Opposing effects of intracerebroventricularly injected norepinephrine on oxytocin and vasopressin neurons in the paraventricular nucleus of the rat

Neuroscience Letters 244 (1998) 13–16

Opposing effects of intracerebroventricularly injected norepinephrine on oxytocin and vasopressin neurons in the paraventricular nucleus of the rat Yaping Ji*, Jun Mei, Shaogang Lu Department of Brain and Nerve, Xi’an Medical University, Xi’an, Shaanxi 710061, PR China Received 5 January 1998; received in revised form 28 January 1998; accepted 28 January 1998

Abstract Our previous study shows that intracerebroventricularly (i.c.v.) injected norepinephrine (NE) had different effects on the discharge of different firing patterns of magnocellular neurons in the paraventricular nucleus (PVN). In the present study we further classified antidromically identified magnocellular neurons into two groups: vasopressin (VP) and oxytocin (OT) secreting neurons, and found that of all 48 cases of magnocellular neurons, NE had mainly excitatory effects on 36 cases of putatively OTsecreting neurons, and inhibitory effects on 12 cases of VP-secreting neurons. The third ventricular injected NE had almost the same effect on two types of neurons as that of lateral ventricular injection, partly ruling out the possibility that the lateral ventricularly injected NE may have acted indirectly on the magnocellular neurons in PVN. The results show that different mechanisms may be involved in mediating the effect of i.c.v. injected NE on VP- and OT-secreting neurons in the PVN.  1998 Published by Elsevier Science Ireland Ltd.

Keywords: Norepinephrine; Oxytocin; Vasopressin; Paraventricular nucleus

The magnocellular neurosecretory neurons in the paraventricular nucleus (PVN) of the hypothalamus synthesize two important hormones, vasopressin (VP) and oxytocin (OT), and release them in the posterior hypophysis. It is well established that the PVN receive a dense noradrenergic projection of brain stem origin [6,10], but the role of noradrenergic afferents in regulating the activity of VP and OT neurons is still controversial. Former research works which iontophoretically applied noradrenaline into the PVN show an inhibitory effect on the firing of neurohypophysially projecting neurons in it [7], but stimulation of noradrenergic cell groups in the brain stem (A1 and A2 area) have an excitatory effect on both VP and OT neurons, and the effect can be abolished by 6-hydroxydopamine (a catecholaminespecific neurotoxin) pre-treatment, suggesting that the excitatory effects were actually mediated by noradrenergic fibers [3]. Recently Day et al. [3] reported that the excitatory effect of electrical stimulation of the A1 cell group can not be affected by pre-injection of phentolamine (a wide-spec* Corresponding author. Fax: +86 29 5267364.

trum noradrenergic antagonist), contradicting the above mentioned results [4]. Given the contradiction of the above mentioned different ways of drug administration, to further clarify the effect of NE on these neurons, we assessed the effect of intracerebroventricular (i.c.v.) injection of NE on PVN magnocellular neurosecreting neurons. It is well known that electrophysiological research on this nucleus shows that these magnocellular neurons display three main patterns of discharge, i.e. slow irregular, fast continuous, and phasic discharge [9]. Our former studies [5] showed that i.c.v. injection of norepinephrine had different effect on these different firing patterns of neurons. But, the effect of i.c.v. injected norepinephrine on different secreting types of neurons is still not known. We first identified PVN magnocellular neurosecreting neurons through antidromical stimulation of neurohypophysis, and then further classified these neurons into two groups, VP- and OT-secreting neurons. The effects of i.c.v. injection of NE were observed on these neurons. The cause of different effects of NE on magnocellular

0304-3940/98/$19.00  1998 Published by Elsevier Science Ireland Ltd. All rights reserved PII S0304- 3940(98) 00117- 7

14

Y. Ji et al. / Neuroscience Letters 244 (1998) 13–16

PVN neurons observed by different researchers was discussed, along with the significance of our findings. Adult Sprague–Dawley rats weighing 220–300 g were anaesthetized with urethane (20%, 1.1 g/kg). The external jugular was catheterized to enable administration of a peripherally active vasoconstrictor metaraminol (2–10 mg). After tracheal intubation, the rat was then placed in a stereotaxic instrument. The position of PVN, the lateral ventricle and neurohypophysis were determined according to the atlas of Paxinos and Waston [8] and then the parietal cortex was exposed and sagittal sinus ligated to insert recording electrode. Two other holes were drilled on the surface of the skull for inserting the stimulation electrodes and the guide cannula for drug administration. Electrocardiogram leads were attached for monitoring of heart rate, and body temperature was maintained at 37°C. A bipolar stimulation electrode was positioned in the neurohypophysis and a 500 ms current pulse of 0.1–2 mA was applied at a frequency of 2 Hz. Single unit activity was recorded in the PVN by using glass micropipettes filled with 2% pontamine sky blue in 0.5 mol/l sodium acetate acid solution (10–20 MQ impedance) and cells were identified as magnocellular neurosecretory neurons according to the criteria reported by Yagi et al. [14]. The magnocellular neurons were further identified as VPor OT-secreting neurons according to their response to intravenous (i.v.) injection of metaraminol (2–10 mg) as reported by Day et al. [3]. That is, units which were inhibited by i.v. injection of metaraminol were identified as VPsecreting, and those unresponsive to baroreceptor activation were identified as OT-secreting. To determine the effect of i.c.v. injection of NE on these antidromically identified magnocellular neurons, after waiting for a period of 3 min for control, a volume of 5 ml NE (10 − 3 mol/l) which were pre-dissolved in 0.9% saline were slowly injected into the lateral ventricle over a period of 2 min. In some experiments when more than one units were

Fig. 1. (A) Oscilloscope record showing the constant latency of a continuously firing PVN neuron after neurohypophysial stimulation. Five sweeps are superimposed. (B) Oscilloscope record showing ‘high frequency following’ character of the neuron. The interval between the two stimuli is 5 ms. (C) Oscilloscope record showing the result of the ‘collision test’. The stimulus delay was adjusted so as to abort the action potential following the first stimulus because of collision. Scale bars, (A,B) 10 ms, 0.5 mV; (C), 50 ms, 0.5 mV.

Fig. 2. Histological photograph showing the location of the recording site of a magnocellular paraventricular neuron. Scale bar, 2000 mm. The arrow indicates the marked spot of pontamine sky blue. Neutral red stain.

tested, a second injection was performed after at least 30 min. In control experiments, the same volume of 0.9% saline were injected. At the end of each experiment, potamine sky blue was iontophoresed by using cathodal current of 10–20 mA for 20–10 min to mark the recording site. The location of drug injection was marked by injecting 1 ml pontamine sky blue into the same place as that of ventricular injection of NE. Stimulation site was examined by re-placing the electrode in the same direction with the same parameters as in the

Fig. 3. (A) Histogram showing the spontaneous discharge of a continuously firing neurosecreting neuron was not affected by i.v. injected metaraminol. (B) Histogram showing the excitatory effect of i.c.v. injected NE on the same neuron.

15

Y. Ji et al. / Neuroscience Letters 244 (1998) 13–16

Fig. 4. (A) Histogram showing the spontaneous discharge of a continuously firing neurosecreting neuron was inhibited by i.v. injected metaraminol. (B) Histogram showing the inhibitory effect of i.c.v. injected NE on the same neuron.

experimental process after the brain tissue was taken out in order to see if it was positioned in the neurohypophyseal. Data were analyzed for statistical significance (P , 0.05) by the student’s t-test. In case that large intensity of stimulation may excite neuron fibres outside the posterior hypophysial, neurons with an intensity threshold of larger than 1 mA were omitted. Under this condition, a total of 48 units were identified as magnocellular neurosecretory neurons in the PVN area, of which 24 discharged continuously, 10 phasically and 14 slow irregularly. The mean threshold for eliciting antidromic action potential was 0.51 ± 0.25 mA, with a mean latency of 11.7 ± 3.2 ms. Fig. 1 shows the process of identifying a unit as a magnocellular neuron. The site of the recording electrode is shown in Fig. 2. All 10 phasically firing units were inhibited by i.v. administration of metaraminol and thus be classified as VP-secret-

ing. Of the 24 continuously firing units, 22 were classified as OT-secreting and two were VP-secreting according to their response to metaraminol. All 14 slow irregularly firing units were not classified by this means because the inhibitory response occured after metaraminol injection was not easy to determine in these neurons. Our observation shows that administration of norepinephrine into the lateral cerebral ventricle had different effects on these two kinds of neurosecreting neurons. Norepinephrine preferentially inhibited the spontaneous discharge of those neurons classified as VP-secreting, and excited those OT-secreting neurons and those 14 unclassified slow irregularly firing neurons (see Figs. 3, 4 and Table 1). Because of this similarity of response to i.c.v. injected NE we finally supposed that these neurons were OT-secreting ones. The effects of NE occured in 10 s after the beginning of i.c.v. injection, and lasted for about 8–10 min. No changes were found in the discharge rate of magnocellular neurons after i.c.v. injection of the same volume of saline. To exclude the possibility that norepinephrine may act directly on the neurons adjacent to the lateral ventricle and then indirectly on the neurons in PVN, in some experiments (on four OT-secreting neurons and one VP-secreting neuron) we compared the effect of injecting norepinephrine (2 ml, the same concentration as that of lateral ventricular injection) into the third ventricle with that of lateral ventricle injection, and the results show that the third ventricle injected NE had the same kind of effects as that of lateral ventricle injected NE, except that the effects lasted for a shorter period of time. Different ways of drug administration have been used to examine the effects of NE on magnocellular neurosecreting neurons of hypothalamus, but up to now the results seem to be controversial. For example, by iontophoresing NE into SON or PVN, some authors observed its inhibitory effect on VP and OT neurons, but, stimulation of A1 and A2 cell groups which send direct projecting fibre to magnocellular neurons show an excitatory effect on VP and OT neurons. To our knowledge, because iontophorese ejection can cause a high tip concentration in the microelectrode [1], and this high level of neurotransmitter may lead to pharmacological effect rather than physiological effect on magnocellular neurons, while though stimulation of A1 or A2 noradrenergic cell group can mimick natural release of NE that acts directly on neurosecreting neurons, it may also cause the fibres release some neuropeptides which coexisted with nor-

Table 1 Classification of PVN magnocellular neurons and their response to NE Discharge pattern

Slow irregular Fast continuous Phasic

No. of neurons 14 24 10

Response to metaraminol

Response to NE

Inhibitory

No response

Not determined

Excitatory

Inhibitory

0 2 10

0 22 0

14 0 0

14 22 0

0 2 10

16

Y. Ji et al. / Neuroscience Letters 244 (1998) 13–16

adrenaline in the somata of A1 and A2 cell group [2], and these peptides had been reported to affect the effects of NE on magnocellular neurons [11]. In our research, i.c.v. injected NE may reach the adjacent paraventricular nucleus through cerebrospinal fluid circulation and then act on the magnocellular neurons in a concentration very similar to that in the synaptic cleft where it was released. The effect of i.c.v. injection of NE on the spontaneous discharge of PVN neurosecretory neurons has not been reported before. The results of our present study show that the i.c.v. injection of NE had different effects on VP and OT neurons of PVN, and further, it partly contradicts the results of electrical stimulation of A1 and A2 cell groups. One may argue that i.c.v. injected NE may have acted directly on neurons somewhere near the lateral ventrical wall first and then indirectly affected the activity of the PVN magnocellular neurons. However, when we injected NE directly into the third ventrical, we saw the same effect on different neurosecreting neurons. This observation partly ruled out the above mentioned argument. The reason why NE had different effects on VP and OT neurons is still not known. In our former research we found that i.c.v. phentolamine can partly abolish the inhibitory effects of NE on VP neurons, but whether it can affect the effects of NE on OT neurons was not studied. To further determine whether different subtype of NE receptor mediated different effect of NE on VP- and OT-secreting neurons, it may be helpful to inject different antagonists of NE directly into the PVN to see if they can abolish the effect of i.c.v. injected NE. In electrophysiological research of PVN and SON magnocellular neurons, two main methods of classifying VPand OT-secreting neurons have been generally used until now. One way is to classify the neurons by observing their response to suckling [12,13], and this method can only be used in the research of female animals. Another way now widely used is to identify VP- and OT-secreting neurons by using baroreceptor activation. But, since those neurons which do not show inhibitory response to i.v. injection of metaraminol were classified as OT-secreting ones, sometimes it is very difficult for us to determine whether an inhibitory response occurred especially in units whose spontaneous discharge rate were very slow. In our experiment we found that slowly firing magnocellular neurons showed mainly excitatory response to i.c.v. injected NE and are likely to be OT neurons, so the secreting type of those slowly firing neurons can then be determined. In other words, our observation may provide a third method for classifying magnocellular neurons.

In conclusion, the present study demonstrated that i.c.v. injection of NE had quite different effects on VP- and OTsecreting neurons in PVN and the effects of NE are likely to be direct ones. Different mechanisms may be involved in mediating the effects of NE on these magnocellular neurons. [1] Armstrong-James, M., Fox, K., Kruk, Z.L. and Millar, J., Quantitative iontophoresis of catecholamines using multibarrel carbon fiber electrodes, J. Neurosci. Methods, 4 (1981) 385–406. [2] Cunningham, E.T. and Saychenko, P.E., Reflex control of magnocellular vasopressin and oxytocin secretion, Trends Neurosci., 14 (1991) 406–411. [3] Day, T.A., Furguson, A.V. and Renaud, L.P., Facilitatory influence of noradrenergic afferents on the excitability of rat paraventricular nucleus neurosecretory cells, J. Physiol., 355 (1984) 237–249. [4] Day, T.A., Renaud, L.P. and Sibbald, J.R., Excitation of supraoptic vasopressin cells by stimulation of A1 noradrenaline cell group: failure to demonstrate role for established adrenergic on amino acid receptors, Brain Res., 516 (1990) 91–98. [5] Ji, Y.P., Jun, M. and Tang, B., Effects of intracerebralventricular administration of norepinephrine on paraventricular neurosecretory cells, Acta. Physiol. Sinica (in Chinese), 47 (1995) 120–126. [6] McNeill, T.H. and Sladek, J.R. Jr., Simultaneous monamine histofluorescence and neuropeptide immunocytochemistry. Correlative distribution of catecholamine varicosities and magnocellular neurosecretory neurons in the rat supraoptic and paraventricular nuclei, J. Comp. Neurol., 193 (1980) 1023– 1033. [7] Moss, R.L., Dyball, R.E.J. and Cross, B.A., Responses of antidromically identified supraoptic and paraventricular units to acetylcholine noradrenaline and glutamate applied iontophoretically, Brain Res., 35 (1971) 573–575. [8] Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, 2nd edn., Academic Press, Sydney, 1986. [9] Poulain, D.A. and Wakerley, J.B., Electrophysiology of hypothalamic magnocellular neurons secreting oxytocin and vasopressin, Neuroscience, 7 (1982) 773–808. [10] Sawchenko, P.E. and Swanson, L.W., The organization of noradrenergic pathways from the brainstem to the paraventricular and supraoptic nuclei in the rat, Brain Res. Rev., 4 (1993) 275– 325. [11] Sibbald, J.R., Wilson, B.K.J. and Day, T.A., Neuropeptide Y potentiates excitation of supraoptic neurosecretory cells by noradrenaline, Brain Res., 499 (1989) 164–168. [12] Wakerley, J.B. and Lincoln. D.W., Intermittent release of oxytocin during suckling in the rat, Nature New Biol., 233 (1971) 180–181. [13] Wakerley, J.B. and Lincoln, D.W., The milk ejection in the rat: a 20–40 fold acceleration in the firing of paraventricular neurons during oxytocin release, J. Endocr., 57 (1973) 477–493. [14] Yagi, K., Azuma, T. and Matsuda, K., Neurosecretory cell: capable of conducting impulse in rats, Science, 154 (1966) 778– 779.