Cardiovascular activity of WIN 65579, a novel inhibitor of cyclic GMP phosphodiesterase 5.

This study describes the phosphodiesterase inhibitory potency and cardiovascular actions of WIN 65579 (1-cyclopentyl-3-ethyl-6-(3-ethoxy-4-pyrridyl)-1H-pyrazolo[3,4-d]p yrimidin-4-one), a potent, new cGMP phosphodiesterase 5 inhibitor. WIN 65579 is a competitive inhibitor of phosphodiesterase 5, with IC50 values of 2-3 nM for phosphodiesterase 5 from human or canine vascular sources. WIN 65579 has low affinity for phosphodiesterases 1, 2 and 3 (IC50 > 3-10 microM), and is somewhat selective for phosphodiesterase 4 (IC50 approximately 100 nM). WIN 65579 is an endothelial-dependent relaxant of rat aortic smooth muscle (EC50 = 60 nM) and lowers mean arterial blood pressure in conscious spontaneous hypertensive rats following intravenous or oral dosing. WIN 65579 also increases plasma cGMP levels, and reinstates vascular responsiveness to nitroglycerin in conscious rats that are nitroglycerin-tolerant. These data show that WIN 65579 is one of the more potent phosphodiesterase 5 inhibitors, and that WIN 65579 possesses cardiovascular activities consistent with vascular phosphodiesterase 5 inhibition in vivo.

Species-dependent pharmacodynamic effects of the selective low Km cyclic AMP phosphodiesterase III inhibitors WIN 58993 and WIN 62005.

We describe the biochemical and pharmacologic effects of two novel fused pyridinones derived from milrinone: WIN 58993 and WIN 62005. Both WIN 58993 and WIN 62005 competitively inhibit cyclic GMP-inhibitable low Km cyclic AMP phosphodiesterase (PDE III) from rat heart and canine aorta with Ki values of 25 +/- 3 and 26 +/- 5 nM, respectively, and are selective (at least 160-fold) for PDE III inhibition relative to other PDE isozymes. WIN 58993 and WIN 62005 were given to conscious, chronically instrumented rats and dogs intravenously (i.v.) or perorally (p.o.). Because the doses of WIN 58993 and WIN 62005 required to decrease mean arterial blood pressure (MAP) by 20% were estimated to be 0.9 and 0.7 mg/kg, respectively, the compounds appear to be equipotent after acute i.v. administration in rats. However, the duration of the depressor responses in rats apparently differs since MAP remained significantly decreased 6 h after i.v. or p.o. administration of WIN 58993, but returned to control levels < or = 4 h after administration of WIN 62005. WIN 58993 may be slightly less potent than WIN 62005 after acute i.v. administration to dogs since significant increases in left ventricular (LV)dP/dtmax first occurred at doses of 0.1 and 0.03 mg/kg, respectively. LVdP/dtmax significantly increased in 30 min and returned to baseline 3 h after p.o. administration of 1 mg/kg WIN 58993. After p.o. administration of 1 mg/kg WIN 62005. LVdP/dtmax was significantly increased in 30 min and remained increased for at least 6 h. These data suggest that WIN 58993 and WIN 62005 are potent, selective, p.o.-active inhibitors of PDE III.(ABSTRACT TRUNCATED AT 250 WORDS)

Cyclic GMP potentiation by WIN 58237, a novel cyclic nucleotide phosphodiesterase inhibitor.

The objectives of this study were to determine the potency and selectivity of the structurally novel cyclic nucleotide phosphodiesterase (PDE) inhibitor, WIN 58237 (1-cyclopentyl-3-methyl-6-(4- pyridyl)pyrazolo[3,4-d]pyrimidin-4-(5H)-one), and to determine if this compound possesses cyclic GMP (cGMP) PDE inhibitory activity in vitro and in vivo. WIN 58237 is a competitive inhibitor of cGMP PDE V from canine aorta, with a Ki value of 170 nM. It is a relatively less potent inhibitor of calmodulin-sensitive PDE I and cGMP-inhibitable cyclic AMP PDE III; but does inhibit cyclic AMP PDE IV with an IC50 value of approximately 300 nM. In vitro, WIN 58237 is a functional cGMP PDE inhibitor at submicromolar concentrations as evident by potentiation of both sodium nitroprusside- and atrial natriuretic factor-mediated vasorelaxation of contracted, endothelial-denuded rat aortic rings. Moreover, WIN 58237 possesses vasorelaxant activity in the presence of an intact endothelium or nitric oxide. Similar results are evident in vivo, as WIN 58237 (0.3-3.0 mg/kg i.v.) decreases mean arterial pressure in conscious spontaneously hypertensive rats with an associated increase in vascular (aortic) cGMP content in vivo. Both the decrease in mean arterial blood pressure and increase in aortic cGMP content are attenuated by the nitric oxide synthase inhibitor, N omega-nitro-l-arginine. However, WIN 58237 may possess an additional depressor mechanism of action. WIN 58237 restores vasorelaxation responsiveness to nitroglycerin in vitro and in vivo in models of vascular tolerance.(ABSTRACT TRUNCATED AT 250 WORDS)

Pharmacologic and pharmacodynamic effects of the selective low Km cyclic AMP phosphodiesterase III inhibitors WIN 63291 and WIN 62582.

We describe the biochemical, pharmacologic, and in vivo pharmacodynamic profiles of two novel inhibitors of the cyclic GMP-inhibitable, low Km cyclic AMP phosphodiesterase (PDE) III; WIN 63291, a 6-quinolinyl analogue of the prototypic PDE III inhibitor milrinone and WIN 62582, an imidazopyridinone. Both WIN 62582 and WIN 63291 competitively inhibit PDE III from rat, dog, and human heart and from rat and canine aorta with IC50 values of 5-37 and 55-80 nM, respectively; the IC50 values for milrinone ranged from 300 to 520 nM. WIN 62582 and WIN 63291 are at least 1,000-fold selective for PDE III relative to inhibition of PDE isozymes I, II, IV, and V. We evaluated WIN 62582 and WIN 63291 in conscious rats and dogs after intravenous (i.v.) and oral (p.o.) administration. The dose of WIN 62582 required to reduce mean arterial blood pressure (MAP) by 20% (ED20) in rats was 1.8 mg/kg, with a pharmacodynamic duration of action of approximately 2 h. In comparison, the estimated i.v. ED20 for WIN 63291 in rats was 0.4 mg/kg, with a pharmacodynamic duration of action > 6 h. In conscious dogs, the i.v. doses of WIN 62582 and 63291 required to increase left ventricular (LV)dP/dtmax significantly were 0.1 and 0.01 mg/kg, respectively. In dogs, WIN 63291 0.1 mg/kg p.o. increased LVdP/dtmax by 86% in 30 min; LVdP/dtmax remained increased by 60% for at least 6 h. In comparison, WIN 62582, 0.3 mg/kg p.o., increased LVdP/dt by 56% in 30 min and remained increased by 40% at 6 h.(ABSTRACT TRUNCATED AT 250 WORDS)

Zaprinast increases cyclic GMP levels in plasma and in aortic tissue of rats.

The purpose of this study was to determine if significant relationships exist between plasma and aortic cyclic GMP (cGMP) levels and pharmacodynamic effect after the i.v. administration of the cGMP-selective phosphodiesterase inhibitor zaprinast to conscious, spontaneously hypertensive rats. Zaprinast dose-dependently increased plasma and aortic cGMP levels at 10, 18 and 30 mg/kg and decreased mean arterial blood pressure (MAP) at 18 and 30 mg/kg. The concentrations of cGMP in the plasma and in the aorta were significantly correlated (r = 0.765, P < 0.0001). The changes in MAP were significantly correlated to aortic (r = -0.750, P < 0.0001) and plasma (r = -0.762, P < 0.0001) cGMP levels. We conclude that plasma cGMP may be an index of cGMP-selective phosphodiesterase inhibition in vivo.

Hemodynamic and renal effects of the protein kinase inhibitor staurosporine in conscious rats.

To evaluate the pattern of hemodynamic responses produced by an inhibitor of protein kinase C (PKC), staurosporine 0.03-0.55 mg/kg was administered intravenously (i.v.) to conscious, normotensive rats chronically instrumented with vascular catheters for direct measurement of blood pressure (BP) and i.v. administration of drugs and either an aortic flow probe for measurement of cardiac output (CO) or miniaturized pulsed Doppler flow probes for measurement of hindquarter, renal, and mesenteric vascular resistances. Staurosporine decreased mean arterial pressure (MAP) and total peripheral resistance (TPR) and increased heart rate (HR) in a dose-dependent manner. Because staurosporine decreased resistance in all three vascular beds monitored (hindquarter, renal, and mesenteric), staurosporine is probably a nonselective vasodilator that decreases MAP by decreasing resistance in a number of peripheral vascular beds. Staurosporine produced biphasic effects on CO, dF/dtmax and peak aortic blood flow; these parameters were significantly increased at doses less than 0.3 mg/kg and decreased to levels equal to or significantly less than control values at doses greater than 0.3 mg/kg. In comparison, the calcium channel blocker nitrendipine decreased MAP and TPR and increased HR, CO, dF/dtmax, and peak aortic flow in a dose-dependent manner over the entire dose range (0.01-1 mg/kg i.v.). Staurosporine (0.3 mg/kg) and nitrendipine (1 mg/kg) produced similar changes in MAP (-44 +/- 3 and -33 +/- 2 mm Hg, respectively), yet staurosporine affected dF/dtmax to a lesser extent than nitrendipine (-5 +/- 36 and 390 +/- 46 ml/s/s, respectively).(ABSTRACT TRUNCATED AT 250 WORDS)

Differential hemodynamic responses to selective inhibitors of cyclic nucleotide phosphodiesterases in conscious rats.

Selective inhibition of either the low Km cyclic AMP (cAMP) or low Km cyclic GMP (cGMP) phosphodiesterase (PDE) promotes vasorelaxation and, consequently, produces depressor effects. To evaluate the systemic and regional hemodynamic effects of selective inhibitors of these PDE isozymes, CI-930 (0.1-10 mg/kg), an inhibitor of low Km cAMP PDE, or zaprinast (3-30 mg/kg), an inhibitor of low Km cGMP PDE, was given i.v. to conscious, normotensive rats. The rats were chronically instrumented with vascular catheters and either an ultrasonic transit-time flow probe around the ascending aorta or miniaturized pulsed Doppler flow probes around the superior mesenteric and left renal arteries and the abdominal aorta. CI-930 and zaprinast, at cumulative doses of 3 and 30 mg/kg, respectively, produced comparable reductions in mean arterial pressure (-22 +/- 3 and -19 +/- 4 mm Hg, respectively) and total peripheral resistance (-0.41 +/- 0.07 and -0.42 +/- 0.06 mm Hg/ml/min, respectively) but affected other hemodynamic variables differently. CI-930 at 3 mg/kg increased the heart rate (HR), maximal aortic flow acceleration (dF/dt), and peak aortic flow and decreased the stroke volume (SV). Cardiac output (CO) was not affected by CI-930. Zaprinast at 30 mg/kg increased the CO, dF/dt, and peak aortic blood flow. The HR and SV were unaffected by zaprinast. Although both CI-930 and zaprinast increased the dF/dt and peak aortic flow, these parameters were affected more by CI-930 than by zaprinast. CI-930 decreased hindquarter, mesenteric, and renal vascular resistances in a dose-dependent manner.(ABSTRACT TRUNCATED AT 250 WORDS)

The effects of iodixanol and iopamidol on hemodynamic and cardiac electrophysiologic parameters in vitro and in vivo.

Iodixanol is a new, nonionic, dimeric contrast medium which, in concentrations appropriate for radiographic use, is hypotonic with respect to plasma. The purpose of these in vivo and in vitro studies was to compare the effects of iopamidol, iodixanol formulated to isotonicity with sodium salts (sodium formulation), and iodixanol formulated to isotonicity with sodium, calcium, and magnesium salts (cationic formulation) on hemodynamic and electrophysiologic parameters. In vitro, the spontaneous rate of contraction by guinea pig right atrial and force development by right ventricular papillary muscles were evaluated in the presence of 1% to 100% (v/v) of the three contrast media. Iopamidol significantly (P less than .05) decreased the rate of atrial contraction to a greater extent than either formulation of iodixanol. Iopamidol decreased papillary muscle force development more than the sodium formulation of iodixanol (P less than .05). The cationic formulation of iodixanol had little effect (less than 30% change) on papillary muscle force development at concentrations up to 100%. The contrast media were also injected into the left coronary arteries of open-chest, anesthetized dogs at 0.8 mL/second for 5 to 30 seconds. All contrast media increased (P less than .05) systolic blood pressure (SBP), mean arterial pressure (MAP), and peak left ventricular pressure (LVP). Iopamidol increased LVP and LV end diastolic pressure to a greater extent (P less than .05) than the cationic formulation of iodixanol. We conclude that iopamidol affected cardiovascular parameters more than iodixanol.

N omega-nitro-L-arginine attenuates the accumulation of aortic cyclic GMP and the hypotension produced by zaprinast.

To determine if N omega-nitro-L-arginine (NNA), an inhibitor of the synthesis and/or release of endothelium-derived relaxing factor (EDRF), alters the response to zaprinast, a selective inhibitor of cyclic GMP (cGMP) phosphodiesterase, zaprinast (3-30 mg/kg) or vehicle (1 ml/kg) was given to conscious, spontaneously hypertensive rats (SHR) in a cumulative i.v. dose-response manner 30 min after pretreatment with NNA (1 or 3 mg/kg) or saline (1 ml/kg). Mean arterial pressure (MAP) was measured 5 min after each dose of zaprinast. Five minutes after the last dose of zaprinast (30 mg/kg), the rats were anesthetized with pentobarbital (25 mg i.v.). A segment of the abdominal aorta was freeze-clamped in situ and removed for the determination of cGMP levels. NNA (3 mg/kg) decreased basal aortic cGMP levels by 54% and increased MAP by 37 +/- 2 mm Hg. Zaprinast (30 mg/kg) increased aortic cGMP by 187% and decreased MAP by 49 +/- 4 mm Hg. NNA (3 mg/kg) reduced the accumulation of cGMP in aortic tissue (from 4.1 +/- 0.4 to 1.3 +/- 0.1 fmol/microgram protein) and attenuated the depressor response (from -49 +/- 4 to -31 +/- 4 mm Hg) produced by zaprinast. These data are consistent with the hypothesis that NNA inhibits the tonic release of EDRF and that the depressor effects of zaprinast are due, at least in part, to the potentiation of the vasodilator effects of EDRF in vivo. Moreover, since the changes in MAP produced by NNA and zaprinast were significantly correlated with cGMP levels in aortic tissue, the concentration of cGMP in vascular tissue may be a determinant of blood pressure in SHR.

Protein kinase inhibitors and blood pressure control in spontaneously hypertensive rats.

Considerable evidence suggests that protein kinase C activation participates in the regulation of vascular smooth muscle tone. The objective of the current study was to examine the relations between inhibition of protein kinase C (PKC) and myosin light-chain kinase (MLCK) and vasorelaxation and blood pressure regulation in spontaneously hypertensive rats (SHR). Putative PKC inhibitors from two chemical classes, staurosporinelike (staurosporine and K252A) and isoquinolinesulfonamides (H7 and HA1004), were tested for their ability to 1) inhibit PKC and MLCK from SHR aorta, 2) relax isolated SHR aorta, and 3) lower blood pressure in conscious SHR. A rank order of potency for the inhibition of PKC and MLCK was established, with the staurosporinelike compounds (staurosporine PKC IC50 = 54 nM) clearly more potent than the isoquinolinesulfonamides (H7 PKC IC50 = 128 microM). The rank order of potency for inhibition of PKC was retained for inhibition of MLCK for all compounds. Staurosporine (EC50 = 75 nM) and H7 (EC50 = 2 microM) caused concentration-dependent relaxation of SHR aorta, but only staurosporine produced vasorelaxation at concentrations consistent with the inhibition of PKC or MLCK. Dose-dependent reductions in arterial pressure of SHR were demonstrated after intravenous injection of staurosporine and HA1004. A single intravenous injection of staurosporine (0.3 mg/kg) lowered blood pressure for more than 10 hours. Staurosporine also lowered blood pressure after oral administration. The depressor response to staurosporine was unaffected by sympathetic beta-adrenergic blockade. In conclusion, the vasorelaxant and antihypertensive actions of staurosporine in SHR are consistent with the inhibition of PKC but could also be equally related to inhibition of MLCK. Not all PKC inhibitors produce vasorelaxation and lower blood pressure. Moreover, the lack of correlation between in vitro vasodilation and PKC or MLCK inhibition for the isoquinolinesulfonamide protein kinase inhibitors H7 and HA1004 suggests that these agents do not cause vasorelaxation in SHR by inhibition of these enzymes.

Inhibition of low Km cyclic GMP phosphodiesterases and potentiation of guanylate cyclase activators by cicletanine.

Cicletanine is an antihypertensive/vasorelaxant/natriuretic agent of unknown mechanism. We wished (a) to determine if cicletanine interacts with guanylate cyclase activators that modulate vasomotor tone and sodium balance [i.e., atriopeptin II (AP II), endothelium-derived relaxing factor (EDRF), and sodium nitroprusside (SNP)], and (b) to define the subcellular basis for this interaction by quantitating the effects of cicletanine on low Km cyclic GMP phosphodiesterase (PDE) activity. In phenylephrine-contracted rat aortic smooth muscle, the vasorelaxant potency of cicletanine was increased twofold in the presence of a threshold-relaxant concentration of AP II, and functional cyclic GMP PDE inhibition was also evident from the three- to sixfold potentiation by cicletanine of AP II- or SNP-induced vasorelaxation. Vasorelaxation produced by cicletanine was not endothelium dependent, however. In further studies, intravenous (i.v.) administration of cicletanine or the low Km cyclic GMP PDE inhibitor, zaprinast, decreased blood pressure (BP) greater than or equal to 20% in conscious spontaneously hypertensive rats (SHR). These results are consistent with the additional finding that cicletanine inhibited Ca2(+)-calmodulin (CaM) cyclic GMP PDE and zaprinast-sensitive cyclic GMP specific PDE over a concentration range (10-600 microM) similar to that for vasorelaxation. Thus, inhibition of low Km cyclic GMP PDEs by cicletanine may be partly responsible for the vasorelaxant effect of cicletanine as well as the potentiation by cicletanine of the vasorelaxant actions of guanylate cyclase activators. The extent to which this mechanism contributes to the antihypertensive efficacy of cicletanine has not yet been fully determined.

Sodium nitroprusside potentiates the depressor response to the phosphodiesterase inhibitor zaprinast in rats.

To determine if the presence of an activator of guanylate cyclase alters the depressor response to a selective inhibitor of low Km cyclic GMP (cGMP) phosphodiesterase (PDE), zaprinast (3-30 mg/kg) was given i.v. to conscious, spontaneously hypertensive rats during a steady state of i.v. infusion of sodium nitroprusside (15 micrograms/kg per min). Sodium nitroprusside significantly increased the magnitude of the depressor response to zaprinast. In contrast, fenoldopam (20 micrograms/kg per min), an activator of adenylate cyclase, did not affect the depressor response to zaprinast. Zaprinast (10 mg/kg) significantly decreased mean arterial pressure (MAP) in rats given an infusion of sodium nitroprusside, an activator of soluble guanylate cyclase, at doses of 15 and 25 micrograms/kg per min but not at a dose of 5 micrograms/kg per min. However, in rats given atrial natriuretic peptide (ANP; 0.5, 1 and 2 micrograms/kg per min), an activator of particulate guanylate cyclase, zaprinast (10 mg/kg) did not affect MAP. In contrast to the potentiation of the depressor response to zaprinast, sodium nitroprusside (15 micrograms/kg per min) significantly attenuated the reductions in MAP produced by CI-930, a selective inhibitor of low Km cAMP PDE. It is concluded that sodium nitroprusside, but not ANP or fenoldopam, potentiates the depressor response to zaprinast. Furthermore, the potentiation of the depressor response to zaprinast is dependent upon the dose of sodium nitroprusside and is selective for zaprinast; the depressor response to CI-930 is attenuated by sodium nitroprusside.

Regional cerebral glucose utilization during vasopressin-induced barrel rotations or bicuculline-induced seizures in rats.

Regional cerebral metabolic rates of glucose (rCMRglu) were measured in conscious rats grouped according to three treatments: control, bicuculline (5.5 mg/kg s.c.), and intracerebroventricular arginine vasopressin (0.5 micrograms/5 microliter). Rats in the latter group were pre-exposed to the peptide 48 h earlier to render them 100% susceptible to a motor output termed barrel rotation (BR). Marked increases in rCMRglu occurred in all brain areas investigated after bicuculline. Increases after intracerebroventricular arginine vasopressin were smaller and occurred in fewer brain areas. Opposite to bicuculline, arginine vasopressin reduced rCMRglu in hippocampus and auditory cortex. The data confirm marked stimulation of rCMRglu during bicuculline-induced seizures and provide initial data for an endogenous peptide that causes BR. Generalized seizures and BR may share some neural substrates, but they produce distinct changes in rCMRglu. The rCMRglu changes are compatible with the interpretation that BR is initiated by brainstem/cerebellar areas that process visual-vestibular information, with subsequent involvement of higher brain structures.

Site-dependent central effects of aldosterone in rats.

A relationship between the subcommissural organ (SCO) and the adrenal glands has long been suspected. This report provides further information about the effects of a continuous D-aldosterone infusion into the SCO area of conscious, adult male Sprague-Dawley rats. A 6-day aldosterone infusion (5 ng/h) increased urinary sodium excretion, decreased adrenal medullary cross-sectional area, elevated adrenal corticosterone content and terminal plasma epinephrine concentration. Mineralocorticoid infusions directly into a lateral cerebral ventricle did not affect these parameters but, unlike SCO area infusions, decreased consummatory behavior. Infusions of tritiated aldosterone into the SCO area revealed that radioactivity was mainly confined to dorsomedial portions of the brain near the SCO, whereas the pineal body contained only background radioactivity. The data support the concept that the SCO area interacts with physiological systems related to both the adrenal cortex and medulla.

Barrel rotation evoked by intracerebroventricular vasopressin injections in conscious rats. II. Visual/vestibular interactions and efficacy of antiseizure drugs.

Intracerebroventricular (i.c.v.) arginine-vasopressin (AVP) injections evoke 'barrel rotation' (BR) in rats. This motor system abnormality was studied in a protocol where conscious rats were injected on day 1 with 1 microgram i.c.v. AVP and reexposed to 0.5 micrograms on day 3. Three paradigms modifying visual/vestibular systems were employed: labyrinthectomy, 3-acetylpyridine (3-AP) destruction of the inferior olive and atropine pretreatment. Ambient illumination (light vs dark) was also modified. Initial (day 1) incidence of BR, increased incidence (i.e. sensitization) on day 3, and day 3 BR latency were differentially affected by the various paradigms and suggest a complex role of visual/vestibular input in modifying i.c.v. AVP-induced BR. For example, 3-AP rats tested in light and atropinized rats had a reduced responsiveness to the peptide on day 1. 3-AP-treated rats tested in dark conditions showed a normal incidence of BR on day 1, but the expected sensitization to AVP on day 3 did not occur. Combined labyrinthectomy and darkness did not modify BR incidence on either day, but altered the distribution of latency data. Four diverse antiepileptic drugs were tested for efficacy against i.c.v. AVP-induced BR in sensitized rats: phenytoin, diazepam, valproic acid and phenobarbital; all drugs reduced the proportion of rats with BR and prolonged the latency. We conclude that brain AVP may be involved in abnormal motor conditions that are modified by visual/vestibular neuronal circuits. The unusual motor output (barrel rotation) can be inhibited by diverse antiepileptic drugs.

Barrel rotation evoked by intracerebroventricular vasopressin injections in conscious rats. I. Description and general pharmacology.

Intracerebroventricular (i.c.v.) arginine-vasopressin (AVP) injections in rats evoke an unusual motor response termed 'barrel rotation' (BR). This report documents several aspects of BR after i.c.v. AVP in conscious, adult male Sprague-Dawley rats: single i.c.v. AVP injections (100-1000 ng/5 microliters) evoke BR in about 50% of naive rats with no relationship to dose and 20% mortality; no directional preference exists for BR, and sensitivity to BR does not vary over a weight range of 301-475 g; continuous i.c.v. AVP infusions at doses of 50-2500 ng/h evoked BR in 13 and 50% of rats tested at the extreme ranges; latency to BR was always within 3-6 min in infusion experiments; a protocol where rats received a single i.c.v. AVP injection (1 microgram) on day 1 followed on day 3 by 0.5 micrograms, increased the proportion of rats with BR from 51% to 83% (P less than 0.05), indicating a sensitization phenomenon; latency to BR after single i.c.v. injections did not fit the assumption of single underlying normal distribution; a novel method to analyze these data, hazard plotting, revealed two phases to the BR latency under ambient illumination. The following paper presents evidence of visual/vestibular involvement and the efficacy of anti-seizure drugs. Collectively, the data are compatible with the hypothesis that brain vasopressin pathways are involved in some abnormalities of motor output.