For patients with bipolar disorder, the antiepileptic drug lamotrigine has been shown to be more effective in treating depression than in treating hypomania or mania.35
Recently, a number of studies have reported that this drug is effective in the treatment of major depressive disorders, including refractory unipolar depression,36,37,38
which suggests promise for the use of lamotrigine as an augmentation drug. Pilot studies have suggested that lamotrigine augmentation of SSRIs such as fluoxetine39
is well tolerated and provides efficacy superior to that of SSRI monotherapy. Lamotrigine may also have antidepressant properties in patients with unipolar depression and may accelerate the onset of action when given in combination with classical antidepressants.40
The aim of the present study was to evaluate the antidepressant-like effect of lamotrigine in the FST. Despite many years of research, no valid and satisfactory animal model of bipolar depression has been developed to evaluate the mechanism of action of mood stabilizers. Animal models mimic only limited aspects of the behaviours seen in this disorder, either mania or depression. Here, we have evaluated the anti-immobility effect of lamotrigine in mice using a simple but robust model of depression, the FST.20,21,22
Furthermore, the FST enables investigation of mechanisms of action of drugs through the use of association studies with receptor-specific ligands. The present work focuses mainly on 5-HT1
In this study, lamotrigine impaired spontaneous locomotor activity, with some effect at a dose of 4 mg/kg and a decrease in activity of more than 50% at 16 mg/kg. In addition, lamotrigine at doses of 8 and 16 mg/kg, administered 30 minutes before the FST, decreased immobility time, thus demonstrating antidepressant-like effects despite the drug's sedative effects at these doses. In a previous study using the FST in mice,41
lamotrigine, at doses of 15 and 25 mg/kg, was found to be inactive relative to saline. This discrepancy might be due to differences in methodology. Indeed, in the previous study,41
lamotrigine was administered 60 minutes before testing instead of 30 minutes as in our study. Furthermore, the size of the glass cylinder was larger (16 cm v. 10 cm). Standardization of the FST is required to avoid difficulties in replicating experiments between laboratories and to ensure reproducibility of results.
Lamotrigine administered alone seems to be more effective than lithium, valproate or carbamazepine in decreasing immobility time, with an effect of about 17% at the upper dose of 16 mg/kg. Lithium and sodium valproate have been previously evaluated with the FST and were found to be without effect,42
whereas carbamazepine was weakly effective at a dose of 32 mg/kg.16
Recently, large double-blind, placebo-controlled trials comparing lamotrigine with lithium in the maintenance treatment of patients with bipolar I disorders have been performed.43
Both drugs were effective as maintenance treatment, and they had potentially complementary mood-stabilizing properties. These data are consistent with the results of the present experimental study and suggest that lamotrigine is more effective in treating depression, whereas lithium is more effective in treating mania.8,9,43
According to the newly available nomenclature for bipolar disorder, which distinguishes “above baseline,” for mania and hypomania and mixed states, from “below baseline,” for depression and subsyndromal depression, lithium would be the prototype of class A drugs in treatment of “above baseline” moods and lamotrigine might belong to class B drugs in treatment of “below baseline” moods.5,8,9
Valproate and carbamazepine also belong to class A.5
Alteration of serotonin transmission is implicated in many mood disorders including bipolar disorders. Lopez-Figueroa et al15
have recently shown alteration in 5-HT1A
mRNA levels using post mortem in situ hybridization in the brains of subjects with a history of major depressive disorder or bipolar disorder.
In the present study, lamotrigine was coadministered with specific 5-HT1A
receptor agonists and antagonists to investigate the role of these receptors in vivo in the mechanism of action of lamotrigine in the FST, as reported previously in our laboratory for the other mood stabilizers (lithium, carbamazepine and sodium valproate).16
Only 8-OH-DPAT (a postsynaptic 5-HT1A
receptor agonist), pindolol (a presynaptic and postsynaptic 5-HT1A/1B
receptor antagonist) and to a lesser extent RU 24969 (a 5-HT1A/1B
receptor agonist that acts essentially via postsynaptic receptors) were able to enhance the antidepressant-like effect of lamotrigine in the FST.
subtype receptors are believed to exist at the presynaptic level as autoreceptors controlling 5-HT release and as heteroreceptors controlling the release of other neurotransmitters44
and at the postsynaptic level (coupled to adenylate cyclase).
8-OH-DPAT, like other high-affinity agonists for 5-HT1A
receptors, has been shown to decrease immobility time in the FST on its own.21
Treatments with PCPA (p
-chlorophenylalanine, inhibitor of 5-HT synthesis) causing serotonergic depletion or with the neurotoxin 5,7-DHT (5,7-di-hydroxytryptamine) do not affect the immobility response to 8-OH-DPAT in the mouse FST, which suggests that this effect is mediated by postsynaptic 5-HT1A
In the present study, the observed additive effect of a subactive dose of 8-OH-DPAT in combination with lamotrigine may suggest that postsynaptic 5-HT1A
receptors play a role in the ability of these drug to reduce immobility in the FST.
Antagonist activity at presynaptic 5-HT1A
receptors, which blocks the negative feedback mechanism involved in 5-HT neurotransmission via autoreceptors,46
has been implicated in the potentiating effects of pindolol in the FST.21
Furthermore, it has been demonstrated that pindolol acts at 5-HT1B
presynaptic receptors in the FST with antidepressant-like activity.47
In this study, the combination of subactive doses of lamotrigine and pindolol also decreased immobility time in the FST, which supports presynaptic activity, but the effect was less pronounced than with 8-OH-DPAT.
RU 24969 is believed to act via postsynaptic 5-HT1B
Only the higher subactive dose of lamotrigine was potentiated by RU 24969. Given that anpirtoline was without effect on lamotrigine-induced immobility time, whereas 8-OH-DPAT potentiated the effect, it seems more likely that a 5-HT1A
agonist activity might be involved in the potentiation induced by RU 24969.
Taking these data together, and considering the effect size of the decrease in immobility time of lamotrigine in association with the various ligands (i.e., 8-OH-DPAT > pindolol > RU 24969), we can speculate that postsynaptic 5-HT1A receptors might be involved in the activity of lamotrigine. Other neurotransmitters regulated by postsynaptic 5-HT1A heteroreceptors might also participate in the antidepressant-like effect of lamotrigine.
Although there is a large clinical literature reporting the use of lamotrigine in bipolar disorders or treatment-resistant depression, few studies have addressed the possible neurochemical basis of its therapeutic effect, and these have yielded conflicting results. Some researchers have found that lamotrigine has little or no effect on 5-HT transmission,6,48
whereas Ahmad et al49
recently found that single administration of lamotrigine (10 and 20 mg/kg) decreased extracellular 5-HT in the hippocampus of freely moving rats (determined by in vivo microdialysis). In contrast, carbamazepine and lamotrigine were the only anticonvulsant drugs found to inhibit 5-HT uptake both in vitro and in vivo,50
which would result in elevated extracellular 5-HT concentrations. However, the relevance of these findings to bipolar depression is unclear. A recent study in rats demonstrated significantly less inhibition of forskolin-stimulated adenylcyclase activity by [3
H]8-OH-DPAT in cortical membranes but not in the hippocampus of rats treated with lamotrigine, relative to control animals. This suggests that one mechanism of action of lamotrigine might be the downregulation of responses mediated by cortical 5-HT1A
The role of 5-HT1A
receptors in the mechanism of action of lamotrigine in bipolar disorders has already been examined in healthy human male volunteers51
using the selective 5-HT1A
agonist ipsapirone. Treatment with lamotrigine for 1 week did not significantly modify the hypothermic responses or cortisol responses to ipsapirone. However, that study was limited by the small sample size and the lack of a placebo control, so further investigations are required in humans. Furthermore, it is difficult to conclude anything about the localization of the pharmacologic activity from such studies, as there is considerable heterogeneity between species and technical protocols. For example, the hypothermic response involves only postsynaptic receptors in rats and both presynaptic and postsynaptic receptors in mice.52,53,54
In a previous study, we examined the antidepressant-like activity of lithium, sodium valproate and carbamazepine in the FST using inactive or subactive doses of these drugs in association with the 5-HT1A/B
ligands. The results suggested that 5-HT1B
receptors may play a role in the potentiating effect of lithium in the mouse FST. Lithium demonstrated anti-immobility effects in association with RU 24969 and with anpirtoline.16
Conversely, pretreatment with carbamazepine increased mobility when administered in combination with 8-OH-DPAT and RU 24969, which suggests a role for 5-HT1A
receptors. Sodium valproate in association with 8-OH-DPAT, pindolol and RU 24969 induced antidepressant-like activity in comparison to saline control animals in the FST.
In the present study, strong potentiation of lamotrigine activity with 8-OH-DPAT suggested a 5-HT1A receptor mechanism, but the role of the 5-HT1B receptor was less clear, as anpirtoline, a more specific 5-HT1B receptor agonist, failed to enhance lamotrigine-induced antidepressant-like activity. To define the mechanism further, it would be of interest to antagonize active doses of lamotrigine with NAN 190. The intracortical administration of selective compounds would also help in understanding the possible implication of 5-HT1A/1B receptors.
The results presented here demonstrate that lamotrigine activity is more closely related to valproate and carbamazepine than to lithium, with the advantage of an anti- immobility effect in the mouse FST when administered on its own. These results are consistent with the finding that lamotrigine was more effective in treating bipolar disorders in patients in whom depressive symptoms predominated over mania or hypomania.55