A mysterious relationship

Obesity is an emerging pandemic, increasing morbidity and mortality worldwide. Obesity is predominantly a result of an energy-dense diet and a sedentary lifestyle.  

The main clinical consequences of obesity include both abnormalities of the metabolic system (such as hypertension or insulin resistance) and an increased risk of diseases such as cancer. Moreover, obesity has been linked to depression, with both epidemiological and clinical studies demonstrating a positive correlation between these two disorders. Nonetheless, the precise mechanism underlying the relationship between obesity and depression has yet to be explained. 

Although the neuropathophysiology of depression remains unclear, abnormalities in monoamine signalling components, such as serotonin and dopamine, have been implicated in the development of this condition. Clinical observations conducted in the mid-90s suggest that depression results from decreased monoamine function in the brain. Popular treatments for depression target monoamine signalling; however, not all patients benefit from such intervention.  

Hypothalamic region consists of a “leaky” blood-brain-barrier and neurons in this region are in direct contact with circulating molecules and factors

Obese or overweight patients with major depression are at risk of resistance to the antidepressant fluoxetine (Prozac) that targets the monoamine pathway. When compared with patients of normal body weight, overweight and obese patients showed a diminished response to antidepressant treatment. This observation suggests the involvement of alternative pathways for depression in the overweight and obese population. 

Intracellular messenger 
We first set to investigate whether obesity, due to either environmental or genetic reasons, would drive the development of depression. As an environmental model, we used the high-fat diet (HFD) to induce obesity in mice. Conversely, a leptin-deficient mouse, was used to model genetic obesity. To study depression-related behaviour in mice, we employed commonly used behavioural depression paradigms (forced swim and tail suspension test). Interestingly, we found that in both mouse models of obesity, the development of obesity led to the development of depression-like behaviour. 

The neurocircuitry of depression is complex and involves multiple regions of the brain, such as hippocampus, amygdala, thalamus, cortex and hypothalamus. The hypothalamus is the master regulator of energy homeostasis that has also been implicated in depression. Given the development of the depression-like phenotype upon the consumption of a high-fat diet in mice, we performed a microarray analysis to determine which molecular signalling pathways are affected. Interestingly, we found that the most affected pathway is the protein kinase A (PKA) pathway. PKA is the primary effector of cAMP, a ubiquitous intracellular second messenger. Signalling via cAMP appears to play a key role in the pathophysiology and pharmacology of depression. It is believed the mechanism of antidepressant treatments involve adaptations of the cAMP signalling. Adenylyl cyclases are the enzymes that produce cAMP, and our microarray data revealed the hypothalamic adenylyl cyclase expression was downregulated upon the consumption of a HFD. The reduced PKA phosphorylation upon the consumption of the HFD was confirmed by protein analysis in hypothalamic extracts. 

In depression, signaling via cAMP may be impaired by cyclic nucleotide phosphodiesterases (PDEs), which provide the sole route for cAMP degradation in cells. Members of the PDE4 gene family play a major role in regulating cognition and depressive disorders. Interestingly, rolipram, a known selective PDE4 inhibitor, exhibits antidepressant action in mice. However, the therapeutic potential of rolipram is limited by adverse side effects including nausea and vomiting. Identifying the specific PDE4 isoform that mediates the antidepressant action of rolipram could enable the development of selective inhibitors to offer therapeutic effects with minimal adverse side effects.  

In both our models of obesity-induced depression phenotype, we found that the mRNA levels of PDE4A5 isoform were specifically upregulated in the hypothalamus of mice. The RNA expression analysis was in agreement with the protein analysis where the consumption of a HFD led to increased total and PKA-phosphorylated PDE4A5 levels. PKA phosphorylation of PDE4A5 has been shown to elicit its activation, which serves as a critical feedback loop by engendering increased cAMP degradation. 
 
Given the potential importance of PDE4A, we next employed PDE4A-/- mice and assessed whether the lack of this gene would play a protective role in the development of depression upon dietary or genetically induced obesity. Interestingly, we found that genetic loss of the PDE4A gene can prevent both the dietary and genetic-induced depression phenotype in mice. That was independent of the increased body weight as both PDE4A-/- as well as their litter mate control mice gained similar body weight in both mouse models. 

Phosphodiesterases are enzymes that interact with scaffolding proteins and results in their compartmentalisation and subcellular localisation. Next, we studied the subcellular regulation of PDE4 upon the consumption of the HFD. Phosphodiesterase assays revealed that PDE4 activity was greater at the membrane fraction of mice fed a HFD compared to the control diet, and that increase was abolished in the PDE4A-/- mice. This finding suggests that the membrane-associated PDE4A, namely PDE4A5, is the functionally relevant PDE4A isoform whose activity is upregulated in the hypothalamus after the consumption of a HFD. We also analyzed the PDE4 activity in other brain regions, including the amygdala, but no statistical differences were noted. These findings suggest that the hypothalamus is a key locus affected by the consumption of an HFD in obesity-induced depression via the upregulation of PDE4 activity. 

The acid test 
Next, we hypothesized that dietary fatty acids might play pivotal roles as molecular transducers of cell signalling in the hypothalamus to regulate depression. In support of this hypothesis, it has been previously shown that the central melanocortin pathway, which is the most important neuronal pathway involved in human obesity, is regulated by dietary fatty acids. More specifically, to convey intracellular signalling pathways in energy expenditure and appetite regulation.  

In our study, fatty acid analysis from mice fed a HFD compared to mice on control diet revealed that hypothalamic extracts from mice on the HFD had higher concentrations of both saturated and unsaturated fatty acids. The increased fatty acid concentration in the hypothalamus was region-specific; analysis of cortical samples from the same mice revealed no difference in the fatty acid concentration from either diet. This agrees with previous work showing that the hypothalamic region consists of a “leaky” blood-brain-barrier and neurons in this region are in direct contact with circulating molecules and factors. 

The effect of diets rich either in saturated or unsaturated fatty acids on depression have been linked in multiple studies, where the consumption of a western diet positively correlates with depression. Western diets are poor in essential nutrients for the brain, such as ω-3 polyunsaturated fatty acids (PUFAs) and notorious for their high levels of saturated fatty acids and trans-fat. Adherence to a Mediterranean diet rich in unsaturated fatty acids and olive oil has been suggested to play a protective role in the development of depression. It is believed that PUFAs increase the fluidity of the neuronal membrane, and this allows more frequent access of the neurotransmitters to their receptors. On the contrary, the presence of high levels of saturated fatty acids is believed to make the neuronal membranes more rigid, resulting in less frequent access.  

Given the differential effect of saturated and unsaturated fatty acids, we next decided to test the direct role of both palmitic acid, a saturated fatty acid rich in western diets and oleic acid, a major component of olive oil in regulating PKA signalling. To do so, we used fluorescence resonance energy transfer (FRET)-based PKA biosensor, which enable quantitative, real-time detection of rapid changes in PKA activity. We found that palmitic acid interfered with the PKA pathway activation in vitro (it disabled forskolin to increase PKA activity). On the contrary, oleic acid treatment did not affect the forskolin-induced PKA activation. 

Free fatty acids bind to their free fatty acid receptors to elicit their action. There are four main free fatty acid receptor (FFAR) divisions. According to the length and saturation of fatty acids, they bind to FFAR1-4. RNA expression analysis in our study revealed that FFAR1 (also known as GFP40) is specifically upregulated in the hypothalamus in both models of obesity-induced depression. Interestingly, FFAR1 binds medium and long chain fatty acids and is the main receptor for palmitic acid. Lastly, using protein analysis, we found that PDE4A5 interacts with FFAR1 at the membrane fraction of cells in vitro, and this interaction was time-dependent on the palmitic acid treatment. 

A complex brew 
In summary, we found that both dietary and genetically induced obesity in mice leads to depression. Findings from our study suggest that the consumption of a HFD regulates the cAMP/PKA signalling pathway in the hypothalamus and may be responsible for the development of depression due to obesity. Furthermore, we found that the consumption of a HFD leads to increased PDE4 activity at the membrane subcellular fraction of the hypothalamus. Moreover, we identified that loss of PDE4A could prevent both dietary and genetically induced depression-like behaviour in mice independent of body weight. From the PDE4A isoforms, the membrane-associated isoform PDE4A5 may be responsible for this phenomenon.  

In addition, we found that the consumption of a HFD leads to an influx of dietary fatty acids, specifically in the hypothalamus. These fatty acids can directly and differentially modulate the PKA signalling. Furthermore, we also found that the consumption of a HFD increases specifically FFAR1 expression in the hypothalamus, which in turn can lead to a potential association of FFAR1 and PDE4A5. The translocation of PDE4A5 to the membrane fraction will lead to a re-programming of the pattern of compartmentalisation of the cAMP signaling pathway in these cells. Namely, consumption of a HFD leads to an influx of dietary fatty acids specifically in the hypothalamus. The entrance of dietary palmitic acid in the brain may bind to FFAR1 to suppress the PKA pathway via the upregulation of PDE4A5 activity. These findings suggest that the influx of saturated fatty acids due to the consumption of a HFD can alter the cAMP/PKA signalling and result in the development of depression. 

Author: Dr Eirini Vagena is American Diabetes Association?Postdoctoral Fellow at the Diabetes Center of University of California, San Francisco.