25 April 2008. Selective serotonin reuptake inhibitors such as fluoxetine (better known as Prozac) have always been a bit of a puzzle. Though they quickly elevate serotonin levels, it can take weeks for these drugs to have a noticeable antidepressant effect. Research from the past decade suggests that these drugs have a much more profound and gradual mode of action, possibly related to increased synthesis of new neurons and new neural connections.

In last week’s Science, researchers led by Jose Maya Vetencourt and Lamberto Maffei at the Institute for Neuroscience, National Research Council, Pisa, Italy, collaborating with Eero Castrén's group at the University of Helsinki, reported that antidepressants do, indeed, induce changes in neuronal connections and that these neural rearrangements are functionally significant. They found that antidepressants can promote restructuring of the mammalian visual cortex, turning an eye with poorly developed neural connections into a well-connected, useful one. Beyond the immediate hope for treating amblyopia, or “lazy eye,” the study offers new insights into the effects of antidepressants and the treatment of mood disorders. Further, because the benefits of the drug depended upon visual stimulation, the study supports a synergistic interplay of the drugs and environmental factors, which might include behavioral or psychotherapy in the case of mood disorders.

These results may be of interest to schizophrenia researchers for several other reasons: at a mechanistic level, Maya Vetencourt and colleagues found that brain-derived neurotrophic factor (BDNF) and γ-aminobutyric acid (GABA), both molecules of interest for schizophrenia researchers, are involved in the restoration of adult neuronal plasticity.

Visual cortical plasticity—a classic neurobiological model
In young mammals, covering one eye leads to a shift in the neuronal circuitry of the visual cortex to favor the other eye. This trick can even be used to correct visual problems. When one eye is weaker than the other, covering the stronger eye (monocular deprivation) shifts the visual circuitry until visual acuity in the weaker eye is restored. Usually this only works during a time when the circuitry is “plastic” or easily rewired, as in children. In adults, covering the good eye usually has no effect on the weaker one. That situation changes if fluoxetine enters the picture, according to Maya Vetencourt and colleagues.

Using a monocular deprivation (MD) model in rats, the researchers found that visual acuity is restored in adult rats if they are chronically treated with fluoxetine. To measure visual acuity, the researchers relied on evoked electrical activity in the visual cortex. In adult amblyopic rats, uncovering the deprived eye during the last two weeks of chronic fluoxetine treatment resulted in a complete recovery of visual evoked potentials (VEPs) in the corresponding visual cortex of the brain. In contrast, there was no restoration of visual acuity in animals that did not receive the antidepressant. Behavioral testing also revealed the restoration of visual acuity to the deprived eye, and the fluoxetine treatment also restored binocular vision, as judged by the ratio of electrical activity in the right and left visual cortices. “Our findings demonstrate that chronic fluoxetine administration reinstates a juvenile-like form of OD [optical dominance] plasticity in adulthood, which is indicated by a decrease in the response to stimulation of the deprived eye and promotes a complete recovery of visual function in adult amblyopic rats,” write the authors.

Neurotransmitters and neurotrophins
How does this antidepressant have such a dramatic effect on the visual cortex? One possibility is that it relieves some restraint on neural reorganization. For example, it is believed that inhibitory neurons put an end to the plasticity of the visual system during postnatal development. To test if these inhibitory influences may have been quashed by the antidepressant, the researchers measured levels of GABA (the major inhibitory neurotransmitter). They found that baseline levels of GABA were significantly reduced in fluoxetine-treated animals, and that long-term potentiation (LTP), a form of neural plasticity not normally found in the adult visual cortex, was restored. The findings indicate that it is the GABA inhibitory neurons that prevent the adult visual system from undergoing major changes in circuitry. This is supported by the fact that diazepam, a GABA agonist, prevents the reorganization of visual circuitry in response to fluoxetine.

Relief from GABA may not be the only way antidepressants have such a dramatic effect on the visual system. The authors also confirmed that BDNF levels are significantly elevated by fluoxetine and that simply injecting the trophin into the cortex is sufficient to start shifting the circuitry in response to monocular deprivation. This observation also fits with some earlier work from these researchers that showed environmental enrichment (EE) restores visual acuity in amblyopic rats (see Sale et al., 2007)—EE is known to elevate brain BDNF levels (see Adlard et al., 2005). The researchers conclude that the combination of reduced GABA inhibition and increased BDNF expression induced by chronic fluoxetine somehow alters the genes that regulate plasticity, allowing a functional modification of neuronal circuitries. They suggest that this side effect of antidepressant treatment might serve as a means to correct amblyopia in humans.

“Our data also indicate a potential clinical application for antidepressants in neurological disorders in which synaptic plasticity is compromised because of excessive intracortical inhibition,” write the authors. In the case of schizophrenia, however, too little intracortical inhibition may be the case, and deficits in GABA transmission are specifically implicated (see SRF related news story). The relevance of this work for schizophrenia at the molecular level is also worth considering. In addition to GABA dysfunction, there are indications that levels of both BDNF (see Durany et al., 2001) and BDNF-containing neurons are reduced in the brains of people with schizophrenia (see Iritani et al., 2003). However, it is unlikely that the answer will be as simple as boosting BDNF levels in any psychiatric disorder. For example, there are indications that too much BDNF in other areas of the brain may be detrimental to proper brain functions (see SRF related news story).—Tom Fagan.

Reference:
Maya Vetencourt JF, Sale A, Viegi A, Baroncelli L, De Pasquale R, O’Leary OF, Castren E, Maffei L. The antidepressant fluoxetine restores plasticity in the adult visual cortex. Science April 18, 2007;320:385-388. Abstract

Q&A with Eero Castrén. Questions by Schizophrenia Research Forum.

Q: What prompted you to do this study?
A: The clinical effects of antidepressants appear with a delay, and increased neuronal plasticity and alterations in network structure have been suggested to underlie this delay (Castrén, 2005). However, a direct demonstration that antidepressants would functionally alter neuronal networks has been lacking.

Q: How did you tackle this problem?
A: We used the visual cortex, a classical model structure for developmental plasticity, to examine the effects of fluoxetine on cortical network rearrangements in rats. Our results demonstrate that fluoxetine reactivates the critical period plasticity in the adult visual cortex and facilitates functional recovery of miswired neuronal networks.

Q: Can you give us some background on this model?
A: Plasticity in the visual cortex is high during a critical period of postnatal development, when inputs from the two eyes compete for the innervation of the visual cortex and the functional networks are laid down (Hensch, 2005). If one eye is deprived of vision during the critical period, the inputs of the active open eye occupy the visual cortex and those from the closed eye are withdrawn, leaving the eye poor in vision, a condition known as amblyopia. Opening the eye and patching of the better eye during the critical period allows the weaker eye recovery of its connectivity and visual acuity. In adulthood, after the end of the critical period, plasticity is more restricted and closure of one eye no longer leads to the loss of innervation; conversely, an amblyopic eye not treated with patching of the better eye during the critical period remains poor in vision.

Q: So what was the main finding?
A: We found that peroral fluoxetine treatment for three weeks reopens the critical period plasticity in the adult visual cortex and closure of one eye leads to the shift in the cortical innervation to favor that of the open eye. Furthermore, if an amblyopic eye covered during the critical period is reopened in adulthood, vision improved in the weaker eye to finally equal that of the healthy eye when fluoxetine treatment was combined with covering the better eye.

Q: What is the basis for this renewed plasticity?
A: The enhanced plasticity was associated with increased BDNF expression in the visual cortex, and infusion of BDNF into visual cortex mimicked the effects of fluoxetine. Intracortical inhibition, which is known to regulate critical period plasticity, was reduced and increasing inhibition by intracortical infusion of diazepam blocked the effects of fluoxetine on plasticity. Thus, BDNF signaling and cortical GABAergic networks play a critical role in the mechanisms through which antidepressants facilitate cortical plasticity.

Q: What does this work tell us about the action of antidepressants?
A: These experiments convincingly demonstrate that antidepressants induce plasticity in the visual cortex and facilitate functional rearrangements leading to a recovery of a faulty network miswired due to imbalanced environmental stimuli during the postnatal development.

Q: Does this have any relevance to human conditions?
A: Antidepressants may have similar effects also in the human visual cortex. Normann et al. showed recently that while neuronal plasticity in the visual cortex of depressed patients was reduced, treatment of healthy volunteers with another SSRI, sertraline, enhanced plasticity over that seen in untreated controls (Normann et al., 2007).

It is important to note, however, that fluoxetine did not repair the network on its own; it merely enhanced plasticity to facilitate the ability of environmental cues to guide the rearrangement of the connectivity. Maffei’s lab has previously shown that raising rats in an enriched environment has effects equivalent to those of fluoxetine: an enriched environment enhances plasticity in adult visual cortex and allowed a recovery of an amblyopic eye in adults (Sale et al., 2007).

Q: What about other areas of the cortex, such as those involved in mood regulation. Is it possible that fluoxetine and other antidepressants might have analogous effects there?
A: That remains to be seen. If this was the case, antidepressants, through BDNF signaling, might facilitate the recovery of dysfunctional cortical networks miswired by disruptive early life experiences or perhaps during extended stress. However, in this case, too, beneficial environmental cues, such as rehabilitation or talk therapy, would be required to guide rearranging networks for functional recovery, which is consistent with the observations that antidepressants and psychotherapy together work better than either one alone.

Q: Is there any reason for clinicians to be concerned about the use of antidepressants in schizophrenia, which is pretty common? A prominent body of work in schizophrenia revolves around the idea of GABA hypofunction in prefrontal cortex. If fluoxetine further depresses GABA function, then might not these drugs counteract any psychotherapeutic interventions?
A: For the schizophrenia, it is true that the GABAergic system is implicated; however, this system is multifaceted and complicated. I would be very cautious about rushing into any direct conclusions based on our study. On the contrary, there are decades of clinical experience about using antidepressants in schizophrenics—if something alarming would happen, that would be known by now.

Incidentally, there is an interesting review from the McGuire lab in Lancet (Fusar-Poli et al., 2007) reviewing data that suggest that antidepressant treatment might protect against the first episode of schizophrenia in at-risk people. Whether this has anything to do with the plasticity found in our study remains to be seen. However, it seems to me that it is not useful, and might even be harmful, to strictly label drugs into slots such as "antidepressants" and "antipsychotics," etc. It should be evident already (they are useful for a variety of neuropsychiatric disorders), and further emphasized by our study, that antidepressants do something very different than simply counteracting depression. A fresh look is what is needed.