Objective: Arginine vasopressin is synthesized in the hypothalamus and secreted by the posterior pituitary gland in response to plasma hypertonicity. Previous studies suggest that in utero and neonatal exposure of rat pups to prolonged alterations of plasma osmolality may permanently alter (imprint) arginine vasopressin synthesis and secretion, thus adult responses to osmotic challenges. Little is known, however, of the potential for imprinting of neuroendocrinologic systems in precocial species. In view of the frequent occurrence of altered maternal and fetal plasma tonicity (eg, maternal dehydration, hyperemesis), we sought to determine the effect of prolonged maternal hypertonicity on arginine vasopressin gene expression and pituitary gland content in neonatal sheep. Study Design: Pregnant ewes at 119 ± 3 days of gestation were water restricted to achieve and maintain plasma hypertonicity (10-20 mOsm/kg above baseline level) until normal term delivery. Newborns were provided maternal nursing ad libitum. Within 24 hours after birth, study neonatal lambs (n = 6) and age-matched control neonatal lambs (n = 5) were killed, and the pituitary gland and hypothalamus were removed and frozen immediately. Pituitary arginine vasopressin content was determined by radioimmunoassay, and hypothalamic arginine vasopressin gene expression was quantified with Northern blot. Differences in pituitary arginine vasopressin content and hypothalamic arginine vasopressin gene expression (arginine vasopressin/ β-actin ratio) between study and control newborn lambs were analyzed by unpaired t test. Results: In response to maternal water restriction, maternal plasma osmolality increased from 307 ± 0.9 mOsm/kg to 325 ± 1.3 mOsm/kg, and plasma sodium increased from 147 ± 1.3 mEq/L to 156 ± 1.2 mEq/L. The maternal plasma hyperosmolality and hypernatremia were maintained until normal term delivery. At the time of death, study (in utero dehydrated) lambs had higher plasma sodium (150 ± 0.4 mEq/L vs 146.5 ± 1.5 mEq/L; P <.05) and chloride (112.8 ± 1.0 mEq/L vs 108.5 ± 1.5 mEq/L; P <.05) levels, and lower potassium (4.5 ± 0.2 mEq/L vs 5.5 ± 0.3 mEq/L; P <.05) concentrations than control newborn lambs. Both newborn groups had similar plasma osmolality (320.0 ± 1.3 mOsm/kg vs 318.0 ± 3.4 mOsm/kg). Total pituitary arginine vasopressin content was significantly greater in the study than in the control newborn lambs (8.3 ± 2.8 μg vs 1.6 ± 1.3 μg; P <.05). Conversely, hypothalamic arginine vasopressin messenger RNA levels were lower in the study newborn lambs than in the control newborn lambs (arginine vasopressin/β-actin ratio: 0.29 ± 0.01 vs 0.68 ± 0.15; P <.05). Conclusion: Despite the presence of plasma hypernatremia, prolonged elevation of fetal plasma tonicity increases newborn pituitary arginine vasopressin content yet decreases hypothalamic arginine vasopressin gene expression. The present study suggests that prolonged prenatal exposure to plasma hypertonicity may imprint the hypothalamic-pituitary arginine vasopressin regulatory system. (Am J Obstet Gynecol 2002;187:191-6.)