We have recently shown that hydrolysis of labeled angiotensin I in canine brainstem homogenate causes a rapid accumulation of the heptapeptide angiotensin-(1-7) [Ang-(1-7)]. Although this angiotensin fragment has no vasopressor activity, its consistent generation in brain homogenate led us to study its potential neurosecretory effects in the rat hypothalamo-neurohypophysial system (HNS) in vitro. Ang-(1-7) or angiotensin II (Ang II) was added to HNS perifusate in concentrations of 0.04, 0.4, and 4 microM, and release of arginine vasopressin (AVP) during each treatment was quantified as a percentage of the AVP release detected in the preceding collection period. Base-line release of AVP averaged 281 +/- 47 pg per 15 min (mean +/- SEM) in HNS explants (five experiments, five explants per chamber) perifused in Krebs solution at 37 degrees C, after a 1-hr equilibration period. At 0.04 microM, Ang II or Ang-(1-7) did not stimulate AVP release. Ang II increased AVP release over the control value by 172% +/- 44% and 268% +/- 66% at 0.4 and 4 microM, respectively; the same concentrations of Ang-(1-7) increased AVP release by 134% +/- 12% and 216% +/- 45%. The responses to Ang II and Ang-(1-7) at the highest concentration were both significant (P less than 0.05), and comparison by two-way analysis of variance indicated that Ang II and Ang-(1-7) were equipotent in stimulating AVP release over the range of concentrations studied. In the presence of the competitive Ang II antagonist [Sar1,Thr8]Ang II (20 microM), the release of AVP increased approximately equal to 2-fold. Neither Ang II nor Ang-(1-7) (4 microM) caused a further enhancement of AVP release in the presence of [Sar1,Thr8]Ang II. These data suggest that a hydrophobic residue in position 8 of the angiotensin peptide is not essential for activation of angiotensin receptors in the rat HNS. Moreover, the equipotence of Ang II and Ang-(1-7) indicates that Ang-(1-7) may participate in the control of AVP release.