Brain Research, 535 (1990) 159-162 Elsevier
159
BRES 24415
Iontophoresis of cortisol inhibits responses of identified paraventricular nucleus neurones to sciatic nerve stimulation David Saphier I and Shaul Feldman 2 1Department of Pharmacology and Therapeutics, Louisiana State University Medical School, Shreveport, LA 71130-3932 (U.S.A.) and 2Laboratory of Neurophysiology, Department of Neurology, Hadassah University Hospital and Hadassah-Hebrew University Medical School, Jerusalem (Israel)
(Accepted 21 August 1990) Key words: Electrophysiology; Sciatic nerve; Paraventricular nucleus; Corticotropin-releasing factor; Iontophoresis; Cortisol; Feedback
Responses of paraventricular nucleus (PVN) neurones were examined following stimulation of the sciatic nerve, and concomitant with iontophoretic application of cortisol. Sciatic nerve stimulation excited the majority of cells (22/24, 92%) and iontophoretic application of cortisol reduced the spontaneous activity of 16 of the cells tested (67%). Cortisol prevented neuronal responses to sciatic nerve stimulation in 11 cases (50%) but some of the cells inhibited by the steroid still responded to the stimulation, whilst some cells unaffected by cortisol alone were found not to respond during exposure to the stimulus. These results indicate an inhibitory role for glucocorticoids in the regulation of PVN neuronal activity and responses to afferent neural stimuli.
Neurosecretory cells of the dorsal medial parvocellular part of the hypothalamic paraventricular nucleus (PVN) synthesize corticotropin-releasing factor (CRF) and send their axon projections to the median eminence 1'2°,32,33, CRF being the primary hypothalamic secretagogue regulating hypothalamo-hypophysial-adrenocortical ( H H A ) activity23'31-33. Other cell types are also present within the parvocellular components of the PVN, such as those secreting thyrotropin-releasing hormone and several other neurohormones 32, but the majority are CRF immunoreactive 1'2°'32'33. Electrophysiological studies have provided information regarding the neural pathways and chemical messengers regulating activity of PVN parvocellular neurones and adrenocortical secretion TM 12,14-16,21,25-29
The mechanism(s) by which glucocorticoid hormones exert their negative feedback actions upon the H H A axis has been the subject of numerous investigations and is still the subject of some controversy. Most investigators now agree that the primary site of action is at the level of the central nervous system, rather than the pituitary gland, but the site, or sites, within the brain is still open to question. A number of studies indicate that the hypothalamus may be an important site of negative feedback 3'4'6"8A3'17'18, although other sites such as the septo-hippocampal complex may also play important roles 5'7"1°'24. In relation to this, electrophysiological
studies have demonstrated that putative CRF-secreting neurones may be inhibited by local exposure to glucocorticoid hormones 15,x6,28, suggesting that such cells may be involved in a feedback action of glucocorticoids. In addition, corticosterone administration has been shown to cause a dose-dependent inhibition of PVN 2t and other hypothalamic activity 1° following exposure to stressful neurogenic stimuli. In order to further evaluate the role of glucocorticoids at the level of the PVN, we herein demonstrate the effectiveness of the iontophoretic application of cortisol in blocking the responses of single putative CRFsecreting neurones following stimulation of the sciatic nerve. Experiments were performed on adult male rats of the Hebrew University strain, weighing 240-280 g, maintained in a temperature-controlled room on a 12 h light:12 h dark photoperiod; food and water were available ad libitum. Anaesthesia was induced with a single i.p. injection of urethane (ethyl carbamate, 1.2-1.3 g/kg; 25% w/v solution). Rats were secured in a stereotaxic head-holder and frame, inverted to present the ventral aspect of the animal. The sciatic nerve was exposed and resting on a pair of silver wire electrodes mounted in a perspex 'cup'. The median eminence and neurohypohysial stalk were exposed using a transpharyngeal approach and a non-concentric bipolar electrode was
Correspondence: D. Saphier, Department of Pharmacology and Therapeutics, Louisiana State University Medical School, P.O. Box 33932, Shreveport, LA 71130-3932, U.S.A.
0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
160 rested lightly on the median eminence at the level of the long preeminential portal capillary plexus, with another placed in the neural stalk. These electrodes served to differentiate between tuberoinfundibular neurones projecting to the median eminence and tuberohypophysial neurones projecting to the neurophypophysis; all latter neurones being excluded from the analysis. Test criteria for antidromic invasion of PVN tuberoinfundibular neutones were: (a) constant latency of the antidromic action potential following median eminence stimulation; (b) collision testing, demonstrating the ability of spontaneous action potentials to collide with and cancel antidromic action potentials; (c) the ability of neurones to conduct, at a constant latency, antidromic action potentials generated by short trains of high frequency stimulation. Stimulation was applied using a Grass model S-88 stimulator to deliver bipolar square-wave (leading edge, cathodic) pulses of up to 1 m A p e a k - t o - p e a k current strength and 1 ms total duration. Glass microelectrodes filled with 2 M sodium chloride were used to record extracellular signals from single PVN units which were amplified and displayed using a conventional electrophysiological recording system. Glued onto the recording electrode was a multibarrel micropipette, with a tip d i a m e t e r of less than 5 /tm, for the iontophoretic ejection of the following agents: (a) monosodium glutamate, 0.025 M at p H 6.8, used to test the micropipette patency; (b) cortisol sodium hemisuccinate, 0.025 M at p H 7.4; (c) 2% Alcian blue dye in 0.5 M sodium acetate, for current balancing and dye-marking of the position of r e c o r d e d units. The drugs were ejected using a W P instruments iontophoresis apparatus, set to deliver between 5 and 250 n A ; an opposing retaining current of 5 - 1 0 n A was used to prevent leakage of the drugs. Effects of iontophoretic application were usually considered to be significant if changes in firing rate of a p p r o x i m a t e l y 25%, or greater, were observed 28. Responses to electrical stimulation of the sciatic nerve were also examined, and during iontophoretic application of cortisol, to d e t e r m i n e whether this drug could alter any responses. A t the end of experiments, rats were perfused transcardially with 10% formalin/saline and the brain was r e m o v e d and stored in the same solution. Frozen sections were cut at 45 /~m for histological verification of the locations of r e c o r d e d PVN units, which were c o m p a r e d with the r e c o r d e d m e a s u r e m e n t s of the movements of the electrode microdrive, as d e t e r m i n e d under visual control, via the dissecting microscope, at the time of the experiment. The Xa-test was e m p l o y e d for statistical examination of response distributions. Extracellular single unit activity was recorded from 24 cells and these were located at a mean position 2.23 + 0.14 m m above the median eminence, corresponding with
Q
._c
20]
f-=5.2 HZ - C o n t r o l
16-
"~. 1 2 •
(/1 4,
b
Of'=4.O Hz - Cortisol, 25 nA
16
c m12.
4, 0, 0
12.5 HI!
25 T i m e - sec
37,5
50
Fig. l, Peristimulus histograms showing the response of a single PVN neurone, identified as projecting to the median eminence, following a 2 s train of 50 Hz stimulation delivered to the sciatic nerve (arrows). a: control trace, with the cell showing a longduration, sustained period of increased activity in response to the stimulation, b: during iontophoretic application of cortisol, an overall decrease in spontaneous activity is observed and there is no response to sciatic nerve stimulation. the CRF-rich c o m p o n e n t of the PVN. The mean ( + S . E . M . ) firing rate was 3.9 + 0.7 Hz and all the cells were antidromically identified as projecting to the median eminence, with a m e a n antidromic invasion latency of 10.8 + 0.9 ms, suggesting a conduction velocity of less than 0.21 m/sec, consistent with the p r o p e r t i e s of small c-fibers. Stimulation of the sciatic nerve e v o k e d excitatory responses from 22 of the cells (92%; Figs. l a and 2a), one cell being inhibited and another, non-responsive (Table I). For 8 of the cells, when the latency was d e a r l y defined, the m e a n onset of responses was r e c o r d e d at 171 + 41 msec and the mean offset was 496 + 117 ms (Fig. l a ) , the remaining cells showing no clear temporal-
TABLE I Responses o f P V N neurones, identified as projecting to the median eminence, following sciatic nerve stimulation and iontophoretic application of cortisol
(n) represents the total number of cells tested in each group or sub-group (superscript letters), n-, number of cells inhibited by the stimulus; n +, number of cells excited by the stimulus; n-NR, number of cells non-responsive to the stimulus. Test
Cortisol iontophoresis Sciatic nerve stimulation Sciatic nerve (cortisol) Sciatic nerve (cortisol-) Sciatic nerve (cortisol NR)
(n)
(24) (24) (22)c* (16)a (8) b
Response n-
n+
n-NR
16a 1 0 1 0
0 22c 11 9 2
8b 1 11 6 6
*P < 0.005 compared with sciatic nerve stimulation alone (XZ-test).
. _
161
~ 1~ t f'=3"13 Hz - C°ntr°l
Ob 10 f'~1.27 Hz - Cortisol, 25 r ~
20-
- 160
0 Stimulus !
160 320 Post-Stimutus Time - msec
480 640 (64 Epochs)
Fig. 2. Peristimulus histograms showing the response of a single PVN neurone, identified as projecting to the median eminence, following 64 single stimuli (0.2 Hz) delivered to the sciatic nerve (arrow). a: control trace, with the cell showing stimulus-locked activation in response to the stimulation, b: during iontophoretic application of cortisol, an overall decrease in spontaneous activity is observed but the response to sciatic nerve stimulation is still seen. locking of the response (Fig. 2a). Iontophoretic application of cortisol using currents ranging from 5-50 nA (mean, 33 + 6 nA), decreased the spontaneous activity of 16 of the 22 excited cells examined (Figs. lb and 2b), the remaining cells not being affected (Table I). The onset of this response was usually rapid (within a few seconds), and sometimes preceded by a tendency to a bursting pattern of firing; offset was slow and the inhibitory effects often lasted for several minutes. Sciatic nerve stimulation delivered during cortisol application was found to be ineffective in evoking excitation in 11 cases (50%; Fig. lb); this included 6 of the 8 cells whose spontaneous activity was unaffected by cortisol (Table I). As we have demonstrated in a previous study 28, iontophoretic application of cortisol or corticosterone is able to inhibit the spontaneous activity of the majority of antidromically identified, putative CRF-secreting neurones 25-29 of the PVN. This has also been confirmed in vitro, using hypothalamic slices from adrenalectomized rats 15. Taken together with data from neuroendocrine studies indicating that the hypothalamus may be a direct site for the negative feedback actions of glucocorticoid hormones 3'4'6's, these results suggest that not only may such hormones (locally) inhibit the synthesis of the A C T H secretagogues, CRF and V P 13'17'1s'23"30 but that a fast feedback effect, probably mediated by a membranebound receptor site, also exists at the level of the PVN. It is believed that the former feedback effects on synthesis and secretion are probably mediated via an interaction with the intracellular glucocorticoid receptor demonstrated within the P V N 1, whilst the fast effects are mediated by an interaction of the steroid with the
~-aminobutyric acid-A (GABAA) receptor complex, at a site on or near the barbiturate site 19. The results of the present study indicate the presence of an additional inhibitory effect of glucocorticoids at the level of the PVN since the responses to sciatic nerve stimulation of cells recorded were prevented, suggesting an effect of the steroid, upon synaptic transmission, akin to the inhibition of noradrenaline-evoked excitation observed by Kasai and Yamashita 16. This probably represents an inhibitory effect either upon neurotransmitter release or postsynaptic neurotransmitter receptor sensitivity but the precise mechanism remains to be further characterized. Stimulation of the sciatic nerve causes an increase in the activity of the putative CRF-secreting PVN neurones 11'25"26'28'29 and we believe that this is specifically related to activation of the H H A axis 9'11. Corticosterone decreases PVN multiunit activity following stressful neurogenic stimuli including stimulation of the sciatic nerve 1°'21 and intrahypothalamic glucocorticoid implants can prevent H H A activation following sciatic nerve stimulation 6'8. In view of these observations, it is significant that we demonstrate that the responses of some PVN neurones to sciatic nerve stimulation may be prevented or reduced by iontophoretic application of cortisol and we suggest that the results of this study represent an electrophysiological correlate of a negative feedback effect of cortisol, upon H H A axis sensitivity, exerted at the level of putative CRF-secreting neurones in the PVNo Interestingly, not all of the cells inhibited by cortisol alone were found to stop responding to stimulation of the sciatic nerve, and some cells unaffected by cortisol were not found to respond to the stimulus, perhaps suggesting a heterogeneity of the population studied, in terms of their sensitivity to cortisol or their neurotransmitter input. A further significant observation was that the majority of responses (64%) recorded following sciatic nerve stimulation were of a generalized nature with poorly defined onset and duration. This result is similar to that observed in two previous reports, following stimulation of catecholaminergic structures 25,29 and may suggest a lack of synaptic specialization at the level of the terminal arbor within the PVN, or at another level of the signal processing 2,22. In conclusion, the increase in the PVN unit activity following the afferent stimulus is probably directly related to an increased adrenocorticotrophic discharge 811, whilst the inhibitory effect of the iontophoretic application of cortisol probably represents an electrophysiological correlate of a negative feedback action of glucocorticoids exerted upon synaptic transmission at the level of the hypothalamus. Thus, cortisol decreased the PVN unit responses to the sensory stimulation, just as
162 local i m p l a n t s of g l u c o c o r t i c o i d s in the h y p o t h a l a m u s r e d u c e h y p o t h a l a m i c C R F activity 3'4"6,13,1718,3° and block a d r e n o c o r t i c a i r e s p o n s e s to n e u r a l stimuli 6'8. 1 Agnati, L.E, Fuxe, K., Yu, Z.Y., Harfstrand, A., Okret, S., Wilkstrom, A.-C., Goldstein, M., Zoli, M., Vale, W. and Gustafsson, J.-A., Morphometrical analysis of the distribution of corticotropin releasing factor, glucocorticoid receptor and phenylethanolamine N-methyltransferase immunoreactive structures in the paraventricular hypothalamic nucleus of the rat, Neurosci. Lett., 54 (1985) 147-152. 2 Beaudet, A. and Descarries, L., The monoamine innervation of rat cerebral cortex: synaptic and nonsynaptic axon terminals, Neuroscience, 3 (1978) 851-860. 3 Bohus, B. and Endroczi, E., Effect of intracerebral implantation of hydrocortisone on adrenocortical secretion and adrenal weight after unilateral adrenalectomy, Acta Physiol. Acad. Sci. Hung., 25 (1964) 11-19. 4 Chowers, I., Conforti, N. and Feldman, S., Effect of corticosteroids on hypothalamic corticotropin releasing factor and pituitary ACTH content, Neuroendocrinology, 2 (1967) 193-199. 5 Dallman, M.E, Levin, N., Cascio, C.S., Akana, S.E, Jacobson, L. and Kuhn, R.W., Pharmacological evidence that the inhibition of diurnal adrenocorticotropin secretion by corticosteroids is mediated via type I corticosterone-preferring receptors, Endocrinology, 124 (1989) 2844-2850. 6 Davidson, J.M., Jones, L.E. and Levine, S., Feedback regulation of adrenocorticotropin secretion in 'basal' and 'stress' conditions: acute and chronic effects of intrahypothalamic corticoid implantation, Endocrinology, 82 (1968) 655-663. 7 Feldman, S. and Conforti, N., Participation of the dorsal hippocampus in the glucocorticoid feedback effect on adrenocortical activity, Neuroendocrinology, 530 (1980) 52-55. 8 Feldman, S., Conforti, N., Chowers, I. and Davidson, J.M., Differential responses to various ACTH-releasing stimuli in rats with hypothalamic implants of corticosterone, Neuroendocrinology, 5 (1969) 290-302. 9 Feldman, S., Conforti, N. and Melamed, E., Involvement of ventral noradrenergic bundle in corticosterone secretion following neural stimuli, Neuropharmacology, 27 (1988) 129-133. 10 Feldman, S., Papir-Kricheli, D. and Dafny, N., Single-cell and multiunit activity in freely moving rats after corticosterone administration, Exp. Neurol., 80 (1983) 427-438. 11 Feldman, S. and Saphier, D., Extrahypothalamic neural afferents and the role of neurotransmitters in the regulation of adrenocortical secretion. In EC. Rose (Ed.), The Control of the Hypothalamo-Pituitary-Adrenocortical Axis, International Universities Press, Madison, CT, 1989, pp. 297-316. 12 Hamamura, M., Onaka, T. and Yagi, K., Parvocellular neurosecretory neurons: converging inputs after saphenous nerve and hypovolemic stimulations in the rats, Jpn. J. Physiol., 36 (1986) 921-933. 13 Harbuz, M.S. and Lightman, S.L., Glucocorticoid inhibition of stress-induced changes in hypothalamic corticotrophin-releasing factor messenger RNA and proenkephalin A messenger RNA, Neuropeptides, 14 (1989) 17-20. 14 Kannan, H., Kasai, M., Osaka, T. and Yamashita, H., Neurons in the paraventricular nucleus projecting to the median eminence: a study of their afferent connections from peripheral baroreceptors, and from the Al-catecholaminergic area in the ventrolateral medulla, Brain Research, 409 (1987) 358-363. 15 Kasai, M. and Yamashita, H., Inhibition by cortisol of neurons in the paraventricular nucleus of the hypothalamus in adrenalectomized rats: an in vitro study, Neurosci. Lea.. 91 (1988) 59-64. 16 Kasai, M. and Yamashita, H., Cortisol suppresses noradrenaline-induced excitatory responses of neurons in the paraventric-
This work was supported by the Etty and Miguel Meilichson Neuroscience Research Fund, and the Lena P. Harvey Endowment Fund for Neurological Research.
ular nucleus; an in vitro study, Neurosci. Lett., 91 (1988) 65-70. 17 Kov~ics, K.J. and Makara, G.B., Corticosterone and dexamethasone act at different brain sites to inhibit adrenalectomyinduced adrenocorticotropin hypersecretion, Brain Research, 474 (1988) 205-210. 18 Kov~ics, K.J. and Mezey, E., Dexamethasone inhibits corticotropin-releasing factor gene expression in the rat paraventricular nucleus, Neuroendocrinology, 46 (1987) 365-368. 19 Majewska, M.D., Harrison, N.L., Schwartz, R.D., Barker, J.L. and Paul, S.M., Steroid hormone metabolites are barbituratelike modulators of the GABA receptors, Science, 232 (1986) 1004-1007. 20 Merchenthaler, I., Hynes, M.A., Vigh, S., Schally, A.V. and Petrusz, P., Corticotropin releasing factor (CRF): origin and course of afferent pathways to the median eminence (ME) of the rat hypothalamus, Neuroendocrinology, 39 (1984) 296-306. 21 Mor, G., Saphier, D. and Feldman, S., Inhibition by corticosterone of paraventricular nucleus multiple unit activity responses in freely moving rats, Exp. Neurol., 94 (1986) 391-399. 22 Pickel, V.M., Ultrastructure of central catecholaminergic neurons. In P. Panula, H. Paivarinta and S. Soinila (Eds.), Neurohistochernistry: Modern Methods and Applications, Alan Liss, New York, 1986, pp. 397-423. 23 Plotsky, P.M. and Sawchenko, P.E., Hypophyseal-portal levels, measured median eminence content, and immunohistochemical staining of corticotropin-releasing factor, arginine vasopressin and oxytocin after pharmacological adrenalectomy, Endocrinology, 120 (1987) 1361-1371. 24 Saphier, D., Cortisol alters firing rate and synaptic responses of limbic forebrain units, Brain Res. Bull., 19 (1987) 519-524. 25 Saphier, D., Catecholaminergic projections to tuberoinfundibular neurones of the paraventricular nucleus: I. Effects of stimulation of A1, A2, A6 and C2 cell groups, Brain Res. Bull., 23 (1989) 389-395. 26 Saphier, D. and Feldman, S., Effects of neural stimuli on paraventricular nucleus neurones, Brain Res. Bull., 14 (1985) 401-407. 27 Saphier, D. and Feldman, S., Effects of stimulation of the preoptic area on hypothalamic paraventricular nucleus unit activity and corticosterone secretion in freely moving rats, Neuroendocrinology, 42 (1986) 167-173. 28 Saphier, D. and Feldman, S., Iontophoretic application of glucocorticoids inhibits identified neurones in the rat paraventricular nucleus, Brain Research, 453 (1988) 183-190. 29 Saphier, D. and Feldman, S., Catecholaminergic projections to tuberoinfundibular neurones of the paraventricular nucleus: II. Effects of stimulation of the ventral noradrenergic bundle: evidence for cotransmission, Brain Res. Bull., 23 (1989) 397404. 30 Sawchenko, P.E., Evidence for a local site of action for glucocorticoids in inhibiting CRF and vasopressin expression in the paraventricular nucleus, Brain Research, 403 (1987) 213224. 31 Swanson, L.W. and Sawchenko, P.E., Paraventricular nucleus: a site for the integration of neuroendocrine and autonomic mechanisms, Neuroendocrinology, 31 (1980) 410-417. 32 Swanson, L.W. and Sawchenko, P.E., Hypothalamic integration: organization of the paraventricular and supraoptic nuclei, Annu. Rev. Neurosci., 6 (1983) 269-324. 33 Swanson, L.W., Sawchenko, P.E., Rivier, C. and Vale, W., The organization of ovine corticotropin releasing factor (CRF)immunoreactive cells and fibers in the rat brain: an immunohistochemical study, Neuroendocrinology, 36 (1983) 165-186.
159
BRES 24415
Iontophoresis of cortisol inhibits responses of identified paraventricular nucleus neurones to sciatic nerve stimulation David Saphier I and Shaul Feldman 2 1Department of Pharmacology and Therapeutics, Louisiana State University Medical School, Shreveport, LA 71130-3932 (U.S.A.) and 2Laboratory of Neurophysiology, Department of Neurology, Hadassah University Hospital and Hadassah-Hebrew University Medical School, Jerusalem (Israel)
(Accepted 21 August 1990) Key words: Electrophysiology; Sciatic nerve; Paraventricular nucleus; Corticotropin-releasing factor; Iontophoresis; Cortisol; Feedback
Responses of paraventricular nucleus (PVN) neurones were examined following stimulation of the sciatic nerve, and concomitant with iontophoretic application of cortisol. Sciatic nerve stimulation excited the majority of cells (22/24, 92%) and iontophoretic application of cortisol reduced the spontaneous activity of 16 of the cells tested (67%). Cortisol prevented neuronal responses to sciatic nerve stimulation in 11 cases (50%) but some of the cells inhibited by the steroid still responded to the stimulation, whilst some cells unaffected by cortisol alone were found not to respond during exposure to the stimulus. These results indicate an inhibitory role for glucocorticoids in the regulation of PVN neuronal activity and responses to afferent neural stimuli.
Neurosecretory cells of the dorsal medial parvocellular part of the hypothalamic paraventricular nucleus (PVN) synthesize corticotropin-releasing factor (CRF) and send their axon projections to the median eminence 1'2°,32,33, CRF being the primary hypothalamic secretagogue regulating hypothalamo-hypophysial-adrenocortical ( H H A ) activity23'31-33. Other cell types are also present within the parvocellular components of the PVN, such as those secreting thyrotropin-releasing hormone and several other neurohormones 32, but the majority are CRF immunoreactive 1'2°'32'33. Electrophysiological studies have provided information regarding the neural pathways and chemical messengers regulating activity of PVN parvocellular neurones and adrenocortical secretion TM 12,14-16,21,25-29
The mechanism(s) by which glucocorticoid hormones exert their negative feedback actions upon the H H A axis has been the subject of numerous investigations and is still the subject of some controversy. Most investigators now agree that the primary site of action is at the level of the central nervous system, rather than the pituitary gland, but the site, or sites, within the brain is still open to question. A number of studies indicate that the hypothalamus may be an important site of negative feedback 3'4'6"8A3'17'18, although other sites such as the septo-hippocampal complex may also play important roles 5'7"1°'24. In relation to this, electrophysiological
studies have demonstrated that putative CRF-secreting neurones may be inhibited by local exposure to glucocorticoid hormones 15,x6,28, suggesting that such cells may be involved in a feedback action of glucocorticoids. In addition, corticosterone administration has been shown to cause a dose-dependent inhibition of PVN 2t and other hypothalamic activity 1° following exposure to stressful neurogenic stimuli. In order to further evaluate the role of glucocorticoids at the level of the PVN, we herein demonstrate the effectiveness of the iontophoretic application of cortisol in blocking the responses of single putative CRFsecreting neurones following stimulation of the sciatic nerve. Experiments were performed on adult male rats of the Hebrew University strain, weighing 240-280 g, maintained in a temperature-controlled room on a 12 h light:12 h dark photoperiod; food and water were available ad libitum. Anaesthesia was induced with a single i.p. injection of urethane (ethyl carbamate, 1.2-1.3 g/kg; 25% w/v solution). Rats were secured in a stereotaxic head-holder and frame, inverted to present the ventral aspect of the animal. The sciatic nerve was exposed and resting on a pair of silver wire electrodes mounted in a perspex 'cup'. The median eminence and neurohypohysial stalk were exposed using a transpharyngeal approach and a non-concentric bipolar electrode was
Correspondence: D. Saphier, Department of Pharmacology and Therapeutics, Louisiana State University Medical School, P.O. Box 33932, Shreveport, LA 71130-3932, U.S.A.
0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
160 rested lightly on the median eminence at the level of the long preeminential portal capillary plexus, with another placed in the neural stalk. These electrodes served to differentiate between tuberoinfundibular neurones projecting to the median eminence and tuberohypophysial neurones projecting to the neurophypophysis; all latter neurones being excluded from the analysis. Test criteria for antidromic invasion of PVN tuberoinfundibular neutones were: (a) constant latency of the antidromic action potential following median eminence stimulation; (b) collision testing, demonstrating the ability of spontaneous action potentials to collide with and cancel antidromic action potentials; (c) the ability of neurones to conduct, at a constant latency, antidromic action potentials generated by short trains of high frequency stimulation. Stimulation was applied using a Grass model S-88 stimulator to deliver bipolar square-wave (leading edge, cathodic) pulses of up to 1 m A p e a k - t o - p e a k current strength and 1 ms total duration. Glass microelectrodes filled with 2 M sodium chloride were used to record extracellular signals from single PVN units which were amplified and displayed using a conventional electrophysiological recording system. Glued onto the recording electrode was a multibarrel micropipette, with a tip d i a m e t e r of less than 5 /tm, for the iontophoretic ejection of the following agents: (a) monosodium glutamate, 0.025 M at p H 6.8, used to test the micropipette patency; (b) cortisol sodium hemisuccinate, 0.025 M at p H 7.4; (c) 2% Alcian blue dye in 0.5 M sodium acetate, for current balancing and dye-marking of the position of r e c o r d e d units. The drugs were ejected using a W P instruments iontophoresis apparatus, set to deliver between 5 and 250 n A ; an opposing retaining current of 5 - 1 0 n A was used to prevent leakage of the drugs. Effects of iontophoretic application were usually considered to be significant if changes in firing rate of a p p r o x i m a t e l y 25%, or greater, were observed 28. Responses to electrical stimulation of the sciatic nerve were also examined, and during iontophoretic application of cortisol, to d e t e r m i n e whether this drug could alter any responses. A t the end of experiments, rats were perfused transcardially with 10% formalin/saline and the brain was r e m o v e d and stored in the same solution. Frozen sections were cut at 45 /~m for histological verification of the locations of r e c o r d e d PVN units, which were c o m p a r e d with the r e c o r d e d m e a s u r e m e n t s of the movements of the electrode microdrive, as d e t e r m i n e d under visual control, via the dissecting microscope, at the time of the experiment. The Xa-test was e m p l o y e d for statistical examination of response distributions. Extracellular single unit activity was recorded from 24 cells and these were located at a mean position 2.23 + 0.14 m m above the median eminence, corresponding with
Q
._c
20]
f-=5.2 HZ - C o n t r o l
16-
"~. 1 2 •
(/1 4,
b
Of'=4.O Hz - Cortisol, 25 nA
16
c m12.
4, 0, 0
12.5 HI!
25 T i m e - sec
37,5
50
Fig. l, Peristimulus histograms showing the response of a single PVN neurone, identified as projecting to the median eminence, following a 2 s train of 50 Hz stimulation delivered to the sciatic nerve (arrows). a: control trace, with the cell showing a longduration, sustained period of increased activity in response to the stimulation, b: during iontophoretic application of cortisol, an overall decrease in spontaneous activity is observed and there is no response to sciatic nerve stimulation. the CRF-rich c o m p o n e n t of the PVN. The mean ( + S . E . M . ) firing rate was 3.9 + 0.7 Hz and all the cells were antidromically identified as projecting to the median eminence, with a m e a n antidromic invasion latency of 10.8 + 0.9 ms, suggesting a conduction velocity of less than 0.21 m/sec, consistent with the p r o p e r t i e s of small c-fibers. Stimulation of the sciatic nerve e v o k e d excitatory responses from 22 of the cells (92%; Figs. l a and 2a), one cell being inhibited and another, non-responsive (Table I). For 8 of the cells, when the latency was d e a r l y defined, the m e a n onset of responses was r e c o r d e d at 171 + 41 msec and the mean offset was 496 + 117 ms (Fig. l a ) , the remaining cells showing no clear temporal-
TABLE I Responses o f P V N neurones, identified as projecting to the median eminence, following sciatic nerve stimulation and iontophoretic application of cortisol
(n) represents the total number of cells tested in each group or sub-group (superscript letters), n-, number of cells inhibited by the stimulus; n +, number of cells excited by the stimulus; n-NR, number of cells non-responsive to the stimulus. Test
Cortisol iontophoresis Sciatic nerve stimulation Sciatic nerve (cortisol) Sciatic nerve (cortisol-) Sciatic nerve (cortisol NR)
(n)
(24) (24) (22)c* (16)a (8) b
Response n-
n+
n-NR
16a 1 0 1 0
0 22c 11 9 2
8b 1 11 6 6
*P < 0.005 compared with sciatic nerve stimulation alone (XZ-test).
. _
161
~ 1~ t f'=3"13 Hz - C°ntr°l
Ob 10 f'~1.27 Hz - Cortisol, 25 r ~
20-
- 160
0 Stimulus !
160 320 Post-Stimutus Time - msec
480 640 (64 Epochs)
Fig. 2. Peristimulus histograms showing the response of a single PVN neurone, identified as projecting to the median eminence, following 64 single stimuli (0.2 Hz) delivered to the sciatic nerve (arrow). a: control trace, with the cell showing stimulus-locked activation in response to the stimulation, b: during iontophoretic application of cortisol, an overall decrease in spontaneous activity is observed but the response to sciatic nerve stimulation is still seen. locking of the response (Fig. 2a). Iontophoretic application of cortisol using currents ranging from 5-50 nA (mean, 33 + 6 nA), decreased the spontaneous activity of 16 of the 22 excited cells examined (Figs. lb and 2b), the remaining cells not being affected (Table I). The onset of this response was usually rapid (within a few seconds), and sometimes preceded by a tendency to a bursting pattern of firing; offset was slow and the inhibitory effects often lasted for several minutes. Sciatic nerve stimulation delivered during cortisol application was found to be ineffective in evoking excitation in 11 cases (50%; Fig. lb); this included 6 of the 8 cells whose spontaneous activity was unaffected by cortisol (Table I). As we have demonstrated in a previous study 28, iontophoretic application of cortisol or corticosterone is able to inhibit the spontaneous activity of the majority of antidromically identified, putative CRF-secreting neurones 25-29 of the PVN. This has also been confirmed in vitro, using hypothalamic slices from adrenalectomized rats 15. Taken together with data from neuroendocrine studies indicating that the hypothalamus may be a direct site for the negative feedback actions of glucocorticoid hormones 3'4'6's, these results suggest that not only may such hormones (locally) inhibit the synthesis of the A C T H secretagogues, CRF and V P 13'17'1s'23"30 but that a fast feedback effect, probably mediated by a membranebound receptor site, also exists at the level of the PVN. It is believed that the former feedback effects on synthesis and secretion are probably mediated via an interaction with the intracellular glucocorticoid receptor demonstrated within the P V N 1, whilst the fast effects are mediated by an interaction of the steroid with the
~-aminobutyric acid-A (GABAA) receptor complex, at a site on or near the barbiturate site 19. The results of the present study indicate the presence of an additional inhibitory effect of glucocorticoids at the level of the PVN since the responses to sciatic nerve stimulation of cells recorded were prevented, suggesting an effect of the steroid, upon synaptic transmission, akin to the inhibition of noradrenaline-evoked excitation observed by Kasai and Yamashita 16. This probably represents an inhibitory effect either upon neurotransmitter release or postsynaptic neurotransmitter receptor sensitivity but the precise mechanism remains to be further characterized. Stimulation of the sciatic nerve causes an increase in the activity of the putative CRF-secreting PVN neurones 11'25"26'28'29 and we believe that this is specifically related to activation of the H H A axis 9'11. Corticosterone decreases PVN multiunit activity following stressful neurogenic stimuli including stimulation of the sciatic nerve 1°'21 and intrahypothalamic glucocorticoid implants can prevent H H A activation following sciatic nerve stimulation 6'8. In view of these observations, it is significant that we demonstrate that the responses of some PVN neurones to sciatic nerve stimulation may be prevented or reduced by iontophoretic application of cortisol and we suggest that the results of this study represent an electrophysiological correlate of a negative feedback effect of cortisol, upon H H A axis sensitivity, exerted at the level of putative CRF-secreting neurones in the PVNo Interestingly, not all of the cells inhibited by cortisol alone were found to stop responding to stimulation of the sciatic nerve, and some cells unaffected by cortisol were not found to respond to the stimulus, perhaps suggesting a heterogeneity of the population studied, in terms of their sensitivity to cortisol or their neurotransmitter input. A further significant observation was that the majority of responses (64%) recorded following sciatic nerve stimulation were of a generalized nature with poorly defined onset and duration. This result is similar to that observed in two previous reports, following stimulation of catecholaminergic structures 25,29 and may suggest a lack of synaptic specialization at the level of the terminal arbor within the PVN, or at another level of the signal processing 2,22. In conclusion, the increase in the PVN unit activity following the afferent stimulus is probably directly related to an increased adrenocorticotrophic discharge 811, whilst the inhibitory effect of the iontophoretic application of cortisol probably represents an electrophysiological correlate of a negative feedback action of glucocorticoids exerted upon synaptic transmission at the level of the hypothalamus. Thus, cortisol decreased the PVN unit responses to the sensory stimulation, just as
162 local i m p l a n t s of g l u c o c o r t i c o i d s in the h y p o t h a l a m u s r e d u c e h y p o t h a l a m i c C R F activity 3'4"6,13,1718,3° and block a d r e n o c o r t i c a i r e s p o n s e s to n e u r a l stimuli 6'8. 1 Agnati, L.E, Fuxe, K., Yu, Z.Y., Harfstrand, A., Okret, S., Wilkstrom, A.-C., Goldstein, M., Zoli, M., Vale, W. and Gustafsson, J.-A., Morphometrical analysis of the distribution of corticotropin releasing factor, glucocorticoid receptor and phenylethanolamine N-methyltransferase immunoreactive structures in the paraventricular hypothalamic nucleus of the rat, Neurosci. Lett., 54 (1985) 147-152. 2 Beaudet, A. and Descarries, L., The monoamine innervation of rat cerebral cortex: synaptic and nonsynaptic axon terminals, Neuroscience, 3 (1978) 851-860. 3 Bohus, B. and Endroczi, E., Effect of intracerebral implantation of hydrocortisone on adrenocortical secretion and adrenal weight after unilateral adrenalectomy, Acta Physiol. Acad. Sci. Hung., 25 (1964) 11-19. 4 Chowers, I., Conforti, N. and Feldman, S., Effect of corticosteroids on hypothalamic corticotropin releasing factor and pituitary ACTH content, Neuroendocrinology, 2 (1967) 193-199. 5 Dallman, M.E, Levin, N., Cascio, C.S., Akana, S.E, Jacobson, L. and Kuhn, R.W., Pharmacological evidence that the inhibition of diurnal adrenocorticotropin secretion by corticosteroids is mediated via type I corticosterone-preferring receptors, Endocrinology, 124 (1989) 2844-2850. 6 Davidson, J.M., Jones, L.E. and Levine, S., Feedback regulation of adrenocorticotropin secretion in 'basal' and 'stress' conditions: acute and chronic effects of intrahypothalamic corticoid implantation, Endocrinology, 82 (1968) 655-663. 7 Feldman, S. and Conforti, N., Participation of the dorsal hippocampus in the glucocorticoid feedback effect on adrenocortical activity, Neuroendocrinology, 530 (1980) 52-55. 8 Feldman, S., Conforti, N., Chowers, I. and Davidson, J.M., Differential responses to various ACTH-releasing stimuli in rats with hypothalamic implants of corticosterone, Neuroendocrinology, 5 (1969) 290-302. 9 Feldman, S., Conforti, N. and Melamed, E., Involvement of ventral noradrenergic bundle in corticosterone secretion following neural stimuli, Neuropharmacology, 27 (1988) 129-133. 10 Feldman, S., Papir-Kricheli, D. and Dafny, N., Single-cell and multiunit activity in freely moving rats after corticosterone administration, Exp. Neurol., 80 (1983) 427-438. 11 Feldman, S. and Saphier, D., Extrahypothalamic neural afferents and the role of neurotransmitters in the regulation of adrenocortical secretion. In EC. Rose (Ed.), The Control of the Hypothalamo-Pituitary-Adrenocortical Axis, International Universities Press, Madison, CT, 1989, pp. 297-316. 12 Hamamura, M., Onaka, T. and Yagi, K., Parvocellular neurosecretory neurons: converging inputs after saphenous nerve and hypovolemic stimulations in the rats, Jpn. J. Physiol., 36 (1986) 921-933. 13 Harbuz, M.S. and Lightman, S.L., Glucocorticoid inhibition of stress-induced changes in hypothalamic corticotrophin-releasing factor messenger RNA and proenkephalin A messenger RNA, Neuropeptides, 14 (1989) 17-20. 14 Kannan, H., Kasai, M., Osaka, T. and Yamashita, H., Neurons in the paraventricular nucleus projecting to the median eminence: a study of their afferent connections from peripheral baroreceptors, and from the Al-catecholaminergic area in the ventrolateral medulla, Brain Research, 409 (1987) 358-363. 15 Kasai, M. and Yamashita, H., Inhibition by cortisol of neurons in the paraventricular nucleus of the hypothalamus in adrenalectomized rats: an in vitro study, Neurosci. Lea.. 91 (1988) 59-64. 16 Kasai, M. and Yamashita, H., Cortisol suppresses noradrenaline-induced excitatory responses of neurons in the paraventric-
This work was supported by the Etty and Miguel Meilichson Neuroscience Research Fund, and the Lena P. Harvey Endowment Fund for Neurological Research.
ular nucleus; an in vitro study, Neurosci. Lett., 91 (1988) 65-70. 17 Kov~ics, K.J. and Makara, G.B., Corticosterone and dexamethasone act at different brain sites to inhibit adrenalectomyinduced adrenocorticotropin hypersecretion, Brain Research, 474 (1988) 205-210. 18 Kov~ics, K.J. and Mezey, E., Dexamethasone inhibits corticotropin-releasing factor gene expression in the rat paraventricular nucleus, Neuroendocrinology, 46 (1987) 365-368. 19 Majewska, M.D., Harrison, N.L., Schwartz, R.D., Barker, J.L. and Paul, S.M., Steroid hormone metabolites are barbituratelike modulators of the GABA receptors, Science, 232 (1986) 1004-1007. 20 Merchenthaler, I., Hynes, M.A., Vigh, S., Schally, A.V. and Petrusz, P., Corticotropin releasing factor (CRF): origin and course of afferent pathways to the median eminence (ME) of the rat hypothalamus, Neuroendocrinology, 39 (1984) 296-306. 21 Mor, G., Saphier, D. and Feldman, S., Inhibition by corticosterone of paraventricular nucleus multiple unit activity responses in freely moving rats, Exp. Neurol., 94 (1986) 391-399. 22 Pickel, V.M., Ultrastructure of central catecholaminergic neurons. In P. Panula, H. Paivarinta and S. Soinila (Eds.), Neurohistochernistry: Modern Methods and Applications, Alan Liss, New York, 1986, pp. 397-423. 23 Plotsky, P.M. and Sawchenko, P.E., Hypophyseal-portal levels, measured median eminence content, and immunohistochemical staining of corticotropin-releasing factor, arginine vasopressin and oxytocin after pharmacological adrenalectomy, Endocrinology, 120 (1987) 1361-1371. 24 Saphier, D., Cortisol alters firing rate and synaptic responses of limbic forebrain units, Brain Res. Bull., 19 (1987) 519-524. 25 Saphier, D., Catecholaminergic projections to tuberoinfundibular neurones of the paraventricular nucleus: I. Effects of stimulation of A1, A2, A6 and C2 cell groups, Brain Res. Bull., 23 (1989) 389-395. 26 Saphier, D. and Feldman, S., Effects of neural stimuli on paraventricular nucleus neurones, Brain Res. Bull., 14 (1985) 401-407. 27 Saphier, D. and Feldman, S., Effects of stimulation of the preoptic area on hypothalamic paraventricular nucleus unit activity and corticosterone secretion in freely moving rats, Neuroendocrinology, 42 (1986) 167-173. 28 Saphier, D. and Feldman, S., Iontophoretic application of glucocorticoids inhibits identified neurones in the rat paraventricular nucleus, Brain Research, 453 (1988) 183-190. 29 Saphier, D. and Feldman, S., Catecholaminergic projections to tuberoinfundibular neurones of the paraventricular nucleus: II. Effects of stimulation of the ventral noradrenergic bundle: evidence for cotransmission, Brain Res. Bull., 23 (1989) 397404. 30 Sawchenko, P.E., Evidence for a local site of action for glucocorticoids in inhibiting CRF and vasopressin expression in the paraventricular nucleus, Brain Research, 403 (1987) 213224. 31 Swanson, L.W. and Sawchenko, P.E., Paraventricular nucleus: a site for the integration of neuroendocrine and autonomic mechanisms, Neuroendocrinology, 31 (1980) 410-417. 32 Swanson, L.W. and Sawchenko, P.E., Hypothalamic integration: organization of the paraventricular and supraoptic nuclei, Annu. Rev. Neurosci., 6 (1983) 269-324. 33 Swanson, L.W., Sawchenko, P.E., Rivier, C. and Vale, W., The organization of ovine corticotropin releasing factor (CRF)immunoreactive cells and fibers in the rat brain: an immunohistochemical study, Neuroendocrinology, 36 (1983) 165-186.
