRACHIAL PROJECTION TO RAT SUPRAOPTIC NUCLEUS well as a centrally generated pressor response (Ward, '88). More recently, electrical and chemical stimulation within the PBN has been shown to evoke VP release (Sved, '861, although the excitatory nature of such an input has been challenged (Ohman et al., '90).Also unclear is the issue of whether such modulation of VP release is achieved through a direct pathway to the SON or whether an intermediate relay is interposed within such a projection. The electrophysiological and anatomical (anterograde tracer) studies reported here were designed to characterize the PBN projection towards the SON. We first performed electrophysiological recordings in vivo to assess the influence of electrical stimulation within the PBN on identified neurosecretory cells of the SON. We then utilized anterograde transport of Phaseolus vulgaris leucoagglutinin (PHA-L) to establish the light microscopic features of a PBN-SON connection.
MATERIALS AND METHODS Electrophysiological studies Twelve adult, male Sprague-Dawley rats weighing 200350 g were anaesthetized with an intraperitoneal injection of urethane (1.2-1.5 g/kg). The femoral artery and vein were catheterized so as to record blood pressure and administer a peripheral vasoconstrictor (metaraminol), respectively. Heart rate was monitored continuously and body temperature maintained at 37°C. Following tracheal intubation, animals were initially positioned in a sterotaxic device for the implantation of a 30 gauge concentric bipolar stimulation electrode (tip-ring separation less than 0.25 mm; impedance 200-400 kR) in the PBN. Prior to fixation with dental cement, the optimal site for the insertion of the electrode within the PBN was determined by evoking a maximal PBN pressor response consequent to electrical stimulation with a brief high frequency train of cathodal pulses (1 second, 50 Hz). Stimulation current intensities (50-300 pA) were verified by monitoring the voltage drop across a 100 kR resistor in series with the stimulating electrode. A transpharyngeal approach was utilized to gain access to the SON and pituitary. A bipolar electrode was positioned in the neural lobe and connected to an isolated stimulation unit (pulse duration 200 p,s, current intensities up to 800 FA) for antidromic activation of neurosecretory cells. Action potentials from SON neurons were recorded extracellularly by means of glass micropipettes filled with 2.0 M NaCl (impedance 5-10 MR), amplified conventionally, bandpassfiltered (100-10 kHz), displayed on an oscilloscope, and led through a window discriminator to an IBMPS-2 personal computer programmed for online spike train analysis. Neurosecretory cells were identified by criteria for antidromic activation which included constant latency and all-ornone responses, ability to follow paired stimuli (5 msec interval), and evidence of collision cancellation between the spontaneous and antidromic spikes. In the rat, neurosecretory cells can be further distinguished as VP- or OXYsecreting according to their pattern of spontaneous activity and response to activation of peripheral arterial baroreceptors, achieved by a brief intravenous administration of the a-adrenergic agonist metaraminol (2-10 pg), which was sufficient to elevate mean arterial blood pressure by 40-60 mm Hg (Renaud et al., '88). Units demonstrating phasic or continuous activity that was transiently suppressed by baroreceptor activation were classified as VP-secreting; those whose continuous firing was unresponsive to this
43
maneuver were deemed to be OXY-secreting. Intracellular recordings from rat hypothalamic neurosecretory cells in vitro also indicate that most phasically firing neurons display VP-like immunoreactivity (Cobbett et al., '86). In most rats, the anatomical configuration of the anterior and middle cerebral arteries prevented access to the rostra1 portion of the SON, which contains a greater number of OXY-immunoreactive neurons. Most cells examined in this study were therefore located in the posterior and ventral portions of the SON, where a majority of neurosecretory cells are VP-immunoreactive (Sawchenko and Swanson, '82). Patterns of response (excitation or inhibition) consequent to PBN stimulation were derived from peristimulus histograms: a 30% stimulus-induced alteration (increase and decrease) of baseline excitability was required to classify a response as being excitatory or inhibitory. Estimation of latencies was determined first by calculating the average firing frequency (spikes/sec) for a duration of 100 msec prior to the PBN stimulus. The first bin (resolution 1msec) of a sustained response which revealed an alteration of 30% from baseline firing rate was deemed to represent the latency of a particular cell to PBN stimulus. The end of a response was derived in a similar fashion from the first bin following the onset of a response with a firing frequency alteration of 30% from baseline. The duration of responses (in msec) was calculated as a difference between the latency and the end of the response. Data were subject to statistical analysis (chi square test) to determine if the change in firing frequency during the synaptically evoked response (excitatory or inhibitory) was significantly different from the background spontaneous firing rate of a cell 100 msec prior to the onset of stimulus. At the conclusion of each experiment, the location of the stimulating electrode in the PBN was marked by a small anodal lesion (200 pA DC, 15 seconds). Animals were deeply anaesthetized with supplemental sodium pentobarbital (30 mgkg) and perfused transcardially with saline followed by 200 ml of 10% formaldehyde. Stimulation sites were identified in 50 pm serial coronal sections of the brain cut with a vibratome and stained with safranin.
Anatomical studies For anterograde tracer studies with PHA-L, thirteen male Sprague-Dawley rats (200-300 g) were anaesthetized initially with an intraperitoneal injection of sodium pentobarbital (50 mgkg). A glass capillary (25-40 pm tip diameter) containing a 2.5% solution of PHA-L (Vector Labs, Burlingame, CA) in 0.05 M sodium phosphate buffered saline (Na PBS, pH 7.4), was stereotaxically directed towards the region of parabrachial and Kolliker-Fuse nuclei through a small burr hole and incision in the dura. PHA-L was iontophoretically deposited into the brain for 15 minutes by using 7s of pulsed positive current (Gerfen and Sawchencko, '85). Following a 14-17 day survival period, the animals were deeply anaesthetized and perfused transcardially with 100 ml saline followed by 500 ml of 4% paraformaldehyde in Na PBS. The brains were removed and postfixed overnight in fixative and 10% sucrose. Brains were then placed into 30% sucrose until they were sectioned. Frozen 40-50 pm thick sections were collected into 0.02 M potassium phosphate buffered saline (KPBS pH 7.4) and then incubated with gentle agitation in primary antibody (anti-PHA-L; 1:1,000, Vector Labs) in KPBS with 2% normal rabbit serum (Vector Labs) and 0.3%Triton X-100 (TX) for 48 hours a t 4°C. After incubation in the primary antibody, the sections were rinsed in KPBS (2 x 15 min-
J.H. JHAMANDAS ET AL.
44 TABLE 1. Numbers (n) of Supraoptic Neurosecretory and Perinuclear Nonneurosecretory Cells Tested With Electrical Stimulation in the Parabrachid Nucleus Cell type
n
% Excitation
% Inhibition
% No response
Neurosecretory cell VP-secreting OXY-secreting Unclassified Nonneurosecretory cell
118 13 10 35 39
70 73 60 63 57
3 6 0 0 10
27 21 40 37 33
utes) and incubated in secondary antibody (fluorescein anti-goat IgG, Vector, 1:200 in KPBS/0.3% TX) for 45 minutes at room temperature. Following rinses in KPBS/TX (2 x 15 minutes), sections were mounted onto glass microscope slides, air dried, and coverslipped with buffered glycerol (glycero1:PBS = 9:l). Injection sites and location of terminals containing PHA-L were identified with a Zeiss fluorescence microscope (excitation filter, 460-485 Fm) and mapped onto line drawings of the rat brain.
RESULTS Influence of parabrachial nucleus stimulation Data were obtained from 73 VP-and 10 OXY-secreting neurosecretory cells classified on the basis of their firing patterns and response to baroreceptor activation (see Materials and Methods). These cells displayed a range of latencies for their antidromic activation from the neural lobe between 6 and 21 ms. An additional 35 neurosecretory cells in the SON could not be readily categorized into either group because of lack of spontaneous activity or owing to a diminution of spike amplitude during blood pressure manipulation required for baroreceptor activation. Among the SON neurosecretory cell population, stimulation sites confined to the parabrachial and the adjacent Kolliker-Fuse nucleus (Fig. 1) altered the excitability of 73% (86/118) of neurons, while the rest were unresponsive. The data in Table 1 based on peristimulus histograms indicate that a majority (82/86) of both W- and OXYsecreting cells displayed an increase in excitability to ipsilat-
- 8.8
50Hz c (
60 sec
-9.2
C -9.7
Fig. 1. A Symbols superimposed on a series of schematic coronal sections of the rat brain (adapted from Paxinos and Watson, '86) depict the location of the electrode tips in the parabrachial nucleus (PBN) where electrical stimulation evoked a response in the excitability of supraoptic nucleus (SON) neurosecretory cells ( 0 ) .Abbreviations: BC, brachium conjunctivum; KF, Kolliker-Fuse nucleus; LPBN, lateral parabrachial nucleus; MPBN, medial parabrachial nucleus; Me5, mesencephalic trigeminal nucleus; DTN, dorsal tegmental nucleus; Mo5,
motor trigeminal nucleus; s5, sensory root trigeminal nerve; 7n, facial nervehoot; py, pyramidal tract. B: Blood pressure (BPI and heart rate (HR) changes evoked as a consequence of a brief (1second duration) burst of stimuli applied to the PBN at 50 Hz with varying current intensities (PA).The pressor response generated was used as a guide for the optimal placement of stimulation electrodes. C : photomicrograph of an electrode tip positioned within the LPBN. Scale bar = 50gm.
45
PARABRACHIAL PROJECTION TO RAT SUPRAOPTIC NUCLEUS
n
39 800
0
1
cn
lo!
a,
"0
0
43 200
400
600
800
10-
4
a,
Y .-
Q
cn
200
400
600
800
Time (msec) Fig. 2. Peristimulus histograms (200 sweeps, 1 ms resolution) for three different SON neurosecretory cells illustrate different patterns of response to an identical PBN stimulus (100 FA, 200 psec duration). Each histogram also displays an antidromic response following a suprathreshold stimulus in the posterior pituitary (PP). Arrows represent stimulus artifacts. Cell 1 (upper record) displays one of the two most commonly observed responses, consisting of a short duration
excitation ( < 100 msec) following PBN stimulation. The response in cell 2 is of a longer duration ( > 100 msec) and in some cells accompanied the short duration excitation seen in cell 1 (see Fig. 3 ) . Cell 3 displays only a decrease in excitability to PBN stimulus. On the right, number of neurosecretory cells (n) which revealed one of the specific response patterns evoked by electrical stimulation within the PBN.
J.H. JHAMANDAS ET AL.
46
A SON VASOPRESSIN-SECRETING CELL PP
PBN
7
I
200
400
800
600
B NON-NEUROSECRETORY CELL 7
200
400
600
800
Time (msec) Fig. 3. A: Peristimulus histogram from a VP-secreting cell in the SON which displays a combination of short and long duration increases in excitability (also see Fig. 2) following PBN stimulation. B: Peristimulus histogram from a nonneurosecretory cell, located within the adjacent perinuclear zone immediately dorsal to the SON. Note the similarity of the PBN-evoked increase in excitability for this cell to that seen for the VP-secreting cell in A.
era1 PBN stimulation. The mean latency of response for all neurosecretory cells was 47.7 f 0.2 msec. Response patterns to PBN stimulation could be further divided into three basic types. An increase in excitability lasting up to 100 msec was evident in 39 of the neurosecretory cells (Fig. 2). More prolonged excitatory responses ( > 100 msec) were obtained in 43 neurons, while 4 cells displayed a depressant response consequent to PBN stimulation. All changes in firing rate during excitatory and inhibitory responses were
statistically significant (x2 > 6.63, P < 0.01). Although it would appear that PBN stimulation influenced the excitability of both VP and OXY cells, further analysis along these lines is limited by the relatively fewer number of recordings from OXY-secreting neurons (see Materials and Methods). For comparison, 39 nonneurosecretory cells, located within the perinuclear zone immediately dorsal to the SON, were examined for their response to PBN stimulation. Fifty-seven percent (22/39) displayed an increase in excitabil-
PARABRACHIAL PROJECTION TO RAT SUPRAOPTIC NUCLEUS
Fig. 4. Schematic coronal sections of the rat brain (at three levels of the brainstem) illustrating the disposition of the injected tracer PHA-L within the region of the dorsolateral pons. Photomicrographs of injections 1and 2, corresponding to the deposition of PHA-Linto the lateral parabrachial and Kolliker-Fuse nuclei respectively, are shown in more detail in Fig. 5. Abbreviations as in Fig. 1.
47
J.H. JHAMANDAS ET AL.
48
Fig. 5. a: Fluorescence photomicrograph of the site of PHA-L injection in the lateral PBN and Kolliker-Fuse nucleus. b: The PHA-L injection shown in a resulted in a preferential labelling of terminals in an area immediately dorsal to the supraoptic nucleus (SON), with a few fibers visible within the nucleus proper. c: Injection located chiefly
within the Kolliker-Fuse nucleus alone fails to reveal any significant labelling of terminals in or around the SON as shown in d. d Dashed lines demarcate the dorsal boundary of the SON. Abbreviations: BC, brachium conjunctiwm; OT, optic tract. Scale bars in a,c = 150 pm and in b, d = 60 pm.
ity with either short ( < 100 msec, 8 of 22 cells) or long ( > 100 msec, 14 of 22 cells) duration responses (Fig. 3). Four nonneurosecretory cells displayed a depression of their firing frequency consequent to PBN stimulation and the remainder of these cells (13139) revealed no response to similar stimulation. Mean latency following PBN stimulation to these adjacent nonneurosecretory cells was slightly lower (46.4 0.8 msec) than for SON W- and OXYsecreting cells.
anterograde tracer within the lateral subdivision were varied in their rostrocaudal position but the pattern of preferential perinuclear innervation adjacent to the SON remained a consistent finding. However, introduction of PHA-L within more medial portions of the PBN or involving the Kolliker-Fuse nucleus (Fig. 5C) revealed no substantive label either around or within the SON (Fig. 5D). In this latter case, however, labelled fibers originating from the region of Kolliker-Fuse nucleus were indeed visualized in previously identified projections from this area to more caudal levels of the brainstem including the nucleus of the solitary tract (Fulwiler and Saper, '84).
*
Distribution of PHA-L relative to SON Five of twelve animals, which had PHA-L injections centered within the subnuclear compartments of the parabrachial complex (cf. Fulwiler and Saper, '84) and the Kolliker-Fuse nucleus, were chosen for detailed light microscopic analysis (Fig. 4).A typical example of the size of the injection and local spread of the labelled cell bodies is depicted in Fig. 5A and C. Animals receiving an injection of PHA-L located mainly within the lateral PBN (Fig. 5A) displayed a moderately dense innervation of PHA-Lcontaining fibers in the area immediately dorsal and lateral to SON (Fig. 5B). Very few PHA-L-labelled fibers were detected within the SON itself. The PBN injections of the
DISCUSSION The projections from the PBN to the hypothalamus and other forebrain structures provide the PBN with the neural framework to influence the functions of body fluid balance, cardiovascular regulation, and the release of the neurohypophyseal hormone VP (Fulwiler and Saper, '84; Lind and Swanson, '84; Moga et al., '89). The present studies provide two major findings which are important with respect to the latter of these functions. First, electrophysiological data
PARABRACHIAL PROJECTION TO RAT SUPRAOPTIC NUCLEUS
49
Fig. 6. Fluorescence photomicrograph of PHA-L-labelled fibers in perinuclear zone adjacent to the supraoptic nucleus (SON), shown at higher magnification. Injection site was similar to that shown in Fig. 5a. OT, optic tract. Scale bar = 150 pm.
indicate that electrical stimulation of the PBN activates a predominantly excitatory projection to the magnocellular neurosecretory cells of the hypothalamic SON. Second, the presence of PHA-L-labelled terminals within the adjacent perinuclear area rather than in the SON proper, suggests that the pathway from the lateral PBN to the neurosecretory cells is predominantly indirect (i.e., at least disynaptic). Previous studies have revealed conflicting data concerning the influence of the PBN upon VP release, with the observation of a depressant (Ohman et al., '90) or a facilitatory (Sved, '86) effect being reported. However, our data suggest that the neural circuitry between the PBN and the neurosecretory cells of the SON appears more complex than might be envisioned on the basis of hormone release into the systemic circulation. Neurosecretory cells display both short ( < 100 msec) and long ( > 100 msec) duration excitatory responses, the latter perhaps indicative of prolonged transmitter action. Moreover, given the diversity of transmitter agents and peptides within the PBN (Block and Hoffman, '87; Shinohara et al., '88; Chamberlin and Saper, '901, it is not surprising that adjacent SON neurons reveal functionally different responses to the same PBN stimulus, suggesting a heterogeneity of neural connections to Wand OXY-secreting cells. Although the relatively small number of OXY-secreting cells precludes firm conclusions, it does appear that the PBN input to the SON is directed at both populations of peptidergic neurons within the SON, unlike the more selective projections from the A1 region of the caudal ventrolateral medulla (Sawchenko and Swanson, '82; Day and Renaud, '84)and diagonal band of Broca
(Jhamandas and Renaud, '86; Jhamandas et al., '89) that appear to direct preferentially towards the VP-secreting cells. The anatomical data presented here permit a n important clarification of the electrophysiological findings, namely that the PBN projection to the SON is an indirect one. In general, prior light microscopic observations, based upon retrograde transport of tracer molecules injected within the region of the SON do reveal the presence of a projection originating from the PBN (Tribollet et al., '85; Anderson et al., '90). However, the relatively large size of some of the injections and the potential for introduction of the tracer into axons of passage when a dorsal stereotaxic approach is utilized, merits a more cautious interpretation of the earlier data. Anterograde transport of PHA-L in our study indicates that the axons of PBN neurons terminate preferentially within the perinuclear zone around the SON rather than within it. Although the possibility exists that the PHA-L-labelled terminals in the perinuclear zone contact dendrites of SON cells extending into this region, anatomical evidence indicates that, in the rat, most of the SON dendritic tree is ventrally directed towards the glial lamina (Armstrong et al.,'82; Randle et al.,'86). However, the probability of a smaller direct projection from the PBN to SON cells is supported by our data which reveal the presence of few PHA-L-labelled fibers within the SON proper (Fig. 6). A more likely proposition is that the majority of the PBN input is directed towards dorsally located perinuclear zone cells, which have been identified to project to SON neurosecretory cells (Tribollet et al., '85;
J.H. JHAMANDAS ET AL.
50
Jhamandas et al., '89). Under such a scheme, one would expect to find nonneurosecretory cells within the perinuclear zone that might also be activated following PBN stimulation. Indeed, the excitatory responses were recorded from nonneurosecretory cells in virtually identical locations as PHA-L-filled axon terminals of PBN origin. In fact, PBN-evoked responses from these cells occur at latencies shorter than or similar to those observed among SON neurosecretory cells, supporting the notion that local interneurons may participate in the transmission of PBN influences on VP- and OXY-secreting neurons. Another contribution of the anatomical data to the characterization of the PBN-SON projection is the rather selective origin of such a pathway from the lateral but not the medial PBN or the adjacent Kolliker-Fuse nucleus (Fig. 5). PHA-L-labelled terminals within the perinuclear zone around the SON were only observed with injections centered over the lateral PBN. Although labelled terminals originating from the Kolliker-Fuse nucleus were found in regions of the brainstem previously recognized to receive inputs from this nucleus, thereby attesting to the specificity of the injection, no anterograde label was evident in or around the SON. These observations are also consistent with the retrograde light microscopic data mentioned earlier that revealed labelled neurons in the lateral PBN following injections of the SON region. In summary, the electrophysiological and anatomical data reported here indicate the presence of a predominantly facilitatory projection originating within the lateral PBN that is directed towards the neurosecretory cells of the hypothalamic SON and that the effects of this input may be mediated, in large part, through a local interneuronal network adjacent to the SON.
ACKNOWLEDGMENTS This work was supported by the Medical Research Council of Canada and the Alberta Heritage Foundation for Medical Research. We thank Vija Jhamandas, Dianne Vincent, and Greg Morrison for their technical assistance and Cynthia Krys for typing this manuscript.
LITERATURE CITED Anderson, W.A., J.E. Bruni, and A. Kaufmann (1990) Afferent connections of the rat's supraoptic nucleus. Brain Res. Bull. 24:191-200. Armstrong. W.E., J. Scholer, and T.H. McNeill(1982)Immunocytochemical, Golgi and electron microscopic characterization of putative dendrites in the ventral glial lamina of the rat supraoptic nucleus. Neuroscience 7579-694. Block, C.H., and G.E. Hoffman (1987) Neuropeptide and monoamine components of the parabrachial pontine complex. Peptides 8:267-283. Bourque, C.W. (1987) Intrinsic features and control of phasic burst onset in magnocellular neurosecretory cells. In J. Ciriello, F.R. Calaresu, L.P. Renaud, and C. Polosa (eds): Organization of the Autonomic Nervous System: Central and Peripheral Mechanisms. New York Alan R. Liss, Inc., pp. 387-396. Cecbetto, D.F., and F.R. Calaresu (1983) Parabrachial units responding to stimulation of buffer nerves and forebrain in the cat. Am. J. Physiol. 245:R811-R519. Chamberlin, N.L., and C.B. Saper (1989) Neuropeptide immunoreactivity of lateral parabrachial neurons that project to the median preoptic nucleus in the rat. SOC.Neurosci. Abstr. 15265. Ciriello, J., and M.M. Caverson (1984) Direct pathway from neurons in the ventrolateral medulla relaying cardiovascular information to the supraoptic nucleus in the cat. Brain Res. 292t221-228. Cobbett, P., K.G. Smithson, and G.I. Hatton (1986) Immunoreactivity to vasopressin- but not oxytocin-associated neurophysin antiserum in phasic neurons of the rat paraventricular nucleus. Brain Res. 362:7-16. Day, T.A., and L.P. Renaud (1984) Electrophysiological evidence that noradrenergic afferents selectively facilitate the activity of supraoptic vasopressin neurons. Brain Res. 303233-240.
Day, T.A., and J.R. Sibhald (1988) Solitary nucleus excitation of supraoptic vasopressin cells via adrenergic afferents. Am. J. Physiol. 254:R711R716. Fulwiler, C.E., and C.B. Saper (1984) Subnuclear organization of the efferent connections of the parabrachial nucleus in the rat. Brain Res. 7229-259. Gerfen, C.R., and P.E. Sawchenko (1984) An anterograde neuroanatomical tracing method that shows the detailed morphology of neurons, their axons and terminals: Immunohistochemical localization of an axonally transported plant lectin, Phaseolus uulgaris leucoagglutinin (PHA-L) Brain Res. 290:219-238. Hatton, G.I., and Q.Z. Yang (1989) Supraoptic nucleus d e r e n t s from the main olfactory bulb-11. Intracellulary recorded responses to lateral olfactory tract stimulation in rat brain slices. Neuroscience 31989-297. Jhamandas, J.H., and L.P. Renaud (1986) A GABAergic baroreceptoractivated pathway from the diagonal band of Broca to supraoptic vasopressin neurones in the rat. J. Physiol. (Lond.) 381:595-606. Jhamandas, J.H., W. Raby, J. Rogers, R.M. Buijs, and L.P. Renaud (1989) Diagonal band projection towards the hypothalamic supraoptic nucleus: Light and electron microscopic observations in the rat. J. Comp. Neurol. 282:15-23. Lind, R.W., and L.W. Swanson (1984) Evidence for corticotrophin releasing factor and leu-enkephalin in the neural projection from the lateral parabrachial nucleus to the median preoptic nucleus: A retrograde transport, immunohistochemicaldouble labelling study in the rat. Brain R ~ s321:217-224. . Moga, M.M., C.B. Saper, and T.S. Gray (1989) Bed nucleus of the stria terminalis: Cytoarchitecture, immunohistochemistry, and projection to the parabrachial nucleus in the rat. J. Comp. Neurol. 283:313-332. Ohman, L.E., and A.K. Johnson (1986) Lesions in lateral parabrachial nucleus enhance drinking to angiotensin I1 and isoproternol. Am. J. Physiol. 251:R504-R509. Obman, L.E., R.E. Shade, and J.R. Haywood (1990) Parabrachial nucleus modulation of vasopressin release. Am. J. Physiol. 258:R35&R364. Oldfield, B.J., and A.J. Silverman (1985) A light microscopic HRP study of limbic projections to the vasopressin-containing nuclear groups of the hypothalamus. Brain Res. 14:143-157. Paxinos, G. and C. Watson (1986) The rat brain in stereotaxic coordinates, second edition. Academic Press Inc., New York. Poulin, D.A., and J.B. Wakerly (1982) Electrophysiology of hypothalamic magnocellular neurones secreting oxytocin and vasopressin. Neuroscience 7~773-808. Raby, W.N., and L.P. Renaud (1989) Dorsomedial medulla stimulation activates rat supraoptic oxytocin and vasopressin neurons through different pathways. J. Physiol. (Lond.) 41 7279-294. Randle, J.C.R., C.W. Bourque, and L.P. Renaud (1986)Serial reconstruction of lucifer yellow labelled supraoptic nucleus neurons in perfused rat hypothalamic explants. Neuroscience 17:453467. Renaud, L.P. (1987) Magnocellular neuroendocrine neurons: Update on intrinsic properties, synaptic inputs and neuropharmacology. Trends Neurosci. 10:498-502. Renaud, L.P., J.H. Jhamandas, R. Buijs, W. Raby, and J.C.R. Randle (1988) Cardiovascular inputs to hypothalamic neurosecretory neurons. Brain Res. Bull. 20:171-777. Sawchenko, P.E., and L.W. Swanson (1981)Central noradrenergicpathways for the integration of hypothalamic neuroendocrine and autonomic responses. Science 214:685-687. Sawchenko, P.E., and L.W. Swanson (1982) The organization of noradrenergic pathways from the brainstem to the paraventricular and supraoptic nuclei in the rat. Brain Res. Rev. 4,275-325. Shinohara, Y., M. Yamano, T. Matsuzaki, and M. Tohyama (1988) Evidence for the coexistence of substance P, neurotensin and calcitonin generelated peptide in single neurons of the external lateral parabrachial nucleus of the rat. Brain Res. Bull. 20:257-260. Sved, A.F. (1986) Pontine pressor sites which release vasopressin. Brain Res. 369: 143-150. Tribollet, E., W.E. Armstrong, M. Dubois-Dauphin, and J.J. Dreifuss 11985) Extrahypothalamic afferent inputs to the supraoptic nucleus area of the rat as determined by retrograde and anterograde tracing techniques. Neuroscience 15135-138. Ward, D.G. (1988) Stimulation of the parabrachial nuclei with monosodium glutamate increases arterial pressure. Brain Res. 462:383-390. Ward, D.G. (1989) Neurons in the parabrachial nuclei respond to hemorrhage. Brain Res. 491:80-92. Weiss, M.L. and G.I. Hatton (1990) Collateral input to the paraventricular and supraoptic nuclei in rat. I. Afferents from subfornicaland anteroventral third ventricle region. Brain Res. Bull. 24.231-238.
Parabrachial Nucleus Projection Towards the Hypothalamic Supraoptic Nucleus: Electrophysiological and Anatomical Observations in the Rat JACK H. JHAMANDAS, KIM H. HARRIS,
AND
TERESA L. KRUKOFF
Departments of Medicine and Neurology (J.H.J., K.H.H.) and Anatomy and Cell Biology (T.L.K.) and the Division of Neuroscience, University of Alberta, Edmonton, Alberta, Canada T6G 2E1
ABSTRACT It has been proposed that the pontine parabrachial nucleus (PBN) participates in the regulation of body fluid balance and the release of vasopressin from the neurohypophysis, although the pathways mediating the latter response are uncertain. This study in the rat, utilizing anatomical and electrophysiological methods, describes a projection from the lateral PBN towards the hypothalamic supraoptic nucleus (SON). Rats received iontophoretic injections of the anterograde tracer Phaseolus vulgaris leucoagglutinin (PHA-L, 2% solution). After 14-17 days, rats were sacrificed and their brains prepared for immunofluorescent visualization of projections to the SON region. PHA-L-labelled terminals were found primarily in perinuclear regions immediately dorsal to the SON. In contrast, injections within the medial PBN and the nearby Kolliker-Fuse nucleus did not reveal labelling in or around the SON, Extracellular recordings from 86 of 118 antidromically identified neurons in anaesthetized rats revealed a set of complex synaptic responses after stimulation in the PBN. Excitatory responses (in 82 of 86 cells) of short ( < 100 msec, 39/82 cells) and long ( > 100 msec, 43/82) duration were observed in both vasopressin- and oxytocin-secreting cells of the SON, while 4/86 cells displayed a depressant response to PBN stimulation. In the adjacent perinuclear zone, 22/39 nonneurosecretory cells responded with an increase in their excitability consequent to an identical stimulus. These data suggest a predominantly facilitatory influence of lateral PBN neurons on SON neurosecretory cells in the rat, and that the PBN-SON projection is an indirect one that utilizes an interneuronal network located in the perinuclear zone adjacent to the SON. Key words: brainstem, Kolliker-Fusenucleus, PHA-L, neurosecretory,vasopressin
The magnocellular neurosecretory neurons within the hypothalamic supraoptic nucleus (SON) release the hormones vasopressin (W) and oxytocin (OXY) from their axon terminals in the neurohypophysis. The amount of hormone released is dependent, in part, upon the unique intrinsic membrane properties of SON neurosecretory cells (Poulin and Wakerley, '82; Bourque, '87), but is also influenced by synaptically generated mechanisms (for review see Renaud, '87). In the rat, the organization of projections to the SON mediating the secretion of VP and OXY has been the focus of several recent anatomical (Sawchenko and Swanson, '82; Oldfield and Silverman, '85; Jhamandas et al., '89; Weiss and Hatton, 1990) and electrophysiological (Day and Renaud, '84; Jhamandas and Renaud, '86; Hatton and Yang, '89) investigations. Of particular interest has been the connectivity of catecholaminecontaining brainstem neurons within the nucleus of the O
1991 WILEY-LISS, INC.
solitary tract (Day and Sibbald, '89; Raby and Renaud, '89) and the caudal ventrolateral medulla (Sawchenko and Swanson, '81, '82; Ciriello and Caverson, '84) to the Vpand OXY-secreting cells in the SON. In contrast, inputs to the hypothalamic SON arising from more rostral levels of the brainstem, including the parabrachial nucleus (PBN) and the locus coeruleus, have been less thoroughly characterized. Anatomical studies that utilize the application of retrograde tracers have identified the PBN as an additional source for the origin of ascending brainstem pathways to the SON (Tribollet et al., '85; Anderson et al., '90). The PBN is a recipient of autonomic-related information (Cechetto and Calaresu, '83; Ward, '89) and is implicated in control of body fluid balance (Ohman and Johnson, '86) as Accepted January 16,1991
PARABRACHIAL PROJECTION TO RAT SUPRAOPTIC NUCLEUS well as a centrally generated pressor response (Ward, '88). More recently, electrical and chemical stimulation within the PBN has been shown to evoke VP release (Sved, '861, although the excitatory nature of such an input has been challenged (Ohman et al., '90).Also unclear is the issue of whether such modulation of VP release is achieved through a direct pathway to the SON or whether an intermediate relay is interposed within such a projection. The electrophysiological and anatomical (anterograde tracer) studies reported here were designed to characterize the PBN projection towards the SON. We first performed electrophysiological recordings in vivo to assess the influence of electrical stimulation within the PBN on identified neurosecretory cells of the SON. We then utilized anterograde transport of Phaseolus vulgaris leucoagglutinin (PHA-L) to establish the light microscopic features of a PBN-SON connection.
MATERIALS AND METHODS Electrophysiological studies Twelve adult, male Sprague-Dawley rats weighing 200350 g were anaesthetized with an intraperitoneal injection of urethane (1.2-1.5 g/kg). The femoral artery and vein were catheterized so as to record blood pressure and administer a peripheral vasoconstrictor (metaraminol), respectively. Heart rate was monitored continuously and body temperature maintained at 37°C. Following tracheal intubation, animals were initially positioned in a sterotaxic device for the implantation of a 30 gauge concentric bipolar stimulation electrode (tip-ring separation less than 0.25 mm; impedance 200-400 kR) in the PBN. Prior to fixation with dental cement, the optimal site for the insertion of the electrode within the PBN was determined by evoking a maximal PBN pressor response consequent to electrical stimulation with a brief high frequency train of cathodal pulses (1 second, 50 Hz). Stimulation current intensities (50-300 pA) were verified by monitoring the voltage drop across a 100 kR resistor in series with the stimulating electrode. A transpharyngeal approach was utilized to gain access to the SON and pituitary. A bipolar electrode was positioned in the neural lobe and connected to an isolated stimulation unit (pulse duration 200 p,s, current intensities up to 800 FA) for antidromic activation of neurosecretory cells. Action potentials from SON neurons were recorded extracellularly by means of glass micropipettes filled with 2.0 M NaCl (impedance 5-10 MR), amplified conventionally, bandpassfiltered (100-10 kHz), displayed on an oscilloscope, and led through a window discriminator to an IBMPS-2 personal computer programmed for online spike train analysis. Neurosecretory cells were identified by criteria for antidromic activation which included constant latency and all-ornone responses, ability to follow paired stimuli (5 msec interval), and evidence of collision cancellation between the spontaneous and antidromic spikes. In the rat, neurosecretory cells can be further distinguished as VP- or OXYsecreting according to their pattern of spontaneous activity and response to activation of peripheral arterial baroreceptors, achieved by a brief intravenous administration of the a-adrenergic agonist metaraminol (2-10 pg), which was sufficient to elevate mean arterial blood pressure by 40-60 mm Hg (Renaud et al., '88). Units demonstrating phasic or continuous activity that was transiently suppressed by baroreceptor activation were classified as VP-secreting; those whose continuous firing was unresponsive to this
43
maneuver were deemed to be OXY-secreting. Intracellular recordings from rat hypothalamic neurosecretory cells in vitro also indicate that most phasically firing neurons display VP-like immunoreactivity (Cobbett et al., '86). In most rats, the anatomical configuration of the anterior and middle cerebral arteries prevented access to the rostra1 portion of the SON, which contains a greater number of OXY-immunoreactive neurons. Most cells examined in this study were therefore located in the posterior and ventral portions of the SON, where a majority of neurosecretory cells are VP-immunoreactive (Sawchenko and Swanson, '82). Patterns of response (excitation or inhibition) consequent to PBN stimulation were derived from peristimulus histograms: a 30% stimulus-induced alteration (increase and decrease) of baseline excitability was required to classify a response as being excitatory or inhibitory. Estimation of latencies was determined first by calculating the average firing frequency (spikes/sec) for a duration of 100 msec prior to the PBN stimulus. The first bin (resolution 1msec) of a sustained response which revealed an alteration of 30% from baseline firing rate was deemed to represent the latency of a particular cell to PBN stimulus. The end of a response was derived in a similar fashion from the first bin following the onset of a response with a firing frequency alteration of 30% from baseline. The duration of responses (in msec) was calculated as a difference between the latency and the end of the response. Data were subject to statistical analysis (chi square test) to determine if the change in firing frequency during the synaptically evoked response (excitatory or inhibitory) was significantly different from the background spontaneous firing rate of a cell 100 msec prior to the onset of stimulus. At the conclusion of each experiment, the location of the stimulating electrode in the PBN was marked by a small anodal lesion (200 pA DC, 15 seconds). Animals were deeply anaesthetized with supplemental sodium pentobarbital (30 mgkg) and perfused transcardially with saline followed by 200 ml of 10% formaldehyde. Stimulation sites were identified in 50 pm serial coronal sections of the brain cut with a vibratome and stained with safranin.
Anatomical studies For anterograde tracer studies with PHA-L, thirteen male Sprague-Dawley rats (200-300 g) were anaesthetized initially with an intraperitoneal injection of sodium pentobarbital (50 mgkg). A glass capillary (25-40 pm tip diameter) containing a 2.5% solution of PHA-L (Vector Labs, Burlingame, CA) in 0.05 M sodium phosphate buffered saline (Na PBS, pH 7.4), was stereotaxically directed towards the region of parabrachial and Kolliker-Fuse nuclei through a small burr hole and incision in the dura. PHA-L was iontophoretically deposited into the brain for 15 minutes by using 7s of pulsed positive current (Gerfen and Sawchencko, '85). Following a 14-17 day survival period, the animals were deeply anaesthetized and perfused transcardially with 100 ml saline followed by 500 ml of 4% paraformaldehyde in Na PBS. The brains were removed and postfixed overnight in fixative and 10% sucrose. Brains were then placed into 30% sucrose until they were sectioned. Frozen 40-50 pm thick sections were collected into 0.02 M potassium phosphate buffered saline (KPBS pH 7.4) and then incubated with gentle agitation in primary antibody (anti-PHA-L; 1:1,000, Vector Labs) in KPBS with 2% normal rabbit serum (Vector Labs) and 0.3%Triton X-100 (TX) for 48 hours a t 4°C. After incubation in the primary antibody, the sections were rinsed in KPBS (2 x 15 min-
J.H. JHAMANDAS ET AL.
44 TABLE 1. Numbers (n) of Supraoptic Neurosecretory and Perinuclear Nonneurosecretory Cells Tested With Electrical Stimulation in the Parabrachid Nucleus Cell type
n
% Excitation
% Inhibition
% No response
Neurosecretory cell VP-secreting OXY-secreting Unclassified Nonneurosecretory cell
118 13 10 35 39
70 73 60 63 57
3 6 0 0 10
27 21 40 37 33
utes) and incubated in secondary antibody (fluorescein anti-goat IgG, Vector, 1:200 in KPBS/0.3% TX) for 45 minutes at room temperature. Following rinses in KPBS/TX (2 x 15 minutes), sections were mounted onto glass microscope slides, air dried, and coverslipped with buffered glycerol (glycero1:PBS = 9:l). Injection sites and location of terminals containing PHA-L were identified with a Zeiss fluorescence microscope (excitation filter, 460-485 Fm) and mapped onto line drawings of the rat brain.
RESULTS Influence of parabrachial nucleus stimulation Data were obtained from 73 VP-and 10 OXY-secreting neurosecretory cells classified on the basis of their firing patterns and response to baroreceptor activation (see Materials and Methods). These cells displayed a range of latencies for their antidromic activation from the neural lobe between 6 and 21 ms. An additional 35 neurosecretory cells in the SON could not be readily categorized into either group because of lack of spontaneous activity or owing to a diminution of spike amplitude during blood pressure manipulation required for baroreceptor activation. Among the SON neurosecretory cell population, stimulation sites confined to the parabrachial and the adjacent Kolliker-Fuse nucleus (Fig. 1) altered the excitability of 73% (86/118) of neurons, while the rest were unresponsive. The data in Table 1 based on peristimulus histograms indicate that a majority (82/86) of both W- and OXYsecreting cells displayed an increase in excitability to ipsilat-
- 8.8
50Hz c (
60 sec
-9.2
C -9.7
Fig. 1. A Symbols superimposed on a series of schematic coronal sections of the rat brain (adapted from Paxinos and Watson, '86) depict the location of the electrode tips in the parabrachial nucleus (PBN) where electrical stimulation evoked a response in the excitability of supraoptic nucleus (SON) neurosecretory cells ( 0 ) .Abbreviations: BC, brachium conjunctivum; KF, Kolliker-Fuse nucleus; LPBN, lateral parabrachial nucleus; MPBN, medial parabrachial nucleus; Me5, mesencephalic trigeminal nucleus; DTN, dorsal tegmental nucleus; Mo5,
motor trigeminal nucleus; s5, sensory root trigeminal nerve; 7n, facial nervehoot; py, pyramidal tract. B: Blood pressure (BPI and heart rate (HR) changes evoked as a consequence of a brief (1second duration) burst of stimuli applied to the PBN at 50 Hz with varying current intensities (PA).The pressor response generated was used as a guide for the optimal placement of stimulation electrodes. C : photomicrograph of an electrode tip positioned within the LPBN. Scale bar = 50gm.
45
PARABRACHIAL PROJECTION TO RAT SUPRAOPTIC NUCLEUS
n
39 800
0
1
cn
lo!
a,
"0
0
43 200
400
600
800
10-
4
a,
Y .-
Q
cn
200
400
600
800
Time (msec) Fig. 2. Peristimulus histograms (200 sweeps, 1 ms resolution) for three different SON neurosecretory cells illustrate different patterns of response to an identical PBN stimulus (100 FA, 200 psec duration). Each histogram also displays an antidromic response following a suprathreshold stimulus in the posterior pituitary (PP). Arrows represent stimulus artifacts. Cell 1 (upper record) displays one of the two most commonly observed responses, consisting of a short duration
excitation ( < 100 msec) following PBN stimulation. The response in cell 2 is of a longer duration ( > 100 msec) and in some cells accompanied the short duration excitation seen in cell 1 (see Fig. 3 ) . Cell 3 displays only a decrease in excitability to PBN stimulus. On the right, number of neurosecretory cells (n) which revealed one of the specific response patterns evoked by electrical stimulation within the PBN.
J.H. JHAMANDAS ET AL.
46
A SON VASOPRESSIN-SECRETING CELL PP
PBN
7
I
200
400
800
600
B NON-NEUROSECRETORY CELL 7
200
400
600
800
Time (msec) Fig. 3. A: Peristimulus histogram from a VP-secreting cell in the SON which displays a combination of short and long duration increases in excitability (also see Fig. 2) following PBN stimulation. B: Peristimulus histogram from a nonneurosecretory cell, located within the adjacent perinuclear zone immediately dorsal to the SON. Note the similarity of the PBN-evoked increase in excitability for this cell to that seen for the VP-secreting cell in A.
era1 PBN stimulation. The mean latency of response for all neurosecretory cells was 47.7 f 0.2 msec. Response patterns to PBN stimulation could be further divided into three basic types. An increase in excitability lasting up to 100 msec was evident in 39 of the neurosecretory cells (Fig. 2). More prolonged excitatory responses ( > 100 msec) were obtained in 43 neurons, while 4 cells displayed a depressant response consequent to PBN stimulation. All changes in firing rate during excitatory and inhibitory responses were
statistically significant (x2 > 6.63, P < 0.01). Although it would appear that PBN stimulation influenced the excitability of both VP and OXY cells, further analysis along these lines is limited by the relatively fewer number of recordings from OXY-secreting neurons (see Materials and Methods). For comparison, 39 nonneurosecretory cells, located within the perinuclear zone immediately dorsal to the SON, were examined for their response to PBN stimulation. Fifty-seven percent (22/39) displayed an increase in excitabil-
PARABRACHIAL PROJECTION TO RAT SUPRAOPTIC NUCLEUS
Fig. 4. Schematic coronal sections of the rat brain (at three levels of the brainstem) illustrating the disposition of the injected tracer PHA-L within the region of the dorsolateral pons. Photomicrographs of injections 1and 2, corresponding to the deposition of PHA-Linto the lateral parabrachial and Kolliker-Fuse nuclei respectively, are shown in more detail in Fig. 5. Abbreviations as in Fig. 1.
47
J.H. JHAMANDAS ET AL.
48
Fig. 5. a: Fluorescence photomicrograph of the site of PHA-L injection in the lateral PBN and Kolliker-Fuse nucleus. b: The PHA-L injection shown in a resulted in a preferential labelling of terminals in an area immediately dorsal to the supraoptic nucleus (SON), with a few fibers visible within the nucleus proper. c: Injection located chiefly
within the Kolliker-Fuse nucleus alone fails to reveal any significant labelling of terminals in or around the SON as shown in d. d Dashed lines demarcate the dorsal boundary of the SON. Abbreviations: BC, brachium conjunctiwm; OT, optic tract. Scale bars in a,c = 150 pm and in b, d = 60 pm.
ity with either short ( < 100 msec, 8 of 22 cells) or long ( > 100 msec, 14 of 22 cells) duration responses (Fig. 3). Four nonneurosecretory cells displayed a depression of their firing frequency consequent to PBN stimulation and the remainder of these cells (13139) revealed no response to similar stimulation. Mean latency following PBN stimulation to these adjacent nonneurosecretory cells was slightly lower (46.4 0.8 msec) than for SON W- and OXYsecreting cells.
anterograde tracer within the lateral subdivision were varied in their rostrocaudal position but the pattern of preferential perinuclear innervation adjacent to the SON remained a consistent finding. However, introduction of PHA-L within more medial portions of the PBN or involving the Kolliker-Fuse nucleus (Fig. 5C) revealed no substantive label either around or within the SON (Fig. 5D). In this latter case, however, labelled fibers originating from the region of Kolliker-Fuse nucleus were indeed visualized in previously identified projections from this area to more caudal levels of the brainstem including the nucleus of the solitary tract (Fulwiler and Saper, '84).
*
Distribution of PHA-L relative to SON Five of twelve animals, which had PHA-L injections centered within the subnuclear compartments of the parabrachial complex (cf. Fulwiler and Saper, '84) and the Kolliker-Fuse nucleus, were chosen for detailed light microscopic analysis (Fig. 4).A typical example of the size of the injection and local spread of the labelled cell bodies is depicted in Fig. 5A and C. Animals receiving an injection of PHA-L located mainly within the lateral PBN (Fig. 5A) displayed a moderately dense innervation of PHA-Lcontaining fibers in the area immediately dorsal and lateral to SON (Fig. 5B). Very few PHA-L-labelled fibers were detected within the SON itself. The PBN injections of the
DISCUSSION The projections from the PBN to the hypothalamus and other forebrain structures provide the PBN with the neural framework to influence the functions of body fluid balance, cardiovascular regulation, and the release of the neurohypophyseal hormone VP (Fulwiler and Saper, '84; Lind and Swanson, '84; Moga et al., '89). The present studies provide two major findings which are important with respect to the latter of these functions. First, electrophysiological data
PARABRACHIAL PROJECTION TO RAT SUPRAOPTIC NUCLEUS
49
Fig. 6. Fluorescence photomicrograph of PHA-L-labelled fibers in perinuclear zone adjacent to the supraoptic nucleus (SON), shown at higher magnification. Injection site was similar to that shown in Fig. 5a. OT, optic tract. Scale bar = 150 pm.
indicate that electrical stimulation of the PBN activates a predominantly excitatory projection to the magnocellular neurosecretory cells of the hypothalamic SON. Second, the presence of PHA-L-labelled terminals within the adjacent perinuclear area rather than in the SON proper, suggests that the pathway from the lateral PBN to the neurosecretory cells is predominantly indirect (i.e., at least disynaptic). Previous studies have revealed conflicting data concerning the influence of the PBN upon VP release, with the observation of a depressant (Ohman et al., '90) or a facilitatory (Sved, '86) effect being reported. However, our data suggest that the neural circuitry between the PBN and the neurosecretory cells of the SON appears more complex than might be envisioned on the basis of hormone release into the systemic circulation. Neurosecretory cells display both short ( < 100 msec) and long ( > 100 msec) duration excitatory responses, the latter perhaps indicative of prolonged transmitter action. Moreover, given the diversity of transmitter agents and peptides within the PBN (Block and Hoffman, '87; Shinohara et al., '88; Chamberlin and Saper, '901, it is not surprising that adjacent SON neurons reveal functionally different responses to the same PBN stimulus, suggesting a heterogeneity of neural connections to Wand OXY-secreting cells. Although the relatively small number of OXY-secreting cells precludes firm conclusions, it does appear that the PBN input to the SON is directed at both populations of peptidergic neurons within the SON, unlike the more selective projections from the A1 region of the caudal ventrolateral medulla (Sawchenko and Swanson, '82; Day and Renaud, '84)and diagonal band of Broca
(Jhamandas and Renaud, '86; Jhamandas et al., '89) that appear to direct preferentially towards the VP-secreting cells. The anatomical data presented here permit a n important clarification of the electrophysiological findings, namely that the PBN projection to the SON is an indirect one. In general, prior light microscopic observations, based upon retrograde transport of tracer molecules injected within the region of the SON do reveal the presence of a projection originating from the PBN (Tribollet et al., '85; Anderson et al., '90). However, the relatively large size of some of the injections and the potential for introduction of the tracer into axons of passage when a dorsal stereotaxic approach is utilized, merits a more cautious interpretation of the earlier data. Anterograde transport of PHA-L in our study indicates that the axons of PBN neurons terminate preferentially within the perinuclear zone around the SON rather than within it. Although the possibility exists that the PHA-L-labelled terminals in the perinuclear zone contact dendrites of SON cells extending into this region, anatomical evidence indicates that, in the rat, most of the SON dendritic tree is ventrally directed towards the glial lamina (Armstrong et al.,'82; Randle et al.,'86). However, the probability of a smaller direct projection from the PBN to SON cells is supported by our data which reveal the presence of few PHA-L-labelled fibers within the SON proper (Fig. 6). A more likely proposition is that the majority of the PBN input is directed towards dorsally located perinuclear zone cells, which have been identified to project to SON neurosecretory cells (Tribollet et al., '85;
J.H. JHAMANDAS ET AL.
50
Jhamandas et al., '89). Under such a scheme, one would expect to find nonneurosecretory cells within the perinuclear zone that might also be activated following PBN stimulation. Indeed, the excitatory responses were recorded from nonneurosecretory cells in virtually identical locations as PHA-L-filled axon terminals of PBN origin. In fact, PBN-evoked responses from these cells occur at latencies shorter than or similar to those observed among SON neurosecretory cells, supporting the notion that local interneurons may participate in the transmission of PBN influences on VP- and OXY-secreting neurons. Another contribution of the anatomical data to the characterization of the PBN-SON projection is the rather selective origin of such a pathway from the lateral but not the medial PBN or the adjacent Kolliker-Fuse nucleus (Fig. 5). PHA-L-labelled terminals within the perinuclear zone around the SON were only observed with injections centered over the lateral PBN. Although labelled terminals originating from the Kolliker-Fuse nucleus were found in regions of the brainstem previously recognized to receive inputs from this nucleus, thereby attesting to the specificity of the injection, no anterograde label was evident in or around the SON. These observations are also consistent with the retrograde light microscopic data mentioned earlier that revealed labelled neurons in the lateral PBN following injections of the SON region. In summary, the electrophysiological and anatomical data reported here indicate the presence of a predominantly facilitatory projection originating within the lateral PBN that is directed towards the neurosecretory cells of the hypothalamic SON and that the effects of this input may be mediated, in large part, through a local interneuronal network adjacent to the SON.
ACKNOWLEDGMENTS This work was supported by the Medical Research Council of Canada and the Alberta Heritage Foundation for Medical Research. We thank Vija Jhamandas, Dianne Vincent, and Greg Morrison for their technical assistance and Cynthia Krys for typing this manuscript.
LITERATURE CITED Anderson, W.A., J.E. Bruni, and A. Kaufmann (1990) Afferent connections of the rat's supraoptic nucleus. Brain Res. Bull. 24:191-200. Armstrong. W.E., J. Scholer, and T.H. McNeill(1982)Immunocytochemical, Golgi and electron microscopic characterization of putative dendrites in the ventral glial lamina of the rat supraoptic nucleus. Neuroscience 7579-694. Block, C.H., and G.E. Hoffman (1987) Neuropeptide and monoamine components of the parabrachial pontine complex. Peptides 8:267-283. Bourque, C.W. (1987) Intrinsic features and control of phasic burst onset in magnocellular neurosecretory cells. In J. Ciriello, F.R. Calaresu, L.P. Renaud, and C. Polosa (eds): Organization of the Autonomic Nervous System: Central and Peripheral Mechanisms. New York Alan R. Liss, Inc., pp. 387-396. Cecbetto, D.F., and F.R. Calaresu (1983) Parabrachial units responding to stimulation of buffer nerves and forebrain in the cat. Am. J. Physiol. 245:R811-R519. Chamberlin, N.L., and C.B. Saper (1989) Neuropeptide immunoreactivity of lateral parabrachial neurons that project to the median preoptic nucleus in the rat. SOC.Neurosci. Abstr. 15265. Ciriello, J., and M.M. Caverson (1984) Direct pathway from neurons in the ventrolateral medulla relaying cardiovascular information to the supraoptic nucleus in the cat. Brain Res. 292t221-228. Cobbett, P., K.G. Smithson, and G.I. Hatton (1986) Immunoreactivity to vasopressin- but not oxytocin-associated neurophysin antiserum in phasic neurons of the rat paraventricular nucleus. Brain Res. 362:7-16. Day, T.A., and L.P. Renaud (1984) Electrophysiological evidence that noradrenergic afferents selectively facilitate the activity of supraoptic vasopressin neurons. Brain Res. 303233-240.
Day, T.A., and J.R. Sibhald (1988) Solitary nucleus excitation of supraoptic vasopressin cells via adrenergic afferents. Am. J. Physiol. 254:R711R716. Fulwiler, C.E., and C.B. Saper (1984) Subnuclear organization of the efferent connections of the parabrachial nucleus in the rat. Brain Res. 7229-259. Gerfen, C.R., and P.E. Sawchenko (1984) An anterograde neuroanatomical tracing method that shows the detailed morphology of neurons, their axons and terminals: Immunohistochemical localization of an axonally transported plant lectin, Phaseolus uulgaris leucoagglutinin (PHA-L) Brain Res. 290:219-238. Hatton, G.I., and Q.Z. Yang (1989) Supraoptic nucleus d e r e n t s from the main olfactory bulb-11. Intracellulary recorded responses to lateral olfactory tract stimulation in rat brain slices. Neuroscience 31989-297. Jhamandas, J.H., and L.P. Renaud (1986) A GABAergic baroreceptoractivated pathway from the diagonal band of Broca to supraoptic vasopressin neurones in the rat. J. Physiol. (Lond.) 381:595-606. Jhamandas, J.H., W. Raby, J. Rogers, R.M. Buijs, and L.P. Renaud (1989) Diagonal band projection towards the hypothalamic supraoptic nucleus: Light and electron microscopic observations in the rat. J. Comp. Neurol. 282:15-23. Lind, R.W., and L.W. Swanson (1984) Evidence for corticotrophin releasing factor and leu-enkephalin in the neural projection from the lateral parabrachial nucleus to the median preoptic nucleus: A retrograde transport, immunohistochemicaldouble labelling study in the rat. Brain R ~ s321:217-224. . Moga, M.M., C.B. Saper, and T.S. Gray (1989) Bed nucleus of the stria terminalis: Cytoarchitecture, immunohistochemistry, and projection to the parabrachial nucleus in the rat. J. Comp. Neurol. 283:313-332. Ohman, L.E., and A.K. Johnson (1986) Lesions in lateral parabrachial nucleus enhance drinking to angiotensin I1 and isoproternol. Am. J. Physiol. 251:R504-R509. Obman, L.E., R.E. Shade, and J.R. Haywood (1990) Parabrachial nucleus modulation of vasopressin release. Am. J. Physiol. 258:R35&R364. Oldfield, B.J., and A.J. Silverman (1985) A light microscopic HRP study of limbic projections to the vasopressin-containing nuclear groups of the hypothalamus. Brain Res. 14:143-157. Paxinos, G. and C. Watson (1986) The rat brain in stereotaxic coordinates, second edition. Academic Press Inc., New York. Poulin, D.A., and J.B. Wakerly (1982) Electrophysiology of hypothalamic magnocellular neurones secreting oxytocin and vasopressin. Neuroscience 7~773-808. Raby, W.N., and L.P. Renaud (1989) Dorsomedial medulla stimulation activates rat supraoptic oxytocin and vasopressin neurons through different pathways. J. Physiol. (Lond.) 41 7279-294. Randle, J.C.R., C.W. Bourque, and L.P. Renaud (1986)Serial reconstruction of lucifer yellow labelled supraoptic nucleus neurons in perfused rat hypothalamic explants. Neuroscience 17:453467. Renaud, L.P. (1987) Magnocellular neuroendocrine neurons: Update on intrinsic properties, synaptic inputs and neuropharmacology. Trends Neurosci. 10:498-502. Renaud, L.P., J.H. Jhamandas, R. Buijs, W. Raby, and J.C.R. Randle (1988) Cardiovascular inputs to hypothalamic neurosecretory neurons. Brain Res. Bull. 20:171-777. Sawchenko, P.E., and L.W. Swanson (1981)Central noradrenergicpathways for the integration of hypothalamic neuroendocrine and autonomic responses. Science 214:685-687. Sawchenko, P.E., and L.W. Swanson (1982) The organization of noradrenergic pathways from the brainstem to the paraventricular and supraoptic nuclei in the rat. Brain Res. Rev. 4,275-325. Shinohara, Y., M. Yamano, T. Matsuzaki, and M. Tohyama (1988) Evidence for the coexistence of substance P, neurotensin and calcitonin generelated peptide in single neurons of the external lateral parabrachial nucleus of the rat. Brain Res. Bull. 20:257-260. Sved, A.F. (1986) Pontine pressor sites which release vasopressin. Brain Res. 369: 143-150. Tribollet, E., W.E. Armstrong, M. Dubois-Dauphin, and J.J. Dreifuss 11985) Extrahypothalamic afferent inputs to the supraoptic nucleus area of the rat as determined by retrograde and anterograde tracing techniques. Neuroscience 15135-138. Ward, D.G. (1988) Stimulation of the parabrachial nuclei with monosodium glutamate increases arterial pressure. Brain Res. 462:383-390. Ward, D.G. (1989) Neurons in the parabrachial nuclei respond to hemorrhage. Brain Res. 491:80-92. Weiss, M.L. and G.I. Hatton (1990) Collateral input to the paraventricular and supraoptic nuclei in rat. I. Afferents from subfornicaland anteroventral third ventricle region. Brain Res. Bull. 24.231-238.
