Lateral hypothalamic region excites the activity of vasopressin neurons in the supraoptic nucleus through subfornical organ neurons

EXPERIMENTAL

NEUROLOGY

97,2 12-2 18 ( 1987)

RESEARCH

NOTE

Lateral Hypothalamic Region Excites the Activity of Vasopressin Neurons in the Supraoptic Nucleus through Subfornical Organ Neurons J. TANAKA, H. SAITO, H. KABA, K. NOJIMA,

AND K. SETO’

Department OfphysioIogy, Kochi Medical School, Nankoku, Kochi 781-51, Japan Received October 27, 1986; revision received December 30, I986 In urethane-anesthetized male rats, microinjection of angiotensin II into the lateral hypothalamic area excited the activity of about half (N = 7) of subfomical organ neurons (N = 15) antidromically identified as projecting to the hypothalamic supraoptic nucleus. Microinjection of angiotensin II also excited the activity of approximately one-quarter (N = 8) of putative vasopressin-secreting neurons (N = 28) in the hypothalamic supraoptic nucleus and these excitatory responses ofputative vasopressin-secreting neurons were blocked (N = 3) or attenuated (N = 3) by pretreatment with the angiotensin II antagonist saralasin, but not by isotonic saline (N = 2), in the subfomical organ. Q 1987 Academic press, Inc.

The subfomical organ (SFO), one of the several circumventricular organs of the brain, participates in the control of vasopressin (VP) release from the posterior pituitary (2-4,6). Neurons in the SF0 project directly to both the hypothalamic supraoptic (SON) and paraventricular nuclei, regions known to contain the cell bodies of VP- and oxytocin-secreting neurons and to regulate the circulating amount of these hormones (7- 10). Recent immunohistochemical tracing studies showed that angiotensin II (AII)-immunoreactive cells exist in the lateral hypothalamic area (LHA) and these neurons have Abbreviations: LILA-lateral hypothalamic area, SFO-subfomical organ, SON-supraog tic nucleus, AII-angiotensin II, Sar-saralasin, VP-vasopressin. ’ This work was supported in part by grant 59770 126 from the Ministry of Education, Science and Culture, Japan. Please address correspondence to J. Tanaka. 212 0014-4886187 $3.00 Copyright0 1987 by Academic All rights of reproduction

Press, Inc. in any form reserved.

LHA

EXCITES

SON

NEURONS

VIA

SF0

NEURONS

213

AII-immunoreactive efferent fibers projecting to the SF0 (7,8). With respect to the pathways from the LHA to the SFO, we reported that the majority of LHA neurons projecting to the SF0 (14) as well as SF0 neurons projecting to the SON (5) have AI1 receptors and that LHA efferent fibers excite the activity of SF0 neurons projecting to the SF0 via AI1 receptors (5). However, it remains to be determined whether or not the LHA neurons projecting to the SF0 actually alter the excitability of SF0 neurons through SF0 neurons projecting to it. In the present study, we examined the effects of microinjection of AI1 into the LHA on the activity of SF0 neurons antidromically identified as projecting to the SF0 or putative VP-secreting neurons in the SON and the effects of pretreatment of the AI1 antagonist into the SF0 on responses of putative VP-secreting neurons in the SON to AI1 injection into the LHA. Male Wistar rats weighing 270 to 330 g were anesthetized with urethane (0.9 to 1.3 g/kg, i.p.) and placed in a stereotaxic frame in a prone position. Coaxial bipolar stainless-steel electrodes (26-gauge, tip-ring separation 0.5 mm, resistance 50 to 100 kQ) were used to stimulate the SF0 or posterior pituitary with cathodic monophasic pulses (duration 0.2 ms). Glass micropipets filled with 0.5 M sodium acetate solution containing 2% pontamine sky blue 6B were used to make extracellular single-unit recordings from the SF0 or SON. Signals were amplified conventionally, displayed on an oscilloscope, and led to the signal processor (Nihon Koden, ATAC-450) programmed for spike train analysis. Stainless-steel cannulas (30-gauge; outside diameter 0.3 mm; length 20 mm), polyethylene tubes (length 80 mm) with which an injection volume can be calculated from the moving distance of an air bubble and the inside diameter (0.18 mm), and microinjectors (N. Takahashi Co., B341) were used for microinjection of drug solution or vehicle. AI1 (Asp’Ile’-AII) salt (Sigma) and saralasin (Sar) ($a?-Va15-Ala’-AII) salt (Peptide Institute), a specific AI1 antagonist, were dissolved in isotonic saline vehicle, and drawn into the cannulas and connecting tubes (including a bubble). AI1 and Sar solutions were injected in doses of lo-l2 and 10-l’ M, respectively. Each microinjection was administered in a volume of 0.2 ~1 during 30 s. In the experiment combining microinjection of AI1 with recording from the SFO, a coaxial bipolar electrode was placed in the SON to stimulate the axons of SF0 neurons antidromically. The criteria for antidromic responses are shown in Fig. 1D. As we had antidromically identified the LHA neurons projecting to the SF0 (14), a stainless-steel cannula filled with drug solution or vehicle was stereotaxically directed toward the LHA sites where these neurons were located. In experiments combining injection of AI1 into the LHA with SON recordings, one stimulating electrode was placed in the posterior pituitary to identify SON neurohypophyseal neurons by antidromic activa-

214

TANAKA

ET AL.

tion. Stainless-steel cannulas were then positioned in the LHA sites described above and SF0 sites where neurons projecting to the SON were located (Fig. 1E), respectively. At termination of each experiment, the stimulation and recording sites were marked by depositing a small amount of iron and dye, respectively. The tip positions of the microinjection sites were marked by electrolytic lesion. The animals were then perfused with 10% Formalin containing potassium ferricyanide and ferrocyanide. The marking sites were confirmed histologically in 50-pm sections stained with neutral red. The stereotaxic coordinates for the marking sites were determined according to the atlas of Paxinos and Watson ( 11). Nineteen neurons in the SF0 were antidromically activated by electrical stimulation of the SON (Fig. ID). The mean latency and threshold of antidromic responses were 14.2 + 2.6 (SD) ms and 0.71 + 0.14 (SD) mA, respectively. These identified SF0 neurons were tested with microinjection of AI1 (N = 15) and isotonic saline (N = 4) into the LHA. Microinjected AI1 excited the activity of 7 identified SF0 neurons (Fig. IA) (response period 45 to 75 s), but did not affect the remaining neurons (N = 8) (Fig. 1B). Microinjected isotonic saline into the LHA did not affect the activity of identified SF0 neurons (N = 4) tested (Fig. 1C). Cells in the SF0 that were responsive to microinjected AI1 into the LHA were located peripherally, and unresponsive neurons were located in the center (Fig. 1E). These results suggest that AIIsensitive LHA neurons projecting to the SF0 have an excitatory influence on the excitability of neurons projecting to the SON in the periphery of the SFO, and are consistent with those of our recent study (5). Neurohypophyseal neurons of the rat SON display two major patterns of spontaneous activity: phasic or continuous firing (12). The phasic activity (Fig. 2) is thought to be a characteristic of VP-secreting neurons (12). Thus, antidromically identified phasically firing SON neurons (N = 32; mean + SD latency, 13.2 + 2.5 ms; mean + SD intraburst frequency, 7.8 + 4.4 Hz) were tested with microinjection of AI1 (N = 28) and isotonic saline (N = 4) into the LHA. Microinjected AI1 excited 8 identified SON neurons (Fig. 2A-C), but did not affect the remainder (N = 20) (Fig. 2D). These excitatory responses of identified SON neurons to microinjected AI1 occurred immediately after the drug was injected into the LHA, and continued for a long period (range 45 to 85 s). As a control, microinjected isotonic saline into the LHA did not affect the activity of 4 identified SON neurons (Fig. 2E). To determine if the responses resulting from microinjected AI1 into the LHA were mediated through SF0 neurons, Sar was injected into the SF0 immediately before AI1 injection into the LHA. Pretreatment with Sar into the SF0 blocked (N = 3) (Fig. 2A) or attenuated (N = 3) (Fig. 2B) the excit-

LHA EXCITES

SON NEURONS

VIA SF0 NEURONS

215

vhc

RG. 1. Effect of microinjection of angiotensin II (AII), or saline into the lateral hypothalamic area (LHA) on the activity of subfomical organ (SFO) neurons antidromically identified as projecting to the hypothalamic supraoptic nucleus (SON). Microinjected AI1 into the LHA produced either an excitation (A) or no effect (B) in firing of identified (SFO) neurons whereas microinjected saline had no effect (C). Superimposed oscilloscope records from a SF0 units in D illustrate the features of antidromic activation: constant latency (top trace); constant latency responses following two SON stimuli presented with an interstimulus interval less than 10 ms (second trace); collision cancellation of antidromic action potentials by spontaneous (*) action potentials (lower trace). Closed and open circles and triangles in E indicate the loci of responsive and unresponsive neurons to microinjected AI1 into the LHA and neurons tested with saline, respectively. Abbreviations: vhc, ventral hippocampal commissure, 3V, third ventricle. Bar = 0.1 mm.

atory responses of identified SON neurons to microinjected AI1 into the LHA, whereas pretreatment with isotonic saline did not affect them (N = 2) (Fig. 2C). With respect to the effect of Sar on the firing activity of SON neurosecretory neurons, Akaishi et al. (1) demonstrated that administration of Sar (1 to 10 pg) into the third ventricle inhibits the activity of SON neurosecretory neurons and that the inhibition occurred within 2 to 3 min after the injection and continued for a long time (more than 7 mm). In the present study, the action of Sar could be observed at 30 s alter the beginning of injection and the firing activity, which seems to be spontaneous activity, was ob-

216

TANAKA

D

LHA-AI

ET AL.

E

LHA-S&ns

FIG. 2. Responses of putative vasopressin (VP)-secreting neurons in the SON to microinjection of angiotensin II (AU) or saline into the LHA and the effect of pretreatment with saraIasin (Sar) into the SF0 on the responses to AI1 injection into the LHA. Microinjected AI1 into the LHA produced either an excitation (A-C) or no effect (D) in firing of putative VP-secreting neurons whereas microinjected saline had no effect(E). The excitatory responses to AI1 injection into the LHA were blocked (A) or attenuated (B) by pretreatment with Sar into the SFO, but not by pretreatment with saline(C).

served within 45 to 135 s after the end time of injection in all identified SON neurons tested (Fig. IA, B), supporting the possibility that the action of Sar observed in this study may not be produced by a leakage of microinjected Sar into the third ventricle. Thus, these results suggest that the excitatory responses of putative VP-secreting neurons in the SON to microinjected AI1 into the LHA may be mediated by AII-sensitive SF0 neurons. Electrical or chemical (AII) stimulation of the SF0 predominantly excites the excitability of putative VP-secreting neurons in the SON (13) and increases plasma VP concentrations (2-4, 6) in the rat. In addition, we have observed that electrical stimulation of the LHA excites the excitability of SF0 neurons with efferent projections to the SON via AI1 receptors (5).

LHA EXCITES

SON NEURONS

VIA SF0 NEURONS

217

Therefore, from our data we propose that activation of AII-sensitive LHA neurons projecting to the SF0 (AII-immunoreactive pathways) may increase the excitability of putative VP-secreting neurons in the SON through an excitatory influence on AI&sensitive SF0 neurons with efferent projections to it. However, we cannot rule out the possibility that AI&sensitive SF0 neurons projecting to other brain sites that in turn supply afferent fibers to the SON may be involved in such an action of the LHA on putative VP-secreting neurons in the SON because activation of SF0 neurons elicits short-latency brief or long-latency prolonged excitation in the activity of putative VP-secreting neurons in the SON (13). Anatomical observations that the LHA receives AII-immunoreactive efferent fibers from the nucleus tractus solitarius where an influence of baroreceptor aIferent fibers is reflected (8) plus our results suggest that the pathways from the LHA to the SON through SF0 neurons may be related to the control of VP release by the baroreceptor reflex. Although further work is necessary to establish the hypothesis and the findings in which these pathways cannot influence on a larger proportion of putative VP-secreting neurons in the SON may raise a problem in the hypothesis itself, our data indicate that these pathways, at least, may participate in regulating VP secretion. In summary, our electrophysiologic data reveal an interconnection between three brain regions and offer the hypothesis that AII-sensitive LHA neurons projecting to the SF0 may act to enhance the excitability of putative VP-secreting netuons in the SON through a facilitatory influence on AIIsensitive SF0 neurons. REFERENCES 1. AKAISHI, T., H. NEGORO, AND S. KOBAYASHI. 1980. Responses of paraventricular and supraoptic units to angiotensin II, Sar’-Iler-angiotensin II and hypertonic NaCI administered into the cerebral ventricle. BruinRex 188: 499-5 11. 2. FERGUSON, A. V., AND N. M. KASTING. 1986. Electrical stimulation in subfomicaI organ increases plasma vasopressin concentrations in the conscious rat. Am. J. Physiol.251: R425-R428. 3. IOVINO, M., AND L. STEARDO. 1984. Vasopressin release to central and peripheral angiotensin II in rats with lesions of the subfomical organ. BrainRex322:365-368. 4. IOVINO, M., AND L. STEARDO. 1985. Thirst and vasopressin secretion following central administration of angiotensin II in rats with lesions of the septaI area and subfomical organ. Neuroscience 15: 6 l-67. 5. KABA, H., J. TANAKA, H. SAITO, AND K. SETO. 1986. Action of the lateral hypothalamic area on subfomical organ neurons projecting to the supraoptic nucleus in the rat. Exp. Neurol.94:431-435. 6. KNEPEL, W., D. NUTTO, AND D. K. MEYER. 1982. Effect oftranscction ofsubfomical organ efferent projections on vasoprcssin release induced by angiotensin or isoprenahne in the rat. BrainRes.24th180-l 84.

218

TANAKA

ET AL.

7. LIND, R. W., L. W. SWANSON, AND D. GANTEN. 1984. Angiotensin II immunoreactivity in the neural afferents and efferents of the subfomical organ of the rat. Bruin Res. 321: 209-2 15. 8. LIND, R. W., L. W. SWANSON, AND D. GANTEN. 1985. Organization of angiotensin II immunoreactive cells and fibers in the rat central nervous system: an immunohistochemical study. Neuroendocrinology 40: 2-24. 9. LIND, R. W., G. W. VAN HOESEN, AND A. K. JOHNSON. 1982. An HRP study of the connections of the subfomical organ of the rat. J. Comp. Neurol. 210: 265-277. 10. MISELIS, R. R. 1981. The efferent projections of the subfomical organ of the rat: a circumventricuiar organ within a neural network subserving water balance. Brain Res. 230: l-23. 11. PAXINOS, G., AND C. WATSON. 1982. The Rat Brain in Stereotaric Coordinates. Academic Press, New York. 12. POULAIN, D. A., AND J. B. WAKFXLEY. 1982. Electrophysiology of hypothalamic magnocellular neurones secreting oxytocin and vasopressin. Neuroscience 7: 773-808. 13. SGRO, S., A. V. FERGUSON, AND L. P. RENAUD. 1984. Subfomical organ-supraoptic necleus connections: an electrophysiologic study in the rat. Brain Res. 303: 7- 13. 14. TANAKA, J., H. KABA, H. SAITO, AND K. SETO. 1986. Angiotensin II-sensitive neurons in the rat lateral hypothalamic area with efferent projections to the subfomical organ. Exp. Neural. 94: 79 l-795.