Salt and water intake in Brattleboro rats with hypothalamic diabetes insipidus.

Brattleboro rats lacking vasopressin have an elevated plasma osmolality and a stimulated renin-angiotensin system relative to Long-Evans rats (LE). The current studies were performed to elucidate the factors controlling water and salt intake in the Brattleboro rat with diabetes insipidus (DI). DI and LE rats were given the choice of water and saline solutions ranging from 0.1-1.0% to assess palatability, dialyzed with isotonic glucose to test for sodium appetite after sodium depletion, and infused intracranially with an angiotensin II analogue (saralasin) to assess the role of angiotensin II in spontaneous salt and water intake. DI rats exhibited spontaneous salt intake which was not significantly different from LE rats and increased their intake of 3.0% NaCl following sodium depletion, although less reliably than LE rats. A significant proportion of those DI rats not developing a sodium appetite showed attenuation of their diabetes following dialysis. No evidence for involvement of angiotensin II in spontaneous salt and water intake was found.

Increased water excretion in hyperprolactinemic rats.

The role of PRL in mammalian salt and water balance remains controversial. To avoid the methodological problems of exogenous PRL administration, endogenous hyperprolactinemia was established in normal rats by implantation of extra anterior pituitary glands under the kidney capsule. To eliminate excess glucocorticoid secretion in these rats, they were adrenalectomized, implanted with corticosterone replacement pellets, and given 0.9% saline drinking fluid. A highly significant increase in urine flow was found in pituitary-grafted male Fischer rats compared to control rats (38.9 +/- 3.8 vs. 20.7 +/- 1.6 ml/24 h X 100 g BW; P less than 0.0005). Urine osmolality was lower in pituitary-grafted rats, but sodium and potassium excretion were not abnormal. Although fluid intake was also greater in hyperprolactinemic rats, the urine flow remained elevated when adjusted for fluid intake. Renal binding sites for radioactive PRL were not decreased in pituitary-grafted rats. Thus, the hyperprolactinemic rat has increased water excretion that may be attributed to a direct renal effect rather than to glucocorticoid excess or a primary change in thirst. Since it is possible that water absorption in the gut is also increased by PRL, multiple effects of PRL may be responsible for the diuresis observed in hyperprolactinemic rats.

Altered catecholamine innervation of the supraoptic nucleus in the nephrogenic diabetes insipidus mouse.

Fluorescence histochemical and immunocytochemical techniques were used to investigate morphologic correlates of the relationship between catecholamine varicosities and vasopressin-containing perikarya in an animal model of vasopressin excess, the nephrogenic diabetes insipidus mouse. Our results show hypertrophy and increased immunoreactivity in vasopressin neurons in these mice were accompanied by a marked increase in the density and to some extent the fluorescence intensity of catecholamine varicosities within the supraoptic nucleus. These results further support the concept of functional interactions between catecholamine and vasopressin neurons and raise the possibility that the target neuron, or one of its products, perhaps vasopressin, either exerts a trophic influence on the catecholamine innervation pattern of the supraoptic nucleus or enhances catecholamine content in existing fibers and terminals.

Identification of medullary loci critical for neurogenesis of gasping.

We hypothesized that a discrete medullary locus, critical for gasping neurogenesis, could be identified. In decerebrate, cerebellectomized, vagotomized, paralyzed, and ventilated cats, activities of phrenic, hypoglossal, and recurrent laryngeal nerves were monitored. Gasping was induced by freezing the brain stem, via a fork thermode, at the pontomedullary junction. By reversible cooling of the medulla, chemical lesions with kainic acid, and radio-frequency lesions, a critical area for gasping neurogenesis was localized bilaterally 2-3 mm rostral to obex, 2.0-2.5 mm lateral to midline, and 3-4 mm ventral to medullary surface. Electrical stimulation in this area elicited premature gasps, whereas unilateral lesions or lidocaine injections eliminated gasping activities in all nerves. These procedures did not cause similar changes during eupnea. In apneusis, however, lidocaine injections markedly altered the pattern or caused apnea. We conclude that discharge of neurons in a discrete portion of the lateral tegmental field of medulla is required for gasping neurogenesis. Our results are consistent with these neurons comprising the central pattern generator for gasping.

Studies of anterior pituitary-grafted rats: I. Abnormal prolactin response to thyrotropin releasing hormone, clonidine, insulin, and fasting.

Although the rat implanted with extra anterior pituitary glands (AP) under the kidney capsule has been widely used as a model of chronic hyperprolactinemia, its hormonal status has not been fully characterized. Using conscious, unrestrained female pituitary-grafted rats and sham-operated littermates, we investigated prolactin (PRL) secretion in response to the following stimuli: thyrotropin releasing hormone (TRH), clonidine, insulin, and fasting. The AP-implanted rats had a greater and more sustained rise in serum PRL after TRH than control rats, reflecting a direct effect of TRH on the ectopic lactotropes. In contrast after clonidine, which acts via the hypothalamus, the serum PRL rose to much higher levels in sham-operated rats than in rats bearing ectopic pituitary tissue. Both insulin-induced hypoglycemia and fasting decreased serum PRL in control rats, but the AP-implanted animals manifested a rise in serum PRL in response to these stimuli. Thus, the AP-implanted rat is not only a valid model of excess and abnormal PRL secretion, but it may also be useful for distinguishing between stimuli requiring an intact hypothalamic-pituitary unit and agents which act directly on the pituitary gland.

Studies of anterior pituitary-grafted rats: II. Normal growth hormone secretion.

Of the various animal models used to study chronic hyperprolactinemia, the otherwise intact rat implanted with extra anterior pituitary glands (AP) under the kidney capsule is assumed to be normal except for excess circulating prolactin (PRL). Since the ectopic glands contain numerous somatotropes in addition to abundant and active lactotropes, it was important to assess growth hormone (GH) secretion as well in this model of hyperprolactinemia. The structural and functional similarities of PRL and GH are such that it is necessary to demonstrate that metabolic abnormalities noted in AP-implanted rats are due to hyperprolactinemia and not to altered GH secretion. AP-implanted female rats have significantly higher resting serum PRL concentrations when compared to sham-operated control rats, but baseline serum GH levels are similar in normal and pituitary-grafted rats. Suppression of GH by insulin and clonidine is comparable in AP-implanted and control rats. The intrasellar pituitary GH concentration is also similar (ca. 20 micrograms/mg wet weight) in hyperprolactinemic and normal rats. We conclude that GH secretion is normal in the non-hypophysectomized AP-implanted rat, in contrast to the hypophysectomized AP-implanted rat model which has been reported to have diminished GH secretion. Despite the presence of recognizable somatotropes, the ectopic anterior pituitary does not appear to secrete significant amounts of GH, making the intact rat bearing multiple pituitary grafts an excellent model of chronic hyperprolactinemia.

Glucose tolerance in rats with elevated circulating prolactin levels.

The consideration of prolactin as one of the anti-insulin hormones has been based on studies describing the ability of injected prolactin to raise the serum glucose. We studied the glucose tolerance of rats made hyperprolactinemic by the implantation of extra anterior pituitary glands under the kidney capsule. Surprisingly the hyperprolactinemic rats had lower serum glucose levels after a bolus injection of dextrose. Furthermore, their serum insulin levels were no higher than sham-operated control rats. AFter a fast of 18 or 24 hours, the serum glucose was lower in the hyperprolactinemic rats, and the serum insulin concentrations were similar in implanted and control rats. These data suggest that chronic hyperprolactinemia does not decrease glucose tolerance. Whether prolactin has a direct insulin-like effect or whether it may potentiate insulin's effects will require further study.

Preparation of a biologically active rat prolactin devoid of antidiuretic activity.

Rat prolactin (NIH rPRL-B2) was purified using Sephadex chromatography and isoelectric focusing. SDS-gel electrophoresis of the original material showed a major band of 23K daltons, as well as several minor bands; the purified prolactin had a single 23K band. The original material contained 33 pg of immunoreactive vasopressin per 0.1 micrograms of material; vasopressin was not detectable in the purified material by either RIA or bioassay. The purified preparation had complete biological activity in a mammary gland bioassay and a receptor binding assay.

Demonstration that rat prolactin has no intrinsic antidiuretic activity in the rat.

The proposition that rat prolactin has an intrinsic antidiuretic activity was examined using a conscious rat model (Brattleboro homozygotes, DI rats) which does not appear to produce arginine vasopressin (AVP). A preparation of rat prolactin obtained from NIAMDD, rat prolactin B1, produced a prompt antidiuresis when administered intravenously to these animals. Prolactin B1 was found to contain 150 ng of AVP/mg protein by RIA. All of the measurable AVP was removed from prolactin B1 by polyacrylamide gel electrophoresis and the electrophoresed prolactin was completely devoid of any antidiuretic activity. Addition of synthetic AVP to electrophoresed prolactin in an amount equivalent to the AVP contamination of prolactin B1 completely restored the originally observed antidiuretic activity. Synthetic AVP given alone produced a similar antidiuresis, but in the presence of electrophoresed prolactin these effects were produced by much smaller amounts of AVP. We believe that our data demonstrate that rat prolactin has no intrinsic antidiuretic activity, and that the antidiuretic activity associated with different prolactin preparations could be entirely due to their contamination with AVP. The data also show that the presence of prolactin enhanced by about five times the antidiuretic activity of AVP.

The effects of elevated circulating prolactin in rats with hereditary hypothalamic diabetes insipidus (Brattleboro strain).

Administration of ovine or rat PRL to animals, including man, has resulted in decreased urine volume and increased urine osmolality. Contamination of PRL preparations with vasopressin is the most likely explanation for the apparent antidiuretic effect. In this study, diabetes insipidus rats lacking vasopressin(homozygous Brattleboro rats) had extra anterior pituitary glands implanted under the kidney capsule, resulting in hyperprolactinemia. The urine of such rats was not more concentrated than that of unoperated littermates or sham-operated littermates with diabetes insipidus. In fact, hyperprolactinemic male rats produced even less concentrated urine than control rats. Furthermore, the hyperprolactinemic rats responded to exogenous vasopressin in a manner similar to normoprolactinemic rats. These studies provide strong evidence against an antidiuretic action of PRL in mammals.

The hypothalamic-neurohypophysial system of the rat: localization and quantitation of neurophysin by light microscopic immunocytochemistry in normal rats and in Brattleboro rats deficient in vasopressin and a neurophysin.

The cellular distribution of neurophysin was examined in hypothalami and neural lobes of normal Long-Evans rats and Brattleboro rats deficient in vasopressin and a major neurophysin. Tissue sections were treated with antisera to bovine, human, and rat neurophysins, using immunoperoxidase bridge techniques. Antisera to oxytocin (OT) and vasopressin (VP) were applied to adjacent sections. Two distinct cell populations were discernible in both magnocellular nuclei on the basis of the intensity of cytoplasmic staining. About half of the magnocellular neurons in the supraoptic (SON) and paraventricular (PVN) nuclei of homozygous Brattleboro rats with diabetes insipidus (DI) were devoid of immunoreactive neurophysin, OT, and VP. These cells were presumably the defective counterparts of those neurons that produce VP and its associated neurophysin in normal and heterozygous Brattleboro rats. The cells in homozygous DI rats which were stained with immunoreaction products to NP and OT were more concentrated in the dorsal part of the SON and in the periphery of the PVN. Spatial segregation of different neurons was also seen in the neural lobe, where clusters of stained axons were surrounded by bundles of nerve fibers lacking immunoreactive material. In normal rats and heterozygotes nearly all magnocellular neurons reacted immunologically with antiserum to neurophysin but with different intensities, so that "dark" and "light" cells could be distinguished. The darker cells in heterozygous Brattleboro rats had the same pattern of distribution as cells which contained OT. In homozygous DI rats, only some of those cells which contained neurophysin and OT exhibited a positive reaction with antiserum to VP due to slight reactivity with OT. The results obtained in the homozygous Brattleboro rat would suggest that OT and VP and their associated neurophysins are produced in different neurons in both the SON and PVN. However, in normal rats and in heterozygous Brattleboro rats, VP appeared to be present in both OT-positive and OT-negative neurons suggesting that some cells may have the capacity to synthesize two hormones.