Differential induction of FosB isoforms throughout the brain by fluoxetine and chronic stress

Highlights

• Fluoxetine induces FosB gene products in ∼25 distinct regions of adult mouse brain.

• Patterns of FosB gene products induced by fluoxetine differ by brain region.

• FosB gene product induction by fluoxetine differs from induction by stress.

• Only ΔFosB levels in PFC and nucleus accumbens correlate with behavioral response to stress.

Abstract

Major depressive disorder is thought to arise in part from dysfunction of the brain's “reward circuitry”, consisting of the mesolimbic dopamine system and the glutamatergic and neuromodulatory inputs onto this system. Both chronic stress and antidepressant treatment regulate gene transcription in many of the brain regions that make up these circuits, but the exact nature of the transcription factors and target genes involved in these processes remain unclear. Here, we demonstrate induction of the FosB family of transcription factors in ∼25 distinct regions of adult mouse brain, including many parts of the reward circuitry, by chronic exposure to the antidepressant fluoxetine. We further uncover specific patterns of FosB gene product expression (i.e., differential expression of full-length FosB, ΔFosB, and Δ2ΔFosB) in brain regions associated with depression – the nucleus accumbens (NAc), prefrontal cortex (PFC), and hippocampus – in response to chronic fluoxetine treatment, and contrast these patterns with differential induction of FosB isoforms in the chronic social defeat stress model of depression with and without fluoxetine treatment. We find that chronic fluoxetine, in contrast to stress, causes induction of the unstable full-length FosB isoform in the NAc, PFC, and hippocampus even 24 h following the final injection, indicating that these brain regions may undergo chronic activation when fluoxetine is on board, even in the absence of stress. We also find that only the stable ΔFosB isoform correlates with behavioral responses to stress. These data suggest that NAc, PFC, and hippocampus may present useful targets for directed intervention in mood disorders (ie, brain stimulation or gene therapy), and that determining the gene targets of FosB-mediated transcription in these brain regions in response to fluoxetine may yield novel inroads for pharmaceutical intervention in depressive disorders.

Introduction

Approximately one in five Americans will experience a depressive disorder within their lifetime (Kessler et al., 2005), and only about 50% of those will fully respond to available treatments (Culpepper, 2010). In recent decades, it has become increasingly clear that the variability in depressive disorders and in the response to existing treatments stems, at least in part, from individual differences in both genetics (Flint and Kendler, 2014) and gene expression (Vialou et al., 2013). Our group and many others suggest that chronic stress or exposure to antidepressants causes induction and altered function of transcription factors which, in turn, modulate gene expression to alter mood and behavior (Vialou et al., 2013). We therefore propose that determining: 1) the specific brain regions, 2) the specific transcription factors, and 3) the specific target genes involved in these processes could potentially lead to the development of novel therapeutic approaches for the treatment and prevention of depressive disorders. Previous studies indicate that the transcription factor ΔFosB is induced by both chronic stress and chronic antidepressant treatment, and that it plays an essential role in mouse models of depression and antidepressant action (Perrotti et al., 2004, Vialou et al., 2010, Ohnishi et al., 2011, Lobo et al., 2013, Robison et al., 2014, Vialou et al., 2014).

The brain's reward circuitry centers on dopaminergic neurons in the ventral tegmental area (VTA) and their mesolimbic projections to nucleus accumbens (NAc), dorsal striatum, amygdala, and hippocampus, as well as their mesocortical projections, in particular to the prefrontal cortex (PFC). Previous studies of the role of ΔFosB in stress responses and antidepressant action have focused on discrete brain regions within the reward circuitry, namely the NAc (Vialou et al., 2010, Robison et al., 2014) and the PFC (Vialou et al., 2014). However, other forms of stimulation, such as chronic exposure to various drugs of abuse, are known to induce ΔFosB in many additional brain regions, both within and outside the reward circuitry (Perrotti et al., 2008). Moreover, the FosB message undergoes complex splicing resulting in multiple proteins, the most prominent of which have apparent molecular weights of 50 kDa (full-length FosB), 35–37 kDa (ΔFosB), and 25 kDa (Δ2ΔFosB) that may have distinct transcriptional target genes and may play distinct roles in behavioral responses (Ohnishi et al., 2011). For instance, ΔFosB appears to drive spontaneous locomotor activity and accumulation of E-cadherin and phospho-Akt, while full-length FosB is essential for stress tolerance (Ohnishi et al., 2011). Though the expression of the distinct gene products in response to stress and various drugs has been examined extensively in NAc and to some degree in PFC (Perrotti et al., 2004), expression of the different FosB gene products outside the NAc in mouse models of depression and antidepressant action has not been reported in detail. Here, we demonstrate that chronic treatment with the antidepressant fluoxetine induces FosB gene products in more than 25 distinct regions of the mouse brain, and that the presence of specific FosB splice products varies in select regions of the reward circuitry under both stress and antidepressant conditions. We chose to focus on NAc and PFC because our previous studies demonstrate the behavioral importance of FosB in these regions (Vialou et al., 2010, Vialou et al., 2014), and we chose hippocampus because is both modulates the function of the mesolimbic and mesocortical dopamine system and has been directly implicated in many studies of depression, both in humans and in pre-clinical models (Duman and Aghajanian, 2012, Eisch and Petrik, 2012).