Fast-acting antidepressants: are we nearly there?

In a recent editorial, RS Duman from Yale University (CT, USA) wrote that a rapid-acting antidepressant “has been the holy grail of drug development for mood disorders for many years, as the action of typical antidepressants requires weeks or even months for a beneficial response” [1]. This ‘quest’ still remains an important challenge in the field of neuropsychopharmacological research. Depression is known as the most common of psychiatric illnesses, with an estimated prevalence of 15–20% within the general population. Such numbers themselves justify the need for a more rapid treatment, as a matter of comfort for millions of patients. In addition, the delayed onset of action of antidepressants becomes a critical factor in the particular cases of major/severe depressions, in which the risks of suicide are strongly increased.

Historically, the vast majority of the antidepressant molecules that have been developed were aimed at increasing extracellular levels of biogenic amines within the brain. This strategy originates from the initial finding that the main biological action of tricyclics, the first generation of effective antidepressants, consists of a potent inhibition of serotonin (5-hydroxytryptamine [5-HT]) and norepinephrine (NE) reuptake. The subsequent development of monoamine oxidase inhibitors (MAO-Is) was based on a similar approach, namely an indirect elevation of amines extracellular concentration, related to the blockade of their catabolism. More recently, it became apparent that, among the biogenic amines, 5-HT plays a crucial role in the mechanism of action of antidepressants. Indeed, although the involvement of other aminergic systems in the pathophysiology of depression is certainly non-negligible (in particular when considering the arguments for a ‘dopaminergic’ theory of depression), it remains that the common feature of all molecules used to date, including the recently developed selective NE reuptake inhibitors, is their ability to increase the central 5-HT neurotransmission [2,3]. In particular, it appears that an enhanced stimulation of the 5-HT_1A receptor type within limbic areas contributes to the beneficial effect of the treatment [2,4]. Importantly, such features appear to be selective of antidepressant molecules, as they have not been observed with other pyschotropes, such as anxiolytics or antipsychotics for instance [2]. In keeping with these observations, selective serotonin reuptake inhibitors (SSRIs) now constitute the most commonly used class of antidepressants, for more than 15 years. Numerous animal studies have been conducted with these compounds, leading to a better understanding of the mechanisms underlying their delayed onset of action. Thus, it is now well known that they initially induce a strong decrease of 5-HT neuronal activity due to an overstimulation of the inhibitory somatodendritic 5-HT_1A autoreceptors [2]. A sustained administration of 2–3 weeks is necessary before 5-HT neurons recover their normal activity, when these autoreceptors become fully desensitized [2].

More recently, new strategies of research have emerged, for the purpose of ‘bypassing’ this 5-HT-related inertia. They were oriented toward assessing the effects of antidepressants beyond the postsynaptic 5-HT receptor level, searching for the cellular/molecular mechanisms that were affected by the treatment. An extensive and seminal work has been conducted in this field for more than a decade, in particular by the groups of EJ Nestler and RS Duman. These studies permitted the isolation of a number of factors that are deeply modified after a sustained antidepressant treatment. Among these, the most noticeable are an elevation/activation of the transcription factor CREB [5] and an enhanced adult neurogenesis [6], the latter being apparently consequent to the former. Changes occur mainly in limbic areas, such as the hippocampus, where modifications of expression of the growth factor brain-derived neurotrophic factor were also evidenced. Altogether, these observations led to the general idea that antidepressants share the common property to positively modulate cellular growth and plasticity in mood-related brain areas. As such, they would act in an opposite manner to stressful experiences, which are also known to favor the emergence of depression in humans. It can be expected that the development of therapeutic strategies able to directly stimulate the above factors would allow a more rapid onset of action than the currently used antidepressants. However, it would also raise the critical question of the regional and/or functional specificity of such treatments. For instance, CREB is a ubiquitous transcription factor, which when activated in rodents, depending on the brain area addressed, can induce opposite effects on various parameters thought to reflect depression in humans [5]. Similarly, an enhancement of adult brain neurogenesis can also have detrimental effects, notably in the general context of epilepsy and seizures [7].

It appears, therefore, that an approach able to combine the regional specificity of the ‘serotonergic’ strategy with the efficacy of the cellular and molecular modulators of neural plasticity could constitute a promising avenue of research. A relevant question in this context is the following: would a rapidly achieved elevation of central 5-HT neurotransmission be able to also induce a rapid onset of neurogenesis and CREB activation? In other words, is the delayed efficacy of typical antidepressants due to the fact that tissue growth requires long-term processes when induced by ‘normal’ mechanisms of neurotransmission, or is it simply related to the indirect, passive action they exert on 5-HT neurons? Indeed, as previously mentioned, SSRIs, tricyclics or MAO-Is do not act directly on 5-HT activity, but rather indirectly increase 5-HT extracellular levels by inhibiting the inactivation/degradation of the transmitter. Consequently, their efficacy is conditioned by the degree of sensitivity of 5-HT_1A autoreceptors, which plays a determinant role in the modulation of 5-HT neuron firing rate (see previously). Thus, instead of trying to ‘bypass’ the 5-HT system as such, one possible strategy could consist of bypassing the 5-HT _1A-mediated homologous control. Obviously, the simplest way to do so would be to find a treatment enhancing both rapidly and steadily the firing activity of 5-HT neurons. On the basis of previous researches conducted in our laboratory, we knew that 5-HT₄ receptor agonists are precisely able to elicit such an effect. We decided, therefore, to test their ability to activate CREB function and neurogenesis in the adult rat hippocampus, after a short-term (3 days) continuous administration. The obtained results showed that such a treatment is indeed effective, with an intensity at least equal, or even superior, to what is observed after 2 weeks with a SSRI [8]. These data were further confirmed by the use of behavioral models of depression known to respond to a long-term (2–3 weeks), but not an acute, administration of typical antidepressants. Again, 5-HT₄ agonists acted four to seven-times more rapidly than SSRIs to reverse the depression-like symptoms [8]. Importantly, these promising results have been recently replicated by using other classes of molecules, also able to directly influence 5-HT electrical activity. Thus, the continuous stimulation of σ-1 receptors for 3 days, which augments 5-HT neuron firing rate by approximately 50% [9], also augments hippocampal neurogenesis [Lucas G et al. (2007), Submitted]. There are also a few reports indicating that the blockade of 5-HT₇ receptors allows a fast recovery of 5-HT impulse flow in the presence of an SSRI, and that this is accompanied by a more rapid activation of CREB and neurogenesis in the hippocampus.

Can we now, therefore, state that we are nearly at the era of fast-acting antidepressants? There are a number of reasons to dampen this enthusiastic proposition. First, the results mentioned previously are very recent, some of them not yet published. They were obtained by conducting preclinical studies in rodents, and no clinical trials have been performed to confirm the potential interest of these new approaches. Second, the use of agonists, in other words, compounds acting directly on a certain type of receptors, may have important consequences in terms of possible peripheral side effects. This is particularly relevant when considering that a subpopulation of 5-HT₄ receptors is located within the heart atrium, where their role and function is not yet fully established. Third, and in a more fundamental point of view, there is still a ‘missing link’ in the literature to relate the effects of 5-HT receptor stimulation with the molecular effectors of brain growth and plasticity. Indeed, although it is known that the selective stimulation of postsynaptic 5-HT _1A receptors has a facilitatory effect on hippocampal neurogenesis [4], it remains that these receptors have been described to be negatively coupled to adenylate cyclase (Ac), and should normally reduce the activation of CREB and, in turn, of neurogenesis (discussed in [1]). On the other hand, 5-HT₄ receptors are positively coupled to Ac, but 5-HT₄ agonists appear to exert their effects in the hippocampus indirectly, via an enhanced stimulation of 5-HT_1A receptors consecutive to the elevation of 5-HT extracellular levels [8]. Therefore, complete and systematic studies have to be performed in the future to reconcile these sparse data, and to better understand the complex mechanics, which is located downstream of the membrane 5-HT receptor, but upstream of transcription factors such as CREB, and that regulates their coupling. Such a strategy would open the exciting perspective of associating proteomics with depression research; this could allow the discovery of very specific pathways linking the ‘pharmacological’ and the ‘molecular’ faces of antidepressants. With the exponential progresses made in proteomics during the last few years, we should nearly be at this stage.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.