Based on previous studies we postulated that, whereas the natriuresis observed following intracerebroventricular (i.c.v.) moxonidine was mediated by a decrease in renal sympathetic nerve activity, the natriuresis observed following intrarenal (i.r.) infusion of moxonidine was mediated by a direct stimulation of renal I1-imidazoline preferring receptors. Sprague-Dawley rats were unilaterally nephrectomized and i.c.v. cannulated 7 to 10 days and 3 days prior to the day of the experiment, respectively. On the day of the experiment, rats were anesthetized (pentobarbital) and the renal function was isolated. Administration of i.c.v. as well as i.r. moxonidine produced an increase in sodium excretion and urine flow fate. Pretreatment with intravenous prazosin (0.15 mg/kg) completely attenuated the response to i.c.v. moxonidine (1 nmol/5 microliters) but only slightly altered the response to i.r. moxonidine (3 nmol/kg/min). Conversely, intravenous pretreatment with the imidazoline preferring receptor antagonist idazoxan (0.3 mg/kg) completely blocked the response to i.r. moxonidine (3 nmol/kg/min) without altering the response to i.c.v. moxonidine (0.3 nmol/kg). These results would be consistent with the natriuresis observed following i.c.v. moxonidine as being mediated by imidazoline preferring receptors located centrally, whereas that following i.r. moxonidine was mediated directly by renal imidazoline-preferring receptors, with a small component of this response conceivably due to activation of central imidazoline preferring receptors. In summary, the antihypertensive effect of imidazoline preferring receptor agonists may be associated with a natriuresis that is due to stimulation of these receptors, found both peripherally (renal) and centrally.