Clinical characteristics of inflammation-associated depression: Monocyte gene expression is age-related in major depressive disorder

Abstract

Increased inflammatory activation might only be present in a subgroup of depressed individuals in which immune processes are especially relevant to disease development. We aimed to analyze demographic, depression, and trauma characteristics of major depressive disorder (MDD) patients with regard to inflammatory monocyte gene expression. Fifty-six naturalistically treated MDD patients (32 ± 12 years) and 57 healthy controls (HC; 31 ± 11 years) were analyzed by the Inventory of Depressive Symptomatology (IDS) and by the Childhood Trauma Questionnaire (CTQ). We determined the expression of 38 inflammatory and immune activation genes including the glucocorticoid receptor (GR)α and GRβ genes in purified CD14⁺ monocytes using quantitative-polymerase chain reaction (RT-qPCR). Monocyte gene expression was age-dependent, particularly in MDD patients. Increased monocyte gene expression and decreased GRα/β ratio were only present in MDD patients aged ⩾28 years. Post hoc analyses of monocyte immune activation in patients <28 years showed two subgroups: a subgroup with a severe course of depression (recurrent type, onset <15 years) – additionally characterized by panic/arousal symptoms and childhood trauma – that had a monocyte gene expression similar to HC, and a second subgroup with a milder course of the disorder (73% first episode depression, onset ⩾15 years) – additionally characterized by the absence of panic symptoms – that exhibited a strongly reduced inflammatory monocyte activation compared to HC. In conclusion, monocyte immune activation was not uniformly raised in MDD patients but was increased only in patients of 28 years and older.

Introduction

Being the leading cause of disability worldwide (WHO, 2012), major depressive disorder (MDD) is a common but clinically and biologically heterogeneous disorder. Although much progress has been made, the pathogenesis of MDD is still largely unknown but may be best described by a complex interplay of social, psychological and biological factors. With regard to biological factors, mounting evidence indicates the involvement of a dysregulated immune response system in the pathogenesis of MDD (Raison and Miller, 2011). In meta-analytic studies, the pro-inflammatory cytokines Tumor Necrosis Factor (TNF)-α and Interleukin (IL)-6 were raised in the serum of patients with MDD (Dowlati et al., 2010), while placebo-controlled randomized studies document that experimentally evoked increases of peripheral pro-inflammatory cytokine levels are followed by the emergence of depression-like symptoms in healthy subjects (Eisenberger et al., 2010, Reichenberg et al., 2001). Furthermore, the inhibition of TNF-α alleviates depressive symptoms in treatment-resistant MDD patients with elevated inflammatory markers (Raison et al., 2013). Also the anti-inflammatory cyclooxygenase-2 (COX-2) inhibitor celecoxib exhibited a therapeutic effect in major depression in a randomized, placebo controlled trial (Muller et al., 2006). Moreover, pro-inflammatory cytokines have been found to interact with many of the pathophysiological domains that characterize depression (Raison et al., 2006).

However, there are numerous cytokine studies that report inconsistent findings. One reason for the heterogeneity might be that cytokine levels are strongly affected by socio-demographic and environmental factors such as age, gender (Pietschmann et al., 2003), exercise (Hayashino et al., 2013), obesity, insulin-resistance (Hotamisligil, 2006), smoking (Rom et al., 2013), as well as by the heterogeneity of major depressive disorder itself, and therefore exhibit high inter-individual variance.

Focusing on the main cellular producers of these cytokines (e.g. microglia, circulating monocytes and macrophages) might be a promising approach to find more stable features for immune activation in MDD. Accordingly, microglia (Beumer et al., 2012) and circulating monocytes have been found to exhibit a pro-inflammatory activation profile with enhanced expressions of immune activation genes in patients with major mental disorders (Bergink et al., 2011, Weigelt et al., 2011, Drexhage et al., 2010), including MDD (Carvalho et al., 2014). In addition, the activated monocyte profile in MDD patients was shown to be accompanied by a decreased ratio of the gene expression of the active glucocorticoid receptor (GR)α over the inactive isoform GRβ (Carvalho et al., 2014). Moreover, GRα/β expression was related to severity of depression and to psychic anxiety in these patients (Carvalho et al., 2014).

Accumulating evidence supports the idea that increased inflammatory markers are only present in a subgroup of depressed individuals in which immune processes are especially relevant to MDD development (Raison and Miller, 2011, Rothermundt et al., 2001) and presumably even treatment (Raison et al., 2013, Carvalho et al., 2013). Still, a valid clinical definition of this subgroup is pending. In this case–control study we determined monocyte inflammatory gene expression profiles with the hypothesis that an increased monocyte immune gene expression only occurs in a subgroup of MDD patients, characterized by either specific demographic characteristics, a specific pattern of depression characteristics and/or a specific form of drug treatment.

Using a heterogeneous and naturalistically treated sample of 56 MDD patients (and 57 age and gender matched healthy controls), we carried out the same quantitative-polymerase chain reaction (qRT-PCR) as in our previous study (Carvalho et al., 2014), using virtually the same panel of 38 inflammatory-related monocyte genes, including the GRα and GRβ genes. Patients’ demographic characteristics, depression course, depression symptomatology, and medication status were analyzed in relation to inflammatory monocyte gene expression. As there is evidence that individuals exposed to childhood trauma exhibit an increased risk of developing both MDD and increased inflammatory markers as adults (Felger and Lotrich, 2013), we also included childhood trauma in our analyses.