A systematic review of in vitro cytokine production in eating disorders

Highlights

• All 12 in vitro studies assessed in vitro cytokine production in anorexia nervosa.

• Only one of these studies included people with bulimia nervosa.

• Applied methods and findings of cross-sectional studies were heterogenous.

• Dietary interventions may alter in vitro cytokine production in anorexia nervosa.

Abstract

Background

Eating disorders (EDs) have been associated with alterations in cytokine concentrations and production. This review examines whether in vitro cytokine production (i) is altered in people with EDs compared to healthy participants; and (ii) changes in response to treatment?

Methods

Using PRISMA guidelines, we systematically reviewed articles reporting group comparisons or longitudinal assessments of spontaneous and/or stimulated cytokine production in vitro in people with EDs.

Results

Twelve studies were included. Cross-sectional results were mixed in anorexia nervosa. Only one study measured cytokine production in bulimia nervosa. Two longitudinal studies showed that daily yoghurt consumption increases phytohemagglutinin-stimulated interferon-γ production in anorexia nervosa.

Conclusion

The mixed results could be accounted for by variations in experimental design. Our findings suggest that cytokine production could possibly be modulated through dietary interventions. However, due to the methodological heterogeneity and shortcomings of the included studies, it seems unreasonable to draw further conclusions.

Introduction

Eating disorders (EDs), including anorexia nervosa (AN), bulimia nervosa (BN) and binge-eating disorder (BED), are characterised by disturbances in eating behaviours and body image. The aetiology of EDs is complex and the underlying pathophysiological mechanisms contributing to their development and maintenance are unclear. Alterations in immunological function and more specifically, the production of cytokines, such as tumor necrosis factor (TNF)-α and interleukin (IL)-6, have been implicated as a possible contributing factor. For example, recent meta-analyses have shown elevated levels of certain pro-inflammatory cytokines including TNF-α and IL-6 (Dalton et al., 2018; Solmi et al., 2015), in people with AN compared to healthy individuals. However, those with BN did not significantly differ in concentrations of pro-inflammatory cytokines in comparison to healthy volunteers and no research in BED was available to be incorporated into the meta-analyses (Dalton et al., 2018).

Cytokines are soluble intercellular signalling proteins that are produced by a range of cells, including microglia and astrocytes, in both the brain and in the periphery (Lichtblau et al., 2013). They have particular importance in the immune system, but also in brain functioning (for a review see Capuron and Miller, 2011). It has been well documented that cytokines are involved in the regulation of appetite and feeding. Multiple cytokines have been shown to have inhibitory effects on food intake (Buchanan and Johnson, 2007; Plata-Salaman, 2001; Wong and Pinkney, 2004). For example, reductions in food intake have been observed following peripheral and/or central administration of IL-1β (Langhans et al., 1993) and TNF-α (Bodnar et al., 1989) in animal models. In addition, in a validated animal model of binge-like eating behaviour, in which cycles of restriction are combined with frustration stress, down-regulation of the anorexigenic IL-18 system was observed (Alboni et al., 2017). The impact on appetite and feeding regulation is due to both direct and indirect effects of cytokines. For example, cytokines directly impact appetite and feeding regulation through interactions with orexigenic and anorexigenic neurohormones, neuropeptides, and neurotransmitters (e.g., Amaral et al., 2006; Romanatto et al., 2007; Wang et al., 2006). They also exert their effects on the central nervous system and on neurons in the hypothalamus, the ‘feeding centre’ of the brain (Holden and Pakula, 1996; Plata-Salaman et al., 1996).

Cytokines have also been shown to play a mediatory role in the complex relationship between the immune and neuroendocrine systems. Administration of cytokines has been shown to activate the hypothalamic pituitary adrenal (HPA) axis, stimulating the expression and release of key hormones, including corticotropin-releasing hormone (CRH), adrenocorticotropic hormone (ACTH) and cortisol (Besedovsky and del Rey, 1996, 2007). This is of particular importance in EDs as dysregulation, and more specifically hyperactivation, of the HPA axis has been identified (Lo Sauro et al., 2008; Warren, 2011). Cytokines have also been shown to influence neurotransmitter synthesis, release and reuptake (Capuron and Miller, 2011; Holden and Pakula, 1996). This is of relevance as dysregulation in neurotransmitter systems, such as those related to serotonin and dopamine, have been observed in EDs (Kaye, 2008).

It is also of interest that cytokines have been implicated in the pathophysiology of mental disorders that are highly comorbid with EDs, such as depression (Lichtblau et al., 2013) and anxiety disorders (e.g., generalised anxiety disorders, obsessive compulsive disorder, post-traumatic stress disorder) (Furtado and Katzman, 2015). Therefore, cytokines may be involved in the psychopathology associated with EDs, their development and maintenance (Brown et al., 2008; Corcos et al., 2003; Holden and Pakula, 1996; Marcos, 1997; Nova et al., 2002b; Slotwinska and Slotwinski, 2017).

Cytokine production indicates an inflammatory response in the body and potential causes for cytokine changes in patients with EDs could include infections or a subsequent excessive use of antibiotics, autoimmune or autoinflammatory diseases, and the composition of the gut microbiota, as these factors have been found to be associated with the development of EDs (Breton et al., 2016; Morita et al., 2015; Morris et al., 2016; Raevuori et al., 2016; Zerwas et al., 2017). Recent reviews have focussed on circulating cytokine concentrations in vivo (Dalton et al., 2018; Solmi et al., 2015), therefore, there has been no recent collation of the evidence related to cytokine production in vitro, even though simply measuring plasma or serum levels of cytokines does not reflect the responsiveness of immune cells following an immune challenge and thus does not provide an insight into the dynamics of an individual's immune response. The current study aims to address this by conducting a systematic review of cross-sectional and longitudinal studies assessing in vitro cytokine production across all EDs. We aim to provide an answer to the following two research questions: (i) is in vitro cytokine production altered in people with EDs compared to healthy controls (HCs); and (ii) does in vitro cytokine production change longitudinally in response to treatment interventions?