Progressive water deprivation increased plasma osmolality, plasma Na+ concentration, and hematocrit in proportion to the severity of dehydration. With increases of 2% in plasma osmolality (24 h dehydration), glucose utilization increased in the supraoptic nuclei and tended to increase in the neural lobe. With further dehydration, glucose utilization also increased in the paraventricular nuclei. These increases were paralleled by depletion of vasopressin and oxytocin contents in the neural lobe and by the enhanced secretion of both hormones into plasma, with a predominant increase of vasopressin. These changes were proportional to the degree of dehydration. With progression of dehydration, decreases in intracellular and extracellular volumes accentuate. Reductions in extracellular volume result in increased angiotensin II (ANG II) formation. Accordingly, glucose utilization in the subfornical organ (SFO), a primary site of ANG II action, increased after 48 and 72 h of dehydration. The median preoptic nucleus, which receives direct inputs from the SFO, also increased glucose utilization at these times. Glucose utilization also increased in the organum vasculosum laminae terminalis, probably in response to the converging inputs from osmoreceptors, volume receptors, and ANG II receptors. Decreases in glucose utilization were observed in the caudal and rostral ventrolateral medulla, perhaps as compensatory responses to decreased extracellular volume to prevent fall in arterial blood pressure.

Intravenous infusion of hypertonic saline increased plasma [Na (+) ] and osmolality and induced a short-latency drinking response. These changes were associated with increased glucose utilization in the supraoptic and paraventricular nuclei and neural lobe, and decreases in the medial septum and nucleus ambiguus. The increases in glucose utilization were more accentuated in the supraoptic nuclei than in paraventricular nuclei, indicating that they are more sensitive to osmotic stimulation than the paraventricular nuclei. In association with enhanced activity in the hypothalamo-neurohypophysial system, plasma vasopressin and oxytocin concentrations increased, with a preferential increase of oxytocin over vasopressin. The hormonal contents in the neural lobe were not depleted by the osmotic stimulus despite the large increases of their concentrations in the plasma.

Intrajugular infusion (200 microliters/min for 10 min) of 0.85 M NaCl or 1.7 M mannitol in conscious adult male Sprague-Dawley rats increased plasma osmolality similarly and had an additive effect when combined. Plasma Na+ concentration, however, increased with infusion of 0.85 M NaCl, decreased with 1.7 M mannitol, and was not significantly altered by the combined solution. Irrespective of changes in plasma Na+ concentration, plasma vasopressin and oxytocin concentrations were elevated to a similar degree after independent infusion of 0.85 M NaCl or 1.7 M mannitol. With the combined infusion, the change in plasma vasopressin was additive but the change in oxytocin tended to be greater. Accordingly, glucose utilization increased throughout the hypothalamoneurohypophysial system after infusion of 0.85 M NaCl and 1.7 M mannitol. With the combined infusion, however, the change in glucose utilization in the paraventricular nucleus was additive but a synergistic effect occurred in the supraoptic nucleus and neural lobe. Drinking responses were similar in all groups receiving hypertonic solutions, with no additive effect after the combined stimulus. Although our results do not completely rule out the participation of cerebrospinal fluid sodium receptors, it is more likely that osmoreceptors regulate the activity of the hypothalamoneurohypophysial system and drinking behavior. Unlike the magnocellular system, however, drinking behavior seems to be negatively influenced by a stress component of the osmotic stimulation.