Prolonged effects of polyriboinosinic:polyribocytidylic acid on spontaneous running wheel activity and brain interferon-alpha mRNA in rats: a model for immunologically induced fatigue.

Following 2 weeks acclimation to the running wheel in the home cages, an i.p. injection of a synthetic double-stranded RNA, polyriboinosinic:polyribocytidylic acid (poly I:C, 3 mg/kg), was performed to produce the immunologically induced fatigue in rats. The daily amounts of spontaneous running wheel activity decreased to about 40-60% of the preinjection level until day 9 with normal circadian rhythm, then gradually returned to the baseline level by day 14. Rats given a heat exposure (36 degrees C for 1 h) for the consecutive 3 days showed an increase in activity except for the first day. In the open field test, the total moving distance and the number of rearing of the poly I:C-injected rats decreased on day 1, but they were not different from the saline-injected group on day 7, suggesting that the poly I:C-induced fatigue on day 7 was not due to the peripheral problems such as muscle/joint pain, but involved the CNS. Quantitative analysis of mRNA levels using a real-time capillary reverse transcriptase-polymerase chain reaction (RT-PCR) method revealed that interferon-alpha (IFN-alpha) mRNA contents in the cortex, hippocampus, hypothalamic medial preoptic, paraventricular, and ventromedial nuclei were higher in the poly I:C group than those in the saline and heat-exposed groups on day 7, although the amount of interleukin-1 beta mRNA showed no differences. Serum adrenocorticotropic hormone and catecholamine levels were not significantly different between groups. The present results indicate that the prolonged fatigue induced by poly I:C, which is evaluated by the spontaneous running wheel activity, can be used as an animal model for the immunologically induced fatigue associated with viral infection, and suggest that brain IFN-alpha may play a role in this model.

Neuroimmunomodulatory actions of hypothalamic interferon-alpha.

Recent studies have revealed that the brain produces interferon-alpha (IFN-alpha) in response to noninflammatory as well as inflammatory stress and that it might have a role in normal physiology. When administered intracerebrally, IFN-alpha causes diverse effects including fever, anorexia, analgesia and changes in the central neuronal activities. These responses are inhibited by the opioid receptor antagonist naloxone. This is consistent with the reports suggesting that recombinant human (rh) IFN-alpha binds to opioid receptors in rodent brain membrane. We revealed that rhIFN-alpha altered the activity of thermosensitive neurons in the medial preoptic area (MPO) and glucose-responsive neurons in the ventromedial hypothalamus in an opioid-receptor-dependent way. As a stress which produces opioid-dependent analgesia is known to suppress the cytotoxicity of splenic natural killer cells, we investigated whether the administration of beta-endorphin and rhIFN-alpha may induce a similar immunosuppression. We found that central, but not peripheral, injection of both compounds inhibited natural killer (NK) cytotoxicity. Further studies revealed that rhIFN-alpha decreased the activity of MPO neurons via opioid receptors and the altered activity of MPO neurons in turn resulted in the activation of corticotropin-releasing factor neurons, thereby suppressing NK cytotoxicity predominantly through activation of the splenic sympathetic nerve and beta-receptor mechanisms in splenocytes. Thus, IFN-alpha may alter the brain activity to exert a feedback effect on the immune system. Further detailed whole-cell clamping analyses on neuronal mechanisms in rat brain tissue slices showed that the inhibitory effect of rhIFN-alpha on N-methyl-D-aspartate-induced membrane current responses of MPO neurons was mediated not only by opioid receptors but also by the local production of reactive oxygen intermediates, nitric oxide and prostanoids, possibly due to neuron-glial cell interaction.

Cerebellar afferents to neuroendocrine cells: implications for adaptive responses to simulated weightlessness.

The hypothalamo-neurohypophyseal system as well as the autonomic nervous system is involved in homeostatic responses associated with changes in head position and orthostatic reflex. The responses induced by body tilt on earth are thought to be attributed to changes in inputs from baroreceptors, vestibular organs and proprioreceptors that are normally required for postural control. The information from these organs is sent to the hypothalamus which thereby influences both neuroendocrine and autonomic systems as well as various kinds of emotional behavior. Our findings showing the fastigial input to the hypothalamus suggested that the FN plays a significant role in these homeostatic responses through its connections with the brain stem and the hypothalamus. Figure 4 shows the input-output organization among the hypothalamus, cerebellum and brain stem, described in detail in sections III to V. This hypothesis may help to account for the autonomic and endocrine disorders often observed in weightlessness.

The hypothalamo-sympathetic nervous system modulates peripheral cellular immunity.

Our findings reviewed in this article have revealed that the stimulation of opioid receptors of the hypothalamic neurons by interferon alpha and beta-endorphin synthesized in the brain or by stress causing the opioid-dependent analgesia suppresses the natural killer cytotoxicity, an important component of immunosurveillance, through an activation of the hypothalamic CRF-sympathetic nervous system.

Hypothalamic modulation of splenic natural killer cell activity in rats.

1. The cytotoxic activity of splenic natural killer cells measured by a standard chromium release assay in urethane and alpha-chloralose-anaesthetized rats was significantly suppressed 20 min after bilateral ablation of the medial part of the preoptic hypothalamus (MPO). The suppression was completely blocked by prior splenic denervation. The splenic natural killer cell activity of MPO sham-lesioned rats or thalamus-lesioned rats, both having an intact splenic innervation, were not different from that of a non-treated control group. 2. Electrical stimulation of the bilateral MPO (0.1 ms, 0.1-0.3 mA, 5-100 Hz) suppressed the efferent activity of the splenic nerve in all six rats examined. The reduction of the nerve activity was accompanied by a transient fall in blood pressure. An I.V. injection of phenylephrine (3 micrograms/0.3 ml) also evoked a suppression of the nerve activity, which was accompanied by transient hypertension, suggesting that the suppressive effect of the MPO stimulation was independent of changes in blood pressure. On the other hand, a bilateral lesion of the MPO resulted in a sustained increase in the electrical activity of the splenic sympathetic nerve filaments which lasted for more than 2 h. 3. Microinjection of monosodium-L-glutamate (0.1 and 0.01 M in 0.1 microliters saline) unilaterally into the MPO evoked a transient suppression of the efferent discharge rate of the splenic nerve activity within 1 min, which was also accompanied by a decrease in blood pressure. The injection of saline (0.1 microliter) into the MPO had no effect. The microinjection of recombinant human interferon-alpha (200 and 2000 U in 0.1 microliter saline) into the MPO dose dependently increased the splenic nerve activity without any change in blood pressure. 4. In contrast, microinjection of interferon-alpha into the paraventricular nucleus of the hypothalamus (PVN) had no effect on splenic nerve activity, although an injection of glutamate increased the nerve activity. 5. The present results, taken together with previous reports, suggest that the neuronal networks between the MPO and the splenic sympathetic nerve, which may be activated by centrally administered interferon-alpha, are important in the suppression of the splenic cellular immunity.

Effects of temperature and neuroactive substances on hypothalamic neurones in vitro: possible implications for the induction of fever.

This paper reviews some of our findings which have shown the usefulness of in vitro methods in the study of hypothalamic neurones. (1) Membrane current analyses of dispersed neurones of the rat preoptic and anterior hypothalamus (POA) during thermal stimulation have revealed that warm-sensitive neurones are endowed with a non-inactivating Na+ channel having a high Q10 in the hyperthermic range (35-41 degrees C). (2) A brain slice study has shown that neurones in the organum vasculosum lamina terminalis (OVLT) region have much higher sensitivity to PGE2 than POA neurones. This provides further evidence of a critical role of the OVLT in translation of blood-borne cytokine signals into brain signals for fever induction. (3) Local application of IL-1 beta and IFN alpha altered the activity of thermosensitive (TS) neurones and glucose responsive (GR) neurones in vitro in an appropriate way to produce fever and anorexia. While the responses to IL-1 beta required the local release of prostaglandins, the responses to IFN alpha were found to be mediated by opioid receptor mechanisms. (4) The responses of POA TS neurones and VMH GR neurones to IL-1 beta but not those to IFN alpha, were reversibly blocked by alpha MSH, an endogenous antipyretic peptide. Thus, immune cytokines and their related neuroactive substances may affect hypothalamic TS and GR neurones thereby producing elaborately regulated changes in homeostatic functions such as thermoregulation (fever) and feeding (anorexia), which are considered as host defence responses.

Immune cytokines and regulation of body temperature, food intake and cellular immunity.

Interleukin-1 (IL-1) and interferon alpha (IFN alpha), cytokines originally detected in immunological cells, now have been shown to produce nonimmunological host defense responses of central and peripheral origins. These cytokines are released from glial cells in the brain in pathological states. Local application of IL-1 beta and IFN alpha to thermosensitive neurons in the preoptic and anterior hypothalamus and glucose responsive neurons in the ventromedial hypothalamus in vivo and in vitro, altered the activity in appropriate ways to explain the cytokines-induced fever and anorexia, respectively. The responses to IL-1 beta, but not to IFN alpha, were blocked by sodium salicylate, suggesting the involvement of synthesis of prostaglandins. alpha MSH, an endogenous antipyretic and a possible antagonist of IL-1 beta at lymphocytes, specifically depressed the responses to IL-1 beta, but not those to IFN alpha. In contrast, the action of IFN alpha was reversibly blocked by naloxone, suggesting the opioid receptor mediation. Intracerebral injection of IFN alpha and beta-endorphin in the rat and mouse resulted in the suppression of cytotoxic activity of natural killer cells in the spleen by activation of brain opioid receptor, which was shown to be mediated predominantly by splenic sympathetic nerves. The results suggest a view that immune cytokines may provide afferent links for the regulatory circuits between the brain and the immune system.

Characterization of opioid-sensitive neurons in the anteroventral third ventricle region of polydipsic inbred mice in vitro.

In previous studies we found that in the extremely polydipsic special strain of mice, STR/N, spontaneous drinking was greatly attenuated by injection of the opioid antagonists given intracerebroventricularly as well as subcutaneously. Therefore, we investigated, using hypothalamic slice preparations, responses of neurons in the anteroventral third ventricle region (AV3V) of the STR/N and its control, Swiss/Webster (S/W) mice to morphine and opiate peptides. An application of morphine at 10(-6) M to the circulating medium inhibited activities of 44% of AV3V neurons (45 of 102) in the STR/N, and 59% (76/129) in the S/W, demonstrating that morphine affected a smaller proportion of neurons of the polydipsic mice than that of controls. Opioid agonists for 3 receptor types, mu, delta and kappa, at 10(-6) to 10(-5) M inhibited AV3V neurons in both the STR/N and S/W mice, but to a different degree. No cell of either strain was excited by morphine or any of the opioids. The mu-receptor agonist, [D-Ala2,N-Me-Phe4,Gly5-ol]-enkephalin (DAGO), was the most potent inhibitor of AV3V neurons; in the STR/N 53% (25/47), and in the S/W 77% (34/44) were inhibited. The kappa-agonist, dynorphin A-(1-13) (DYN), inhibited fewer cells in the STR/N (9%, 4/47), compared with the S/W (36%, 16/44). Only a few cells responded to the delta-agonist, [D-Pen2,5]enkephalin (DPDPE), in both strains. The inhibitory actions of the opiates were reversibly blocked by naloxone, and persisted under synaptic blockade. The threshold concentration of morphine or DAGO for inhibition of AV3V neurons was higher in the STR/N (approximately 10(-8) M for both morphine and DAGO) than in S/W mice (approximately 10(-9) M for morphine and less than 10(-9) M for DAGO). Although the AV3V also contains angiotensin II-sensitive neurons, they were not affected by morphine (10(-6) M). Similarly neurons inhibited by morphine were not excited by angiotensin II (10(-7) M); some neurons were unresponsive to both chemicals. We conclude that morphine and opiate peptides directly inhibit the AV3V neurons of both the STR/N and S/W strains of mice and the sensitivity of these neurons to the opiates is lower in the polydipsic inbred mice compared to their controls. The results, together with our behavioral studies, suggest involvement of the central opioid system in the polydipsia of the STR/N mice.

Effects of hypothalamic stimulation and lesion on adrenal nerve activity.

Activity changes of efferent adrenal sympathetic nerves in response to bilateral manipulations of the hypothalamus, partly after intra-third cerebroventricular injection of 2-deoxy-D-glucose (2-DG) were investigated in anesthetized rats. Stimulation of the middle part of the lateral hypothalamic area (LHAm) increased adrenal nerve activity, whereas lesion caused rapid and remarkable decrease. Stimulation of the anterior part of the LHA (LHAa) tended to decrease the activity, and lesion produced either rapid decrease or late moderate increase. Stimulation of the ventromedial hypothalamic nucleus (VMH) did not affect the nerve activity, but lesion increased it gradually and then remarkably. Cerebroventricular infusion of 2-DG caused remarkable increase in activity that was suppressed by LHAm lesion. Subsequent infusion of 2-DG during the period of suppressed activity was no longer effective. The increased firing rate after 2-DG was suppressed by stimulation of the VMH, whereas lesion caused no change. These findings indicate that the central regulation of adrenal nerve activity is connected with individual hypothalamic regions and consequently depends on the degree and mode of activation of the sympathoadrenal system.

Hypothalamus and sodium-potassium pump activity in skeletal muscles of DOCA-hypertensive rats.

The effects of hypothalamic lesions on Na+ and K+ content ([Na+]i and [K+]i) in both slow tonic muscle [soleus (SOL)] and fast-twitch muscle [extensor digitorum longus (EDL)] were investigated in deoxycorticosterone acetate (DOCA)-salt hypertensive rats. In DOCA-treated rats, [Na+]i was increased and [K+]i decreased in both SOL and EDL muscles compared with controls. Cellular K+ loss and Na+ accumulation in SOL were restored after tibial nerve sectioning (delta [Na+]i, -14.2 +/- 2.6 and delta [K+]i, 13.9 +/- 2.6 mmol/l fiber water (FW), n = 18, P less than 0.01) or bilateral lesioning of the ventromedial hypothalamic nucleus (VMH) (delta [Na+]i, -15.5 +/- 1.5 and delta [K+]i, 17.4 +/- 4.6 mmol/l FW, n = 6, P less than 0.01 and P less than 0.05). On the other hand, the anomalous electrolyte content in EDL was counteracted by lesions of anteroventral portion of the third ventricle (AV3V) (delta [Na+]i, -8.8 +/- 1.6 and delta [K+]i, 5.2 +/- 1.2 mmol/l FW, n = 6, P less than 0.01 and P less than 0.05) or paraventricular hypothalamic nucleus (PVN) (delta [Na+]i, -6.8 +/- 0.6 and 5.7 +/- 1.2 mmol/l FW, n = 6, P less than 0.01 and P less than 0.05), but aggravated by denervation (delta [Na+]i, 13.4 +/- 1.8 and delta [K+]i, -9.6 +/- 1.8 mmol/l FW, n = 18, P less than 0.01). These results suggest that there are at least two hypothalamic mechanisms of suppression of muscle Na-K pump activity in DOCA-hypertensive rats; i.e., neurally mediated inhibition in SOL by the VMH and, presumably, humorally mediated inhibition originating from the AV3V or PVN with greater influence in EDL.

Hypothalamic regulation of sodium-potassium pump activity in skeletal muscle.

The suppression of active Na+-K+ transport in rat skeletal muscle during hypokalemia was counteracted by bilateral electrolytic lesion of the ventromedial hypothalamic nucleus. This reversal effect was unaffected even after pancreatectomy or adrenalectomy. The anomalous electrolyte content in hypokalemic rat muscles was aggravated by lesion of the dorsomedial hypothalamic nucleus and of the anterior hypothalamus. The results indicate that the hypothalamus is involved in the regulation of the Na+-K+ transport system in skeletal muscle during hypokalemia.

Effects of hypothalamic lesions on active sodium-potassium transport in the extensor digitorum longus muscles of hypokalemic rat.

The CNS-induced suppression on muscle Na+-K+ pump was studied in "twitch" muscle, extensor digitorum longus (EDL), of hypokalemic rats which were fed a K+ deficient diet for several weeks. Peripheral nerve section or bilateral lesion of the ventromedial hypothalamic nucleus had no effect on the Na+ and K+ contents in EDL of hypokalemic rats. However, lesions of the paraventricular nucleus caused the net Na+ loss and the net K+ uptake in the muscles. Lesions in either the dorsomedial nucleus or anterior hypothalamus also caused significant net K+ uptake but the net Na+ loss was not significant. The results were compared with those of "tonic" muscle, soleus, reported previously.

Effects of hypothalamic lesion on pancreatic autonomic nerve activity in the rat.

The effects of hypothalamic lesions and intravenous glucose infusion on the efferent activity of vagal and splanchnic nerves to the pancreas were studied in anesthetized rats. Lesions of the ventromedial hypothalamic (VMH), the dorsomedial hypothalamic (DMH) and the paraventricular (PVN) nuclei increased vagal and reduced splanchnic nerve activity. Lesion of the lateral hypothalamic area (LHA) decreased pancreatic vagal nerve activity, and produced either increased or decreased activity of pancreatic splanchnic nerve. Intravenous glucose infusion increased activity of the vagal nerve and reduced that of the splanchnic nerve. These glucose responses were influenced by hypothalamic lesions only slightly or not at all. The findings suggest that hypothalamic modulation of pancreatic hormone secretion involves both the parasympathetic and sympathetic nervous systems, and provide evidence that not only the VMH and the LHA but also the DMH and the PVN are involved in this mechanism.