GABA and circadian timekeeping: implications for manic-depression and sleep disorders

Could the underestimation of bipolarity obstruct the search for novel antidepressant drugs?

INTRODUCTION

Despite the clinical and social relevance of depression, and the availability of numerous antidepressants and non-pharmacological interventions, response rates remain unsatisfactory and novel therapeutic targets are being explored.

AREAS COVERED

This review starts with a brief overview of the evolution of the current antidepressant drug scenario and ends with a focus on the potential influence of the underestimation of bipolarity on the exploration of novel antidepressant drugs.

EXPERT OPINION

The field of antidepressant drug development has suffered from a relative decline recently and, with the exception of agomelatine, innovative non-monoaminergic antidepressants have yet to be developed. The need for more effective compounds is evident. Clinicians and researchers should pay greater attention to the impact of bipolarity in depression. The ultimate goal of this review is not to discourage the use of antidepressants but rather to encourage judicious prescriptions, and also to solicit a better collaboration between clinicians and preclinical researchers so that more reliable diagnostic criteria can be adopted.

Alpha-keto-beta-methylvaleric acid increases the in vitro phosphorylation of intermediate filaments in cerebral cortex of young rats through the gabaergic system

In this study we investigated the effects of alpha-ketoisovaleric (KIV) and alpha-keto-beta-methylvaleric acids (KMV), metabolites accumulating in the inherited neurometabolic disorder maple syrup urine disease (MSUD), on the in vitro incorporation of 32P into intermediate filament (IF) proteins from cerebral cortex of young rats during development (9-21 days of age) We observed that KMV significantly increased the in vitro incorporation of 32P into the IF proteins studied in cortical slices of 12-day-old rats through the PKA and PKCaMII, with no alteration at the other ages. In contrast, KIV was ineffective in altering the phosphorylating system associated with IF proteins at all ages examined. A similar effect on IF phosphorylation was achieved by incubating cortical slices with gamma-aminobutiric acid (GABA). Furthermore, by using specific GABA antagonists, we verified that KMV induced a stimulatory effect on IF phosphorylation of tissue slices from 12-day-old rats mediated by GABA(A) and GABA(B) receptors. In conclusion, our results indicate the involvement of the GABAergic system in the alterations of IF phosphorylation caused by KMV, one of the branched-chain keto acids accumulating in MSUD.

Valpromide increases amplitude of heart rate circadian rhythm in remitted bipolar and unipolar disorders. A placebo-controlled study

The aim of this study was to investigate for the effects of valpromide on heart rate circadian rhythm in remitted recurrent unipolar and bipolar disorders (DSM-III-R). It consisted of a comparative, randomized, double-blind, repeated cross-over study of valpromide versus placebo over four four-week periods. The primary evaluation criteria was heart rate (HR). Secondary criteria comprised motor activity (MA) and the Bech and Rafaelsen mania assessment scale, Horne and Ostberg questionnaire, Montgomery and Asberg depression rating scale, Spiegel questionnaire, a sleep diary, and Clinical Global Impression. Fifteen patients were included, giving 60 one-month periods (30 valpromide periods and 30 placebo periods). Cosinor analysis of HR and MA data revealed a difference in amplitude (P = 0.037, analysis of variance, one-tailed test). The clinical sleep study shows that the duration of sleep was greater with valpromide than with placebo (P = 0.007, one-tailed test). Similarly, evaluation of the quality of sleep by patients themselves showed valpromide to be superior to placebo (P = 0.045, one-tailed test). The results of analysis of the Spiegel questionnaire also confirm the superiority of valpromide over placebo. Safety and compliance were comparable for the active drug and the placebo. In conclusion, the relatively small sample size requires cautious interpretation of this study. Nevertheless, these initial results show a definite effect of valpromide on a biological rhythm that leads one to suppose that it may be effective through a 'synchronizing' effect.

Valproate for sleep consolidation in periodic limb movement disorder

In this study, open-label valproate (VPA) was administered to patients as a treatment for periodic limb movement disorder (PLMD). Six patients aged 28 to 62 years with complaints of sleep disturbance and at least five periodic limb movements (PLMs) per hour of sleep underwent polysomnograms (PSGs) with and without low-dose VPA treatment (125-600 mg at bedtime). After a baseline PSG, patients received VPA therapy from 2 weeks to 14 months, until the time of the follow-up PSG on VPA (median, 5 months; mean, 6 months). All six patients experienced subjective improvement in daytime alertness. Sleep efficiency was improved from 76% to 88% (p = 0.003), stage 1 (light) sleep decreased from 26% to 13% (p = 0.04), stage 3 and 4 (deep) sleep increased from 19% to 30% (p = 0.01), and rapid eye movement sleep was unchanged. There was a trend toward a reduction in the number of PLMs per hour of sleep and in the percentage of arousals associated with PLMs. All of the patients continued taking VPA after the PSGs were completed. One patient discontinued VPA 1 month after completion of the last PSG because of short-term side effects, and one patient stopped VPA 22 months after the last PSG because of weight gain. Thus, these data indicate that VPA has a long-term beneficial effect on sleep consolidation in patients with PLMD.

The emerging role of valproate in bipolar disorder and other psychiatric disorders

Although valproate, a simple branched-chain fatty acid, is generally considered to be an antiepileptic agent, a large literature dating back to 1966 describes its use in primary psychiatric disorders. The significant role that gamma-aminobutyric acid plays in mood provided the rationale to examine valproate in this regard. Numerous uncontrolled as well as placebo- and lithium-controlled studies verified the drug's efficacy in the short-term management of bipolar and schizoaffective disorders. The response appears to be independent of response to traditional therapies such as lithium, neuroleptics, and carbamazepine, and may be maintained for extended periods of time. Valproate should be considered not only in patients with mood disorders who are intolerant of or nonresponsive to traditional therapies, but also in those with rapid cycling, electroencephalographic abnormalities, head trauma antedating the onset of psychiatric illness, or any other factor suggesting an organic component. Preliminary uncontrolled studies suggest that the drug may also eventually play a role in the management of panic disorder and behavioral dyscontrol (agitation, aggression, temper outbursts). Its adverse event profile is well known from years of experience in the management of epilepsy and does not appear to be altered in the presence of psychiatric disorders. Similarly, the drug-drug interaction potential of valproate is reasonably well known, although further research into interactions with psychotropic agents is warranted.

Rhythms and the pharmacology of lithium

Lithium is the treatment of choice for bipolar affective disorder (manic-depression) and is useful in other recurrent affective and nonaffective illnesses. This review discusses lithium's actions on period, phase, amplitude and coupling of biological rhythms that may relate to its therapeutic effectiveness. Alternatively, lithium might interact with environmental light to influence circadian rhythms by an action on the retina. The mechanisms responsible for lithium's chronopharmacological actions are not known, but cellular cations, phosphoinositide or adenylate cyclase second messenger systems, hormones and neurotransmitters may all be involved.

Demonstration of GABAergic cell bodies in the suprachiasmatic nucleus: in situ hybridization of glutamic acid decarboxylase (GAD) mRNA and immunocytochemistry of GAD and GABA

The existence of GABAergic neurons in the rat suprachiasmatic nucleus (SCN) was demonstrated by three specific markers; mRNA coding for glutamic acid decarboxylase (GAD) and visualized by in situ hybridization using a 35S-labelled cDNA probe, and GAD protein and GABA were identified by immunocytochemistry using specific antisera. In situ hybridization demonstrated well labelled GAD mRNA positive cells throughout SCN, and GABA and GAD immunoreactive cells showed similar distributions. These results indicate that GABA is a transmitter of a large portion of the SCN neuronal population.

Psychosis, lithium-induced antipsychotic response, and seasonality

Lithium alone has previously been shown to effect antipsychotic response in approximately one-quarter of mood incongruent psychoses. Such lithium-responsive psychoses and/or the antipsychotic effects of lithium appear to be seasonal, with maximal response rates occurring in the 3-month periods surrounding January and July peaks. Within patients, retrials of lithium in formerly lithium-responsive psychotic patients demonstrate a trend toward greater continued antipsychotic effects of lithium during the periods surrounding January and July (80% response rates) vs. the remaining 6 months of the year (33% response rates).

Different responses of the circadian system to GABA-active drugs in two strains of mice

Recent work in our laboratory has shown that sodium pentobarbital injections can induce phase-dependent phase shifts of the circadian rhythm of locomotor activity with the maximum advance at circadian time (CT) 8 and the maximum delay at CT0 in SK/Nga mice but no phase shifts in C57BL/6 mice. In the present study, the possibility that the differences in the effects of pentobarbital on the circadian rhythm may be due to different contributions of the GABA-ergic system to circadian organization in the two strains was tested by comparing the responses of SK mice with those of C57BL mice to muscimol (2 mg/kg), a GABA receptor agonist, and triazolam (25 mg/kg), which is thought to act by potentiating the action of GABA. The hypothesis that pentobarbital-induced phase shifts of SK mice are mediated by the GABA receptor system was also tested by observing whether the phase-shifting effects of pentobarbital were blocked by bicuculline (0.5 mg/kg), a selective antagonist of GABA, injected 3 min prior to pentobarbital (30 mg/kg). The results indicated that muscimol induced phase advances at CT8 and phase delays at CT0, and triazolam induced phase advances at CT8 in SK mice. No phase shifts were induced by any treatment in C57BL mice. These results suggest that the role of GABA-ergic systems in circadian organization may be different in SK and C57BL mice. In addition, bicuculline could block the phase-shifting effects of pentobarbital in SK mice, suggesting that the GABA receptor system may mediate phase-shifting effects of pentobarbital in SK mice.

Effects of protein restriction, melatonin administration, and short daylength on brain benzodiazepine receptors in prepubertal male rats

The possibility that there are changes in brain benzodiazepine binding sites controlled by photoperiod was investigated in two strains of male rats. The hypothesis was tested by 3H-diazepam binding studies in various brain regions of prepubertal rats maintained in 14 or 10 h of light or treated with late-afternoon injections of melatonin (50 micrograms/day). Protein restriction was applied during the experiment to sensitise the animals to the treatments. Under the conditions employed, rats kept in short daylength throughout or kept on long photoperiod and given late-afternoon melatonin injections showed evidence of delayed puberty (seminal vesicle, ventral prostate, and testis weight decreased by 45%, 55%, and 60% respectively, compared to control rats). Binding measurements were made 1 h before and 2 and 5 h after the onset of darkness in the pubertal (42-day-old) or experimentally prepubertal rats. In the rats of the Porton strain (for which protein restriction was obligatory for the gonadal response) there was no consistent treatment or time effects on specific binding of 3H-diazepam to washed membranes of the hypothalamus, midbrain, or striatum. Similarly, there were no differences in the stimulation of 3H-diazepam binding by 100 microM GABA or the inhibition of binding by 50 microM N-acetyl 5 methoxy kynurenamine. By contrast, in Wistar rats, specific binding to midbrain membranes was reduced 5 h after dark compared to 2 h (37% saline; 20% melatonin) and the extent of stimulation by GABA in the hypothalamus was increased 5 h after darkness (35.6% to 46.7% saline; 37.4% to 50% melatonin). Melatonin treatment resulted in significantly higher specific binding in the hypothalamus 2 h after dark (10%, control fed; 20%, protein restricted) but reduced the GABA induced stimulation of binding in the midbrain (35.5% to 25%, control fed; 33.7% to 23.5%, protein restricted). The Bmax of benzodiazepine binding to unwashed cortical P2 synaptosomal membranes has been reported to increase twofold in adult Wistar rats at mid-dark. By contrast the Bmax of juvenile Wistar rats in this study increased only 17% (116 +/- 2.4 fmol/mg protein to 140 +/- 3 fmol/mg protein) between 2 and 5 h after darkness. In melatonin-treated animals the increase in Bmax of 3H-diazepam binding was blocked (124 +/- 5 fmol/mg protein at 2 h; 127 +/- 3 fmol/mg protein at 5 h) and the Kd reduced (4.5 +/- 0.5 to 4.0 +/- 0.2 nM).(ABSTRACT TRUNCATED AT 400 WORDS)