Studies examining cerebral glucose metabolism in MDD using
Studies examining cerebral glucose metabolism in MDD using 18 F-fluorodeoxyglucose in positron emission tomography (FDG-PET) showed altered regional glucose metabolism in MDD (Dunn et al., 2002, Kennedy et al., 2001, Kimbrell et al., 2002, Saxena et al., 2001). In a recent meta-analysis by Su and colleagues, ten studies using FDG-PET were analyzed. The results showed that in MDD patients the cluster size of reduced glucose metabolism was higher than the cluster size of increased glucose metabolism (Su et al., 2014). Our results of increased GLUT1 methylation in MDD may underscore these results by suggesting that altered expression of GLUT1 may be in part responsible for altered atp citrate lyase glucose metabolism in depression. Disturbance in the brain energy metabolism has been linked to the development of type-2 diabetes mellitus and obesity. According to the “selfish brain theory” the brain is the main consumer of glucose in the body, spending around 130 g glucose per day in normal weight individuals (Peters et al., 2004). Under stressful conditions, global brain glucose uptake increases (Hitze et al., 2010). Given an altered glucose metabolism during depression as indicated by the meta-analysis by Su et al. (2014), activation of the HPAA and subsequent hypercortisolism may be interpreted as a compensating mechanism to redistribute glucose to the brain via impairment of insulin action. Furthermore, the observed enhancement of pro-inflammatory cytokines may be put in this hypothesis. Tumor-necrosis factor-α is a long discussed factor antagonizing insulin action by interrupting insulin-stimulated tyrosine phosphorylation (Feinstein et al., 1993). Furthermore, TNF-α and related inflammatory cytokines have also been discussed in the context of cerebral insulin resistance in Alzheimer's disease (Clark et al., 2012, Liu et al., 2011). MDD has been linked to the development of insulin resistance and type-2 diabetes mellitus (T2DM), documented by the nearly doubled prevalence of T2DM in depressed patients (Kan et al., 2013, Mezuk et al., 2008). MDD and T2DM have been shown to be independently associated with a greater risk for dementia, and the combined association of both disorders confer a stronger risk for the development of all-cause dementia than each disorder alone (Katon et al., 2015). Interestingly, T2DM and Alzheimer's dementia have been associated with alterations in brain glucose metabolism, and with altered expression of brain glucose transporters (Shah et al., 2012). Our finding of an altered GLUT1 methylation may point to a common pathogenetic origin in the development of these disorders. Smoking was higher in remitters compared to non-remitters, although this difference did not reach statistical significance. To date, little is known on the influence of smoking on glucose transporters in the brain. In a recent study GLUT1 function and expression were examined in a focal brain ischemia mouse model. Pre-exposure to ischemia with nicotine resulted in reduced GLUT1 function and expression. The authors conclude that nicotine may be a vulnerability factor for enhanced stroke injury (Shah et al., 2015). However, it is unclear whether different smoking patterns between remitters and non-remitters observed in our study may be sufficient to differentially influence GLUT1 methylation. GLUT1 expression may be upregulated by psychopharmacologically active drugs. In a recent study by Nagai and colleagues, upregulation of GLUT1 was observed after administration of fluoxetine and pergolide (Nagai et al., 2014). Furthermore, valproic acid was shown to upregulate GLUT1 in cultured brain astrocytes in GLUT1 heterozygous mice (Kim et al., 2013). However, remitters and non-remitters were not different in drug treatment. Methylation of GLUT4, the key glucose transporter for peripheral glucose homeostasis, was not different between MDD and CG, and between remitted and non-remitted patients. However, patients with MDD had higher levels of cortisol, insulin, relative insulin resistance and pro-inflammatory cytokines. These results are in line with former studies showing endocrine and immune dysfunctions in MDD. In the context of altered GLUT1 methylation and subsequently altered brain glucose metabolism in MDD, our findings may be interpreted as a mechanism to counteract impeded glucose transport to the brain. According to the selfish brain theory, which was put into the scene by Peters and colleagues ten years ago, the brain priorizes its energy demands and activates the HPAA and autonomous system during states of current or anticipated central glucopenia. According to this theory, HPAA upregulation with subsequent hypercortisolism, activation of the sympathoadrenergic system and the upregulation of pro-inflammatory cytokines may be the result of impeded central energy metabolism (Peters et al., 2004).