Intestinal Dysbiosis, Gut Hyperpermeability and Bacterial Translocation: Missing Links Between Depression, Obesity and Type 2 Diabetes?

Abstract and Figures

The comorbid prevalence of major depressive disorder (MDD) with obesity and type II diabetes mellitus reflects the existence of a subset of individuals with a complex common pathophysiology and overlapping risk factors. Such comorbid disease presentations imply a number of difficulties, including: decreased treatment responsivity and adherence; altered glycemic control and increased risk of wider medical complications. A number of factors link MDD to metabolic-associated disorders, including: higher rates of shared risk factors such as poor diet and physical inactivity and biological elements including increased inflammation; insulin resistance; oxidative and nitrosative stress; and mitochondrial dysfunction. All of these biological factors have been extensively investigated in the pathophysiology of obesity and type 2 diabetes mellitus as well as MDD. In this review, we aim to: (1) overview the epidemiological links between MDD, obesity and type 2 diabetes mellitus; (2) discuss the role of synergistic neurotoxic effects in MDD comorbid with obesity, and type 2 diabetes mellitus; (3) review evidence of intestinal dysbiosis, leaky gut and increased bacterial translocation, in the pathophysiology of MDD, obesity and type 2 diabetes mellitus; and (4) propose a model in which the gut-brain axis could play a pivotal role in the comorbidity of these disorders.
Content may be subject to copyright.
Send Orders for Reprints to
Journal Name, Year, Volume 1
Authors copy, Current Pharmaceutical Design (in press)
Intestinal dysbiosis, gut hyperpermeability and bacterial translocation: missing links
between depression, obesity and type 2 diabetes?
Anastasiya Slyepchenkoa,b, Michael Maesc,d,e,f,g, Rodrigo Machado-Vieirah,i, George Andersonj,
Marco Solmik,l,m, Yolanda Sanzn, Michael Berkd,o, Cristiano A. Köhlerp and André F. Carvalhop *
a MiNDS, McMaster Integrative Neuroscience Discovery and Study, McMaster University, Hamilton,
b Women’s Health Concerns Clinic, St Joseph’s Healthcare Hamilton, Hamilton, Canada.
c Department of Psychiatry, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand.
d Deakin University, IMPACT Research Center, Geelong, Australia.
e Department of Psychiatry, Faculty of Medicine, State University of Londrina, Londrina, Brazil.
f Department of Psychiatry, Medical University of Plovdiv, Plovdiv, Bulgaria.
g Revitalis, Waalre, the Netherlands.
h Laboratory of Neuroscience (LIM-27), Department and Institute of Psychiatry, Faculty of Medicine,
University of São Paulo.
i Experimental Therapeutics and Pathophysiology Branch, National Institute of Mental Health, National
Institute of Health, Bethesda, MD, USA.
j CRC Scotland & London, Eccleston Square, London, UK.
k Neuroscience Department, University of Padua, Padua, Italy
l Mental Health Department, Local Health Unit 17, Monselice, Padua, Italy
m Insitute for clinical Research and Education in Medicine, I.R.E.M., Padua, Italy
n Microbial Ecology, Nutrition & Health Research Unit. Institute of Agrochemistry and Food Technology,
National Research Council (IATA-CSIC). Av. Agustin Escardino 7, 46980, Paterna-Valencia, Spain.
o Department of Psychiatry, The Florey Institute of Neuroscience and Mental Health, and Orygen, The
National Centre of Excellence in Youth Mental Health, Centre for Youth Mental Health, University of
Melbourne, Parkville, Vic., Australia.
Send Orders for Reprints to
Journal Name, Year, Volume 2
p Department of Clinical Medicine and Translational Psychiatry Research Group, Faculty of Medicine,
Federal University of Ceará, Fortaleza, CE, Brazil.
*André F. Carvalho, MD, PhD; Department of Clinical Medicine, Faculty of Medicine, Federal
University of Ceará, Rua Prof. Costa Mendes, 1608, 4º andar, 60430-040, Fortaleza, CE, Brazil.
Phone/Fax: +55 85 33668054. E-mail:;
Co-authors e-mails:
Anastasiya Slyepchenko ---
Michael Maes ---
Rodrigo Machado- Vieira ---
George Anderson ---
Marco Solmi ---
Yolanda Sanz ---
Michael Berk ---
Cristiano A. Köhler ---
Send Orders for Reprints to
Journal Name, Year, Volume 3
The comorbid prevalence of major depressive disorder (MDD) with obesity and
type II diabetes mellitus reflects the existence of a subset of individuals with a
complex common pathophysiology and overlapping risk factors. Such comorbid
disease presentations imply a number of difficulties, including: decreased
treatment responsivity and adherence; altered glycemic control and increased
risk of wider medical complications. A number of factors link MDD to
metabolic-associated disorders, including: higher rates of shared risk factors
such as poor diet and physical inactivity and biological elements including
increased inflammation; insulin resistance; oxidative and nitrosative stress; and
mitochondrial dysfunction. All of these biological factors have been extensively
investigated in the pathophysiology of obesity and type 2 diabetes mellitus as
well as MDD. In this review, we aim to: (1) overview the epidemiological links
between MDD, obesity and type 2 diabetes mellitus; (2) discuss the role of
synergistic neurotoxic effects in MDD comorbid with obesity, and type 2
diabetes mellitus; (3) review evidence of intestinal dysbiosis, leaky gut and
increased bacterial translocation, in the pathophysiology of MDD, obesity and
Send Orders for Reprints to
Journal Name, Year, Volume 4
type 2 diabetes mellitus; and (4) propose a model in which the gut-brain axis
could play a pivotal role in the comorbidity of these disorders.
Keywords: microbiome, gut, probiotics, type 2 diabetes mellitus, obesity, metabolic syndrome,
major depressive disorder, psychiatry
The prevalence of metabolic disorders such as obesity and type 2 diabetes mellitus
(T2DM) has grown markedly in recent decades. This has given rise to an emergent population of
individuals presenting with a complex pathology. Apart from numerous metabolic complications,
such as cardiovascular disorders [1] and non-alcoholic fatty liver disease [2], an increased
prevalence of psychiatric comorbidities has also been reported in this population [3-5]. Major
depressive disorder (MDD) is more highly prevalent in populations with both obesity [6] and
type 2 diabetes mellitus (T2DM) [7], particularly in women. Furthermore, population-based
evidence has ascertained that those with a history of mood dysregulation have a higher likelihood
of being obese [8]. Obesity, T2DM and MDD are all linked to the presence of comorbid chronic
health conditions [9-11], poor health outcomes [12, 13], and extensive use of health services [10,
12, 14], indicating high disease burden and disability in these populations [9, 15, 16]. The
condition of the metabolic syndrome (MetS) is defined by a set of conditions that predisposes
Send Orders for Reprints to
Journal Name, Year, Volume 5
individuals to cardiovascular morbidity insulin resistance, low levels of high-density
lipoprotein (HDL) cholesterol, hypertension and increased triglyceride levels [17].
A number of pathophysiological mechanisms link MDD to T2DM and obesity, including:
activated immune-inflammatory pathways, evident in all three conditions [18-20]; mitochondrial
dysfunction, which is linked to problems in adipocyte differentiation and fatty acid oxidation, as
well as reduced clearance of harmful reactive oxygen species (ROS) and increased
atherogenicity and insulin resistance [21, 22]. MDD, T2DM, and obesity have a
pathophysiological overlap in oxidative and nitrosative stress (O&NS), whereby increased
concentrations of ROS and reactive nitrogen species (RNS) cause damage to lipids, proteins
and/or nucleic acids, leading to morbid effects such as vascular damage and the failure of
pancreatic β-cells [22-25]. Insulin resistance, especially centrally, has also been associated with
MDD through mitochondrial dysfunction, including raised levels of O&NS [26], and the
activation of central immune-inflammatory pathways that impair insulin signalling [27, 28].
Finally, it needs to be noted that T2DM and depression share common environmental and
genetic risk factors [29, 30].
Given the high rates of comorbidity and overlapping pathophysiology among T2DM,
obesity and MDD, indicative of commonalities in etiology and course, an integrative treatment
strategy seems to be a realistic target. Indeed, there is a compelling argument more broadly for
an integrated preventive and therapeutic approach to the non-communicable disorders and
common mental disorders [31]. In particular, emerging evidence now links intestinal microbiota
to the etiology of MDD, T2DM and obesity, indicating the microbiome as a potential target for
the management of both of these disorders [32-38].
This article reviews the above areas of research, focussing on: the epidemiological links
among MDD, obesity and T2DM; the synergistic neurotoxic effects in MDD when comorbid
with obesity and/or T2DM; evidence for a role of intestinal microbiota in the pathophysiology of
MDD, obesity and T2DM; finally we provide an integrative model that highlights the potential
role that alterations of intestinal microbiota could play in the overlapping pathophysiology of
MDD, obesity and T2DM.
Send Orders for Reprints to
Journal Name, Year, Volume 6
This narrative review primarily utilized PubMed and Cochrane databases using the
following search terms: depression, major depressive disorder, diabetes, type 2 diabetes mellitus,
obesity, metabolic syndrome, insulin resistance, microbiota, gut microbiome, leaky gut, bacterial
translocation, mucosal immunity, cytokine, immune, inflammation, inflammatory, probiotics,
insuli resistance, leptin resistance and epidemiology. Papers utilized within this review were
considered in terms of the quality of their methodology, and were limited to publications in the
English language, up until January 1st, 2016.
In large-scale epidemiological studies, having a history of mood disorders is associated
with obesity, defined as a body mass index (BMI) >30, especially in females [8]. Lifetime and
12-month prevalence of MDD is increased among obese people in the general population,
although not in those defined as overweight [39]. Past-month prevalence of MDD has been
associated with obesity, primarily accounted for by Class 3 obesity (BMI ≥40) [6]. This
relationship appears bidirectional, with meta-analyses showing that young women with MDD are
at higher risk of becoming obese [40], with obesity increasing MDD risk [41]. Women with
obesity evident during adolescence have an increased risk of developing first anxiety, then MDD
[42]. In diabetic populations, female gender is associated with MDD, as is smoking, older age,
socioeconomic status and perceived health [43]. Prospective meta-analyses have found MDD to
be linked to a 37-60% increase of risk of T2DM, although T2DM only increases MDD risk by
15% [44, 45]. Socioeconomic factors, particularly socioeconomic disadvantage, also influence
the development of these metabolic and mood disturbances [46]. The effects of early-life stress
may be linked to this, with epigenetic mechanisms, such as DNA methylation, histone
modifications, and non-coding RNAs being potentially relevant causal factors [47, 48].
3.1 Phenomenology of Comorbid Major Depressive Disorder in Obesity and Type 2
The presentation of MDD in individuals with obesity is more complex, entailing higher
representation of atypical features such as increased mood reactivity, appetite, and sleep [49, 50].
In recent years, the idea of a metabolic subtype of depression, or metabolic mood syndrome has
been proposed due to the differentiated treatment outcomes, atypical symptomatology and
Send Orders for Reprints to
Journal Name, Year, Volume 7
overlapping synergistic mechanisms involved in MDD comorbid with metabolic disorders [51,
52]. Obesity and diabetes are associated with increased chronicity of MDD, as well as the
severity of its symptomatology [53, 54].
MDD comorbid with T2DM also leads to a more pernicious course of both disorders,
with MDD having an adverse impact on treatment adherence and disability levels, as well as
leading to a 50-75% increase in healthcare costs associated with diabetes [55-58]. The
comorbidity of MDD and T2DM also dramatically increases the odds of functional disability
(OR 7.15) [59], and comorbidities [60], including diabetic retinopathy, neuropathy, sexual
dysfunction and cardiovascular complications [61, 62], as well as being associated with a higher
BMI [63].
Hyperglycemia, a source of complications in T2DM, has been associated with depression
in those with T2DM, including levels of glycosylated haemoglobin HbA1C [63-65]. Depression
in those with T2DM is further linked to work disabilities, such as absenteeism and work
performance difficulties, as well as increased unemployment per se [16]. Moreover, mortality
increases drastically (OR 1.5) for individuals with type 1 and 2 diabetes when comorbid with
MDD [45]. Atypical MDD is also a predictor of longitudinal increases in adiposity [66]. Obesity
is additionally associated with increased levels of suicidal ideation and suicide attempts [67],
while both obesity and MDD correlate with an increased risk of cardiovascular disease and
associated mortality [68, 69].
Population-based evidence shows antidepressant use, tricyclics and selective serotonin
reuptake inhibitors (SSRI), as a risk factor for T2DM [70]. Use of atypical antipsychotics can
lead to rapid weight gain, increasing the risk of developing diabetes, in part via the induction of
insulin resistance and hyperglycemia [71]. Antidepressant use may also interact with glucose
metabolism, thereby further impacting on the management of diabetes [72], whilst T2DM may
diminish the hypothalamic-pituitary-adrenal (HPA) axis response to SSRIs [73]. It is curious that
several antidepressants may have effects on mitochondrial metabolism [74]. Individuals with a
higher BMI present with a decreased treatment responsivity to antidepressants, even though their
medication-induced weight gain is not as significant as in those with a normal BMI [75].
Lifetime MDD additionally affects the outcome of weight reduction programs, in part via
treatment non-adherence [76], with MDD also affecting weight loss retention after surgical and
Send Orders for Reprints to
Journal Name, Year, Volume 8
non-surgical interventions [77, 78]. Combined treatment approaches to obesity and MDD have
been more efficacious in both short and long term outcomes [79, 80].
3.2 Synergistic Mechanisms of Major Depressive Disorder Comorbid with Type II Diabetes
Mellitus and Obesity
A number of synergistic mechanisms have been implicated in the development of MDD
when comorbid with type II diabetes and/or obesity, with consequences as described above
(Table 1). However, no common causal pathway or directionality has been established in the
overlapping course of these disorders, leaving many unanswered questions, including as to the
potential existence of a metabolic subtype of MDD.
<Insert Table 1 here>
MDD and metabolic disorders are both environmentally mediated disorders with a strong
inherited component. Data on the genetic epidemiology of MDD suggesting a 31-42%
heritability of depression [81], while the heritability of T2DM ranges between 31-69% [82, 83],
and that of obesity is approximately 50-90% [84, 85]. As such, genetic factors in overlapping
processes across these conditions may contribute to their overlap, though it is difficult to dissect
genes exclusively from habits transmitted by family environment.
3.2.1 Immune activation, low-grade inflammation and the HPA Axis
Immune deregulation and inflammation have been extensively implicated as mediators of
both metabolic and mood disorder pathophysiology. Dysregulation of the hypothalamic-
pituitary-adrenal (HPA) axis associated with chronic overproduction of glucocorticoids, which
usually constitute the standard acute stress response, has been increasingly reported in
populations with mood and metabolic disorders, providing a common pathophysiological
substrate across these disorders. Poor socioeconomic conditions can also contribute to
hypercortisolemia in response to sustained perceived stress, with poor socioeconomic status also
associated with increased visceral adiposity [86]. Social determinants of health play a role, with
greater exposure to adverse lifestyle risk profiles in those with poor socioeconomic status [87].
For overweight and obese individuals, both lower and higher cortisol reactivity to acute
stress have been reported [88, 89], whereas those with a high waist to hip ratio undergoing
chronic stress show an elevated cortisol response compared to controls [90]. Stress reactivity is
Send Orders for Reprints to
Journal Name, Year, Volume 9
disrupted in those with MDD, including during current episodes as well as in euthymic states,
possibly more so in the melancholic versus atypical subtype of MDD [91].
Patients with MDD coupled to a higher BMI display less improvement on the
dexamethasone-corticotropin releasing hormone (CRH) test, in response to antidepressant
administration, as well as a generally poorer treatment response versus people with MDD who
have a lower BMI [75]. Atypical depressive features have been linked to higher BMI and
abdominal fat mass, as well as to higher levels of plasma adrenocorticotropic hormone (ACTH)
[50], indicating a synergistic worsening of HPA axis hyperactivity in MDD when comorbid with
metabolic disorders.
Continuous low-grade activation of the immune-inflammatory response has been
extensively implicated as a pathophysiological characteristic of MDD, and is also evident in
obesity and T2DM. Table 1 provides an overview of traditional classifications of immune-
inflammatory biomarkers implicated in MDD, obesity and/or T2DM, with alterations in both
anti-inflammatory (e.g. interleukin-10, IL-10) and pro-inflammatory (IL-1, IL-6, tumor necrosis
factor(TNF)-α) cytokines, as well as in cytokines involved in cell-mediated immunity (interferon
(IFN)-γ, IL-2).
While levels of numerous immune-inflammatory biomarkers show alterations in
numerous individual MDD studies, meta-analyses report TNF-α, IL-1β, IL-6, soluble IL-2
Receptor (sIL-2R) and C-reactive protein (CRP) as being consistently elevated in the plasma and
serums of MDD patients [92, 93]. A causal model of MDD has been proposed whereby different
triggers, including auto-immune disorders, inflammatory medical conditions and psycho-social
stress [94] may contribute to the activation of immune-inflammatory and O&NS pathways
leading to a deteriorating and changing pathophysiology, termed neuroprogression, which drives
chronicity and treatment resistance in MDD [95-97], including cognitive dysfunction [98]. This
allows for a bidirectional relationship to exist between inflammation and MDD, whereby each
exacerbates the other [99].
One meta-analytic study shows selective serotonin reuptake inhibitors (SSRIs) to reduce
IL-6 and TNF-α levels, but not IL-1β levels, whilst other studies show antidepressant treatment
reduce the levels of IL-1β and potentially IL-6 [100]. It remains uncertain if this reflects a direct
immune effect of treatment, or mediation via changing state. The soluble IL-2R receptor plays a
major role in triggering T cell response [101]. Raised levels of the receptors of transferrin, an
Send Orders for Reprints to
Journal Name, Year, Volume 10
acute phase protein, have also been noted in the pathophysiology of MDD [101]. T2DM, as with
most non-communicable chronic disorders, also shows evidence of increased inflammation,
including higher levels of fibrinogen [102], plasma IL-6, TNF-α [103] and IL-[104], as well
as upregulated soluble IL-1 receptor antagonist (sIL-1RA) levels [105]. Elevated concentrations
of acute-phase proteins, including haptoglobin, CRP, plasminogen activator inhibitor, sialic acid
and serum amyloid A, have been shown in T2DM patients, providing further evidence of
inflammatory disruption in this disorder [106, 107]. In T2DM patients, only individuals with
depressive symptoms had elevated CRP and urine free cortisol levels [108].
Raised levels of CRP have been proposed as a risk factor for T2DM, although a meta-
analysis showed that the utility of CRP as a predictive factor was confounded by adiponectin
levels, obesity, and markers of liver dysfunction [109]. Higher levels of plasma IL-6, particularly
in conjunction with elevated IL- levels, have also been reported as a risk factor for T2DM
pathogenesis [104]. Other cytokines associated with T2DM include chemokine CC motif ligand
(CCL) 2, CCL5, IL-8, whereas CCL2, CCL7, CCL8, CCL11, CCL13, IL-8 and chemerin are
associated with obesity. Furthermore, chemokine CXCL5 levels are connected to both obesity
and insulin resistance [110].
Obesity is additionally linked to persistent low-grade activation of immune-inflammatory
pathways, especially raised levels of CRP, as reported in a meta-analysis [111]. Levels of plasma
TNF-α correlate with insulin resistance and increase with increasing adiposity [112].
Proportionally, expression of TNF-α mRNA and protein increases in the adipose tissues of obese
individuals [113]. After bariatric surgery, weight loss is accompanied by decreased levels of IL-
6, CRP, insulin and adiponectin, in addition to an improved lipid profile [114].
While pro-inflammatory effects especially IL-6 trans-signaling [115] work to induce
fever, the production of positive acute phase proteins by the liver and activation of cell-mediated
immune processes, also the compensatory anti-inflammatory response system (CIRS) is
activated thereby inactivating an overzealous inflammatory process, with responses involving
increased levels of IL-10, IL-2R and IL-1RA, and HPA-axis hyperactivity [116].
On the other hand levels of anti-inflammatory cytokines, such as IL-10, which appeared
elevated in single studies, do not seem to be elevated in depression, accordingly to a
comprehensive meta-analysis [93]. The primary role of compensatory anti-inflammatory
responses such as the activation of the HPA-axis is to reduce inflammation via production of
Send Orders for Reprints to
Journal Name, Year, Volume 11
glucocorticoids that upon binding on immune cell receptors lead to inhibition of pro-
inflammatory cytokine expression (e.g. TNF-α and IL-1β); however, chronic activation of the
HPA also leads to malfunction of these compensatory mechanisms for example by induction of
glucocorticoid resistance in immunocompetent cells such as macrophages, contributing to
chronic low-grade inflammation [117].
In summary, there are several HPA axis alterations and related cytokine elevations which
are common to the pathophysiology of MDD, T2DM and obesity. Elevations in plasma cortisol
levels, as well as of TNF-α, IL-6 and CRP are evident across all three conditions, indicating the
presence of an immune-inflammatory nexus of these conditions. Further investigations of
immune-inflammatory markers in metabolic disorders comorbid with MDD may contribute to
knowledge as to the potential synergistic effects of these disorders on the immune system.
3.2.2 Cell-Mediated Immunity
Cell mediated immune activation consists of the reciprocal relationship between
macrophages/monocytes and activated T helper (Th) cells, which may further activate other
immune cells such as natural killer cells, macrophages and antigen-specific cytotoxic T-cells, all
contributing to the secretion of various cytokines in response to pathogens [93]. Th cells may be
subdivided into different subsets: Th1 cells, which participate in the extracellular antigen
response, activating cytotoxic T cells, macrophages, and B cells; Th2 cells, which participate in
intracellular antigen response; Th3 or T regulatory (Treg) cells, which participate in the immune
response of mucosal tissues, thereby promoting IgA switching, as well as inhibiting Th1 and Th2
cells [93] and generally dampening pro-inflammatory processes [118]; and Th17 cells, which, by
inducing a more prolonged pro-inflammatory response, are involved in autoimmune responses.
Abnormal cytokine levels may reflect imbalance of the proportion of T cell subsets, such as
Th1/Th2 cell levels or Th17/Treg cells [119]. Indeed, the Th17 system appears to be fundamental
to the development of a state of heightened immune surveillance, as seen in depression [120].
Compelling evidence indicate that depression is associated with aberrations in both innate and
adaptive immunity [121].
In obesity, there is a bias toward Th17 cell proliferation, which is also related to
autoimmune diseases as well as more prolonged and higher levels of inflammation, as evidenced
in high fat diet induced murine obesity [122]. Dendritic cells in adipose tissue from obese mice
Send Orders for Reprints to
Journal Name, Year, Volume 12
are able to induce the proliferation of Th17 cells, which also associate with insulin resistance
[123]. In animal models of obesity and T2DM, B cells have been shown to promote
inflammation through regulation of T-cell function and induction of an inflammatory cytokine
profile associated with reductions in Treg cells [124, 125].
In a comorbid obesity and colitis model, a diet rich in omega-3 polyunsaturated fatty
acids can reduce Th17 and Th1 cell abundance, as well as attenuate the resultant inflammation,
as indicated by elevations in IL-6, IL-17A, IL-17F, IL-21, IL-23, and IFNγ levels [126]. In
T2DM patients, a skewed balance of Treg/Th17 cells, and Treg/Th1 cells was also observed
[126]. B cells induce pro-inflammatory mechanisms, which may lead to increased Treg cell
levels [124]. The Th17 cell associated cytokine IL-17 acts to regulate glucose homeostasis and
adipose tissue, with increased levels of circulating Th-17 cells correlating with increased central
obesity, including in children [127, 128]. In murine models and MDD patients, Th17 cells have
been implicated in the pathogenesis of depression [129], with Th17 cell levels rising in the brains
of mice expressing depression-like behaviour [130], whilst MDD patients show a Th17/Treg
imbalance in association with increased levels of serum IL-17 [131]. A skewing of
inflammatory-associated cells, especially the Treg/Th17 cell ratio, is evident in mood and
metabolic disorders.
Obesity is also characterized by macrophage migration into adipose tissue, which may
represent up to 40% of all cells. The obesity-associated adipose tissue inflammation is
characterized by an increased ratio of “classically activated” macrophages (M1) to “alternative
activated” macrophages (M2). M1 are highly inflammatory macrophages via induction of pro-
inflammatory cytokines and other factors (e.g. primarily TNF-α IL-1β, IL-6 and resistin) and
inducible nitric oxide synthase [iNOS]). By contrast, M2 are the predominant macrophages in
lean adipose tissue and exert anti-inflammatory effects via induction of IL-10 and IL-4 cytokine
production [125, 132]. In addition, M2 may produce catecholamines that sustain adaptive
thermogenesis, increasing energy expenditure in a macrophage-dependent manner [133].
Similarly, a possible role of microglial M1/M2 polarization in relapse and remission of
psychiatric disorders has been suggested [134]. M1 polarized microglia can produce pro-
inflammatory cytokines, ROS, and nitric oxide, suggesting that these molecules contribute to
dysfunction of neural network in the CNS. Alternatively, M2 polarized microglia express
Send Orders for Reprints to
Journal Name, Year, Volume 13
cytokines and receptors that are implicated in inhibiting inflammation and restoring homeostasis
3.2.3 Adipokines
Adipocyte derived factors may have a role in the pathophysiology of both MDD and
metabolic disorders, in part by contributing to inflammatory and immune changes. Such fat-
associated factors, including adipokines such as adiponectin and leptin, may also have a role in
the regulation of mood. However, a recent meta-analysis failed to demonstrate altered peripheral
levels of leptin and adiponectin in individuals with depression compared to healthy controls
[135]. Decreased adiponectin levels leads to a loss of its many anti-inflammatory functions,
including the inhibition of pro-inflammatory cytokine synthesis IL-6 and TNF-α manufacturing
[136], with adiponectin commonly found to be decreased in MDD patients, independent of BMI
and an array of other potential metabolic confounds [137]. Leptin has been classically
conceptualized as a regulator of appetite and wider energy processes, with recent data suggesting
that elevated leptin might predate depression [138] and it could show alterations in MDD,
although with some mixed results. Leptin signalling disturbances, modulated by adiposity, in
particular seem to be longitudinally associated with MDD [139]. Leptin is additionally
associated with immune responses, where it alters the balance of Th cells, skewing the balance
toward Th1 cell response, and otherwise modulating the actions of leukocytes [136].
Additionally, levels of resistin, an adipokine linked to insulin resistance, and leptin have been
correlated to atypical MDD symptomatology [137, 139].
3.2.4 Insulin resistance
In population-based, cross-sectional studies, insulin resistance has been linked to the
presence of MDD, as a possible mechanism linking depression to T2DM and MetS [140].
Further, in individuals with MDD, increased cortisol circulation has been linked to insulin
resistance [141]. Thus, insulin resistance has been hypothesized to result from HPA axis
dysregulation, through overabundant cortisol circulation, its changes to glucose metabolism,
subsequent insulin overabundance, insulin resistance, and finally, T2DM [90]. An alternate
hypothesis states the relationship between depression and insulin resistance to stem from
common risk factors, such as abdominal fat, a lack of physical activity and eating patterns
characteristic of those with obesity, T2DM and MetS [140, 141]. Nevertheless, some studies
Send Orders for Reprints to
Journal Name, Year, Volume 14
reported that MDD is not accompanied by insulin resistance but rather by increased
atherogenicity [25]. Glucose homeostasis is differentially affected by antidepressant use: use of
certain SNRIs such as desipramine leads to hyperglycemia and lowered insulin sensitivity while
certain SSRIs, including fluoxetine exercise normalizing effects on hyperglycemia and insulin
sensitivity. Antidepressants which have dual SNRI and SSRI effects, including duloxetine
seemingly exert no effect on glucose metabolism. Finally, the effects of monoamine oxidase
inhibitors entail low blood glucose levels and increased metabolism thereof [142].
3.2.5 Neural Substrates of Metabolic and Mood Disorders
Neuroprogressive mechanisms, including structural and functional changes within neural
networks, underlie disruptions in processes such as emotional processing, and cognitive function
observed in MDD. A meta-analysis of neuroimaging studies in MDD reports deficient networks
connecting the rostral anterior cingulate and anterior insular cortices to underlie cognitive
impairments in MDD, as well as frontal networks. Meanwhile, emotional dysregulation appears
to rely on network change in frontal networks, the thalamus and striatum [143]. In individuals
with obesity, several network disruptions are present, including the resting state networks,
default network, and fronto-occipital reward processing networks, with magnitude of the
alterations corresponding to BMI [144, 145]. These disruptions include decreased connectivity in
the insula, anterior cingulate and increased connectivity of the precuneus. Further, increases in
orbitofrontal and putamen connectivity are linked to heightened fasting insulin levels and
lowered insulin sensitivity [145]. Further, investigations of the interactions of high BMI and
MDD indicate that while increased BMI is associated with reduced volume of subcortical and
white matter areas, while MDD diagnosis had no such effect. However, reductions in the volume
of the frontal lobe and cerebellum were associated with antidepressant use [146].
A number of neurotransmitter systems are implicated in the circuitry of both MDD and
food consumption, including serotonin, dopamine, and opioids. Opioids acting in conjunction
with cannabinoids underlie the motivational mechanisms of food intake [147], whereas in MDD,
opioids are implicated in emotional dysregulation and stress response alterations characteristic of
the disorder [148]. While monoamines have been long implicated in the pathogenesis of
depression, and continue to be a prime target for its treatment [149], dopaminergic circuits
additionally play an important role in mechanisms key to metabolic disorders, such as the reward
Send Orders for Reprints to
Journal Name, Year, Volume 15
processes of food intake [150]. Neural insulin resistance leads to changes in dopamine
metabolism, as well as changes in mitochondrial function, associated with murine anxiety-like
and depression-like behaviour [26]. In addition, agents that stimulate dopaminergic transmission,
such as amphetamines, may inhibit mitochondrial function [151]. As well, a D2 dopamine
receptor agonist may be used to treat T2DM, which functions through elevating decreased
dopamine in the hypothalamus, and resultant inhibition of hepatic glucose release [152]. Pro-
inflammatory cytokines stimulate the serotonin transporter (SERT) the primary target for
SSRIs -- thereby linking the monoaminergic and inflammatory systems, and inflammatory
conditions can lead to a SERT-mediated reduction in brain-derived neurotrophic factor (BDNF)
gene expression [153].
In individuals without psychiatric disorders, serotonergic activity is linked to insulin
sensitivity [154]. In vitro, SSRIs were linked to pancreatic β cell apoptosis, and led to insulin
resistance, indicating a potential role of SSRIs in contributing to the onset of T2DM [155].
Meanwhile, the presence of T2DM may decrease the effects of SSRIs on ACTH and cortisol
release, which is stimulated in individuals without T2DM or MDD receiving SSRIs [73].
BDNF, a neurotrophin involved in mechanisms such as memory, neuroplasticity and the
survival and differentiation of neurons, has been implicated in the mechanisms involved in both
MDD and obesity. A meta-analysis found serum BDNF levels to be linked to depression score
changes, and for BDNF levels to rise upon treatment with antidepressants [156]. Higher levels of
BDNF are linked to low appetite in older human adults [157], while mice heterozygous for a
BDNF knockout are obese and hyperactive, and display an abnormally increased appetite [158].
Levels of BDNF are additionally inversely linked to BMI, levels of HDL cholesterol, and
triglycerides, suggesting its connection with other MetS components [159]. Recently, a BDNF
SNP rs12291063 minor C allele has been associated with decreased ventromedial hypothalamic
expression of BDNF and obesity, providing a therapeutic target obesity through targeting the
BDNF gene expression [160]. Furthermore, levels of plasma BDNF are lower in individuals with
T2DM [161, 162], while increase in levels of blood glucose lead to the inhibition of neural
BDNF release [161]. Immunoreactive insulin levels have been linked to BDNF levels in women
with T2DM [162].
3.2.6 Oxidative and Nitrosative stress
Send Orders for Reprints to
Journal Name, Year, Volume 16
Inflammation and mitochondrial dysfunction lead to the production of ROS and reactive
nitrogen species (RNS), including nitric oxide (NO), peroxynitrite and peroxides, which may
cause extensive damage of cellular components. This damage to nucleic acids, proteins and
lipids constituent of cellular function is normally in homeostatic, balance with the defensive
mechanisms of antioxidants and antioxidant enzymes. While these reactive species play an
important cell signalling role, as well as provide defensive mechanisms against pathogens, their
excessive presence may result in cell death [163]. Overproduction of ROS and RNS is the
mechanism of O&NS, which occurs when the system is unable to counter the effects of ROS and
RNS, due to the insufficiency of enzymes and other neutralizing pathways present in this system.
In metabolic disorders and MDD, accumulating O&NS results partly from damaged
antioxidant pathways and mitochondrial dysfuction. There are numerous pathways of ROS and
RNS formation. The production of ROS relies on the superoxide anion radical O2*-, which may
act on other molecules to create further ROS. Superoxide is largely produced within the electro
transport chain of the mitochondria, where leakage of electrons during energy transduction in
complexes I and III leads to the formation of the ROS. The hydroxyl radical, on the other hand,
(*OH), is formed when superoxide leads to the release of free iron or other cations, leading to the
genesis of the hydroxyl radical through a number of reactions that produce ROS from iron and
hydrogen peroxide. Peroxyl radicals (ROO*), such as HOO* (the hydroperoxyl radical), are
linked to fatty acid oxidation. Furthermore, peroxisomes, in which cellular oxygen consumption
leads to the synthesis of peroxide (H2O2), which then participates in a different oxidative
processes. In tissue, nitric oxide (NO*) a molecule involved in numerous function including
immune regulation and neurotransmission -- is synthesized by nitric oxide synthases (NOS)
Nucleic acid damage may occur due to the interaction of *OH with nucleic bases or the
deoxyribose backbone [165], leading to permanent oxidative damage to these structures, with
consequences including mutations, ageing and cancer. Fatty acid peroxidation can occur through
ROO*, leading to the formation of malondialdehyde (MDA), a substance resultant from
oxidative damage to lipids. MDA owns to a group of substances often used as markers of
oxidative stress, named thiobarbituric acid reactive substances (TBARS) [166]. Oxidative
damage to protein, occurring through *OH or O2, may be detected through carbonyl group
concentration [167]. Nitrosylation of protein may endanger the functioning of hundreds of
Send Orders for Reprints to
Journal Name, Year, Volume 17
proteins through hypernitrosylation [96]. An IgM-mediated autoimmune response may be
mounted directed against oxidative specific epitopes (neoepitopes such as MDA, oxidized LDL)
nitrosative specific epitopes, e.g. nitros-ryptophan and nitrososo-cysteine [96, 168]. Some of the
formed antibodies may be neurotoxic, e.g. IgM against nitroso-cysteine [96].
The presence of ROS and RNS is countered by a series of mechanisms, including
preventative, repair mechanisms and the presence of antioxidants molecules which protect
against oxidative stress. These include enzymes, such as superoxide dismutase (SOD),
glutathione peroxidase (GPx), catalase, and other molecules such as ascorbic acid and α-
tocopherol (Vitamins C and E, respectively), coenzyme Q10, zinc, carotenoids, flavonoids and
glutathione (GSH) [163].
Zinc acts as an antioxidant and plays a role in hippocampal and glutamatergic affective
regulation and cognitive function circuits [169]. Furthermore, zinc is linked to inflammation
[170], which may cause decline in its levels [171], as well as a role in endocrine functions. Zinc
is additionally linked to the development of T and B cells, and fatty acid metabolism and serum
lipid abnormalities, and its decline is linked to depression [172] and cardiovascular disorders
[169, 173].
The pathophysiology of many non-communicable diseases includes O&NS
cardiovascular disease such as atherosclerosis and cardiomyopathies, various cancers, aging, and
rheumatoid arthritis are associated with the overabundance of ROS and RNS. In the case of
T2DM, high blood sugar is thought to generate ROS from mechanisms such as NADPH oxidase,
NOS, and oxidative phosphorylation. In T2DM, mitochondrial O2* formation is shifted to
complex II, as opposed to complexes I and III [174], while NADPH oxidase additionally
contributes to complications associated with diabetes, such as hypertension [175]. Finally,
glucose auto-oxidation has been proposed as a potential mechanism of ROS genesis, forming
*OH [176]. Finally, hyperglycemia is linked to NOS and peroxynitrite production, and inhibits
NO* synthesis and while reducing its level, linked to endothelial dysfunction in T2DM [177].
Antioxidant depletion is characteristic of diabetes, accounting for lower Vitamin E and C levels
[178] in those with diabetes, though effects on GSH peroxidase activity are not consistent.
Oxidative stress markers of T2DM consist of MDA, increased ratio of GSH to oxidized
glutathione, F2-isoprostanes (results of the oxidation of arachidonic acid), nitrotyrosine,
Send Orders for Reprints to
Journal Name, Year, Volume 18
advanced glycoxidation end products (resultant from glucose reacting with free amine groups on
lipid and protein molecules) and S-glutathionylated proteins [163].
High intake of free fatty acids contributes to the synthesis of H2O2 within the
mitochondria, shifting cells towards an oxidized state. Within murine skeletal muscle, inhibiting
this process has been observed to prevent insulin resistance in high fat diet conditions, indicating
the role of mitochondrial oxidative disruptions in developing insulin resistance [179]. Oxidative
markers found in individuals who are obese include increases in F2-isoprostane, rising levels of
MDA and other TBARS, as well as oxidised low-density lipoprotein levels. Paraoxonase 1 and
total antioxidant capacity in these individuals is decreased, indicating the presence of O&NS in
these individuals, particularly as witnessed by increased levels of lipid peroxidation, in
conjunction with decreased antioxidant defenses against these processes [180]. Furthermore,
increased oxidative stress in adipose tissue can contribute to the development of metabolic
syndrome and insulin resistance through the synthesis of adipokines such as adiponectin, and
cytokines including plasminogen activator inhibitor-6, as well as monocyte chemotactic protein-
1 [181, 182].
A meta-analytic study found levels of 8-hydroxy-2’-deoxyguanosine (8-OHdG), a marker
of oxidative damage to DNA, and F2-isoprostanes, indicative of oxidative damage to lipids were
elevated in MDD [183]. Further meta-analytic findings in MDD include a linear relationship
between MDD symptomatology and elevation of lipid peroxidation, including studies that
measured this through several markers of lipid peroxidation (peripheral MDA, 4-
hydroxynonenal, low-density lipoprotein and F2-isoprostane). In this meta-analysis,
antidepressant treatment was not associated with altered lipid peroxidation marker levels [184].
Serum and plasma zinc is also depleted in MDD [169]. Another meta-analysis found levels of
red blood cell MDA to lower after antidepressant treatment, while serum and red blood cell
MDA as well as F2-isoprostane levels were elevated in individuals with MDD. Levels of
antioxidant substances including serum paraoxonase, uric acid, albumin, HDL cholesterol and
zinc were low in individuals with depression, and uric acid, albumin and vitamin C levels
increased after antidepressant therapy [185]. Further studies found lowered serum levels of
antioxidants such as high-density lipoprotein cholesterol, coenzyme Q10, and melatonin in those
with MDD [186, 187]. The major cortical antioxidants glutathione peroxidase and glutathione
Send Orders for Reprints to
Journal Name, Year, Volume 19
are diminished in the prefrontal cortex as well [188]. In acute episodes of MDD, superoxide
dismutase levels and catalase levels are increased [189].
Overall, this evidence points to impaired mechanisms of protection against ROS and
RNS to be common to MDD and metabolic disorders, potentially acting in a synergistic manner
to produce damaging O&NS effects.
3.2.7 IDO and TRYCATS
As the production of serotonin (5-HT) is contingent on plasma tryptophan levels [190],
excess degradation of tryptophan through the oxidative kynurenine pathway may detract from
levels of cortical 5-HT. This pathway, also known as the tryptophan catabolite (TRYCAT)
pathway includes the metabolism of tryptophan to kynurenine and other TRYCATS by a
pathway including either tryptophan 2,3-dioxygenase (TDO), induced by glucocorticoids, or
indoleamine 2,3-dioxygenase (IDO), induced by pro-inflammatory cytokines and ROS [191-
193]. In particular, IDO, a IFNγ -induced gene product, is widely expressed in a range of human
tissues included immune system, and is stimulated by TNFα, IL-2, and IL-, with LPS and
prostaglandin as well, consequently increasing TRYCAT production and decreasing plasma
tryptophan. The activity of the TRYCAT pathway has been implicated in metabolic disorders
and MDD [194, 195].
In MDD, elevation of glucocorticoids, pro-inflammatory cytokines, and O&NS thus
contributes to diversion of tryptophan to the TRYCAT pathway. Furthermore, levels of
melatonin and N-acetylserotonin, which play antioxidant roles and are synthesized from
tryptophan, are depleted upon TRYCAT pathway activation [196-198].
In the cerebrospinal fluid of individuals with MDD, kynurenine and quinolinic acid, a
metabolite of kynurenine, rise, and are linked to low levels of tryptophan, as well as severity of
affective symptomatology [199]. A murine lipopolysaccharide-induced model of MDD exhibits
activity of the IDO enzyme [200], and is linked to quicker cortical 5-HT turnover and more
prominent depression-like behaviour [201].
Plasma tryptophan levels are reduced in individuals with obesity, an effect which persists
in weight loss [202]. Furthermore, in a murine model with lipopolysaccharide-induced
depression, the expression of hippocampal TNF-α, IFN-γ was exacerbated by obesity, as were
the levels of IDO expression [203]. In MetS, IDO activation by pro-inflammatory cytokines, can
Send Orders for Reprints to
Journal Name, Year, Volume 20
contribute to its pathophysiology through neurotoxic and pro-oxidative effects, including
upregulation of inducible NOS, arachidonic acid, prostaglandin, and phospholipase A2 [194].
Metabolites resultant from the TRYCAT pathway have been implicated in the pathogenesis of
insulin resistance and T2DM [204]. For example, in IFN-γ-induced pancreatic islets, IDO
activity can exert short-term cytotoxic protection, though in the long term, tryptophan
metabolites can induce changes in the β- cells which lead to mitochondria-mediated cytotoxicity
3.3 Toward a metabolic-mood syndrome
Metabolic and mood disorders share many underlying pathological mechanisms. Chronic
stress exposure is associated with symptom severity and persistence in MDD [206], while acute
exposure to stress may be associated with increased food intake and high-energy food intake
[207]. Chronic stress exposure, in turn, can lead to BMI increases, and higher adiposity [208,
209]. Visceral obesity, marked by cardiovascular disease risk, is associated with a cognitive-
affective MDD symptomatology [210]. A “metabolic-mood syndrome” has been proposed to
encompass a distinct subtype of disorder, where chronic stress exposure and mutual
pathophysiological mechanisms will produce a disorder characterized by altered mood and
metabolism. Additionally, in this metabolic-mood syndrome, the neuropsychological deficits
characteristic of both of these subsets of disorders are found to be exacerbated [211].
Furthermore, in this subset of MDD, individuals might also have difficulties in responding to
antidepressant treatment, altered sleep architecture and higher levels of inflammation [212, 213].
The gut-brain axis is conceptualized to maintain intestinal permeability, enteric reflexes
and immune activations, as well as to influence cognitive function and emotion [214]. The gut
microbiota and its genome (microbiome) communicate in a bidirectional manner with the central
nervous system (CNS), potentially through the gut-brain axis, which includes elements of the
autonomic nervous system (ANS) and the HPA axis. An important component of the ANS,
particularly in relation to the gut-brain axis, is the enteric nervous system (ENS), a network of
glia and approximately 500 000 000 neurons in the bowel, surrounding the gastrointestinal tract
[215]. Thus, neural and hormonal communicative pathways connect the enteric system with the
Send Orders for Reprints to
Journal Name, Year, Volume 21
central nervous system, allowing for mutual influence between immune, epithelial, enteric
neuronal cells, and other components of the CNS and the gastrointestinal tract. The gut
microbiota, further, affects intestinal cell physiology, leading to changes in brain-gut
communication [214].
4.1 Appetite control: the gut-brain axis
The physiological mechanisms responsible for food intake and body weight regulation
comprise the gut-brain axis, whereby nutrients and metabolites generated from food activate G
protein receptors on enteroendocrine cells, initiating signal transduction to the CNS. This
nutrient signalling activates gut hormone release, such as ghrelin an orexigenic hormone,
peptide YY3-36 and Glucagon Like Peptide (GLP)-1, which can reduce appetite acting directly on
the vagus nerve, the brainstem, and the hypothalamus. The hypothalamic arcuate nucleus houses
neuropeptide Y/agouti related peptide neurons which induce hunger, while proopiomelanocortin
neurons decrease food intake [216]. Neural afferents and hormonal signals send information
about food intake and adiposity levels to the hypothalamus and brainstem, which contribute to
higher hedonic and reward signals in regulating food intake and energy expenditure [217]. As the
arcuate nucleus is located immediately next to the median eminence, an organ with only a partial
blood-brain barrier, gastrointestinal hormones, insulin and leptin can directly communicate
information about immediate and long-term nutrient availability [218]. In the brainstem, the
nucleus tractus solitarus is a place of vagal afferent input, from which information is further
carried into the hypothalamus. These vagal inputs can also be enhanced by gastrointestinal
hormones, such as ghrelin [219].
4.2 Enteroendocrine cells
The epithelium of the gastrointestinal tract consists of four cellular lineages descended
from pluripotent stem cells at the base of the intestinal crypt: enteroendocrine cells, absorptive
enterocytes, goblet cells and paneth cells. Differentiation of intestinal cells occurs as they
migrate from the intestinal crypts to the surface of the epithelium. Enteroendocrine cells
comprise less than 1% of the epithelial cells, and have most diverse subtype expression in the
small intestine [220]. Numerous subtypes of enteroendocrine cells are responsible for the release
of different products, particularly gastrointestinal hormones that induce varied principal effects
Send Orders for Reprints to
Journal Name, Year, Volume 22
on appetite control, gastric acid secretion, insulin modulation, gastrin release modulation,
intestinal motility and contractions, and appetite regulation [221]. These hormones include A
cell-released ghrelin, D cell-released somatostatin, GLP-1, as released by L cells, and leptin,
released by P cells. Finally, serotonin is released by enterochromaffin cells, as well as certain I,
K and L cells. The release of 5-HT plays a role in intestinal motility, including emesis and
nausea in response to harmful toxins [221]. Enterochromaffin cells are the most profuse subtype
of enteroendocrine cells [221], and may interact with T lymphocytes, which lead to a boost of
serotonin production [222]. Undigested carbohydrates are fermented by anaerobic bacteria,
forming short-chain fatty acids (SCFAs) enhance the release of serotonin by enterochromaffin
cells, thereby augmenting intestinal transit [223].
The enteroendocrine cells also play a role in chemosensory functions, through which the
gut-brain axis is able to regulate metabolism, food digestion and consumption, and pancreatic
secretion. For example, L cells are able to sense nutrients in the luminal contents and release gut
hormones such as PYY and GLP-1 in accordance [216, 224].
4.3 The role of adipokines in gastrointestinal functions
Gastrointestinal hormones influence several subsets of tissues, including exocrine glands,
smooth muscle, and the peripheral nervous system [217]. Hormones such as peptide YY, GLP-1
and oxyntomodulin are released in response to stomach distention and intestinal nutrient intake.
These hormones lead to the sensations of hunger or satiety, correspondingly increasing or
decreasing orexigenic signalling, while correspondingly inhibiting or enhancing anorexigenic
signalling within the hypothalamus [220]. Gut hormones also moderate intestinal transit
mechanisms, leading to sustained distention of the gut and encourage the persistence of satiety
Weight loss is linked to the lack of plasma leptin, which is an adipokine that modulates
the sensation of hunger [225]. This hormone is produced from the ob gene, and is elevated in
those with obesity, as hyper-leptinemia in obesity occurs through leptin resistance lack of
response to leptin. The functions of leptin include adapting energy metabolism, regulating
adipose tissue metabolism, particularly by inhibiting fatty acid synthesis and stimulating
Send Orders for Reprints to
Journal Name, Year, Volume 23
triglyceride breakdown [226]. Another function of leptin is the improvement of pancreatic β-cell
function, inducing insulin secretion [227].
Adiponectin, is another adipokine, associated with increased fatty acid oxidation and
insulin sensitivity [228]. In those with obesity and T2DM, plasma adiponectin is decreased. Low
levels of adiponectin are associated with insulin resistance [229], while adiponectin itself has
been linked to pro-inflammatory effects [230].
Levels of ghrelin increase prior to, and decrease upon the act of food consumption,
indicating its role in meal initiation [231]. Ghrelin additionally suppresses insulin secretion,
increases serum cortisol concentration [232], decreases fat oxidation, and increases murine
adipose tissue [233]. In individuals with obesity, plasma ghrelin is decreased [234], while weight
loss is associated with elevated ghrelin levels [235].
Glucagon-like peptide 1 induces insulin release incrementally in response to glucose
release by affecting β cells and insulin transcription [236], acting through the arcuate nucleus of
the hypothalamus [237]. Individuals with morbid obesity and type 2 diabetes mellitus have
reduced GLP-1 secretion in response to meal consumption [238, 239].
Roux-en-Y gastric bypass surgery works to decrease leptin and acylation-stimulating
protein, while improving plasma lipids and insulin resistance. Furthermore, adiponectin is lower
in those with obesity, and is increased after Roux-en-Y gastric bypass. However, levels of
ghrelin are elevated upon gastric bypass surgery, despite the weight loss experienced by these
patients [226].
4.4 The microbiome-gut-brain axis
The gut microbiota comprises approximately 1014 organisms, which in human adults are
mainly represented by members of the phyla Bacteroidetes and Firmicutes, though the exact
microbiome composition varies greatly between individuals [240]. Though infants are born with
an almost sterile gut, its colonization begins from maternal vaginal and fecal microbiomes [241,
242], and further develops through milk diet, and gradually, solid food, finally stabilizing into an
adult profile upon weaning [243]. Due to the diversity of the microbiome and its increasing
relevance as a therapeutic and pathogenic target, there are many large, ongoing microbiome
projects in North America and around the world, targeting the human microbiome, such as the
National Institutes of Health Human Microbiome Project, the Canadian Microbiome Initiative,
Send Orders for Reprints to
Journal Name, Year, Volume 24
the French Metagenopolis programme, the Irish Alimentary Pharmacobiotic Centre Microbiome
Institute and a number of EU collaborative projects (e.g. MyNewGut, Metacardis, Florinash).
Metagenomic initiatives such as the Human Microbiome Project aimed to characterize the
bacterial genome common to all individuals, finding at least 160 shared bacterial species [240].
Further initiatives are now progressing towards a better understanding of the functional role of
the microbiota and its implication in the progression from health to disease in longitudinal
studies and translational research approaches [244]. This extensive symbiotic addition to the
human genome and its diversity has been increasingly implicated in not only immune and
metabolic processes, but in the pathogenesis of many other disorders. The gut microbiota
constitute a large portion of the response to epithelial cell damage [245], digestion of food, drug
metabolism and toxicity, and is involved in the development and maintenance of intestinal
immune functions [245-247].
The intestinal microbiome breaks down dietary complex polysaccharides that utilizes as
main energy source and, in turn, provides to the host secondary potentially beneficial metabolites
(e.g. butyrate), thereby exemplifying a mutualistic relationship within the gut. Moreover, it is
involved in amino acid metabolism from protein fractions that escape digestion in the small
intestine and is partly responsible for metabolic effects of dietary fats via indirect mechanisms
[248]. Metabolic diversity thus arises in humans from the variation within the microbiome [249].
Furthermore, microbiota-derived metabolites generated from dietary fiber (short-chain fatty
acids), amino acids (Tyr, Trp) or other food constituents may also signal via the gut-brain axis,
influencing mental health. For example, γ aminobutyric acid can be produced by Lactobacillus
spp. and Bifidobacterium spp. from monosodium glutamate, constitutes a putative link between
the microbiome and depression [250]. Furthermore, endogenous bacteria can lead to
upregulation of serotonin synthesis and secretion to both the lumen and the bloodstream through
the microbiota metabolites’ interactions with ECs [251].
An important mechanism of interaction between the microbiome and the gut-brain axis is
the production of short-chain fatty acids (SCFAs), which are fermented by subtypes of bacteria
from starch and fibre. The production of SCFAs, including acetate, propionate and butyrate, aids
sodium and water absorption by the colon, as well as mucus release, cellular differentiation, and
intestinal cell proliferation. These are easily absorbed from the lumen, and perform numerous
functions: for example, acetate increases ileal motility and blood flow to the colon [252]. Finally,
Send Orders for Reprints to
Journal Name, Year, Volume 25
butyrate plays signalling and glucose-metabolism related roles [253]. One of the functions of
propionate and butyrate is their joint effect of activation of intestinal gluconeogenesis, indicating
the necessity of SCFAs for weight loss and glucose control activation of intestinal
gluconeogenesis [254]. It has also been suggested in humanized animal models that via
production of SCFA the microbiota increases expression of tryptophan hydroxylase 1 mRNA in
colon, leading to mucosal serotonin synthesis, and increased serotonin concentration compared
to germ-free models [255]. Microbiota additionally generate hydrogen sulphide, which together
with ROS, NO, SCFAs modulate host mitochondrial activity, depending on bacterial strain
quality and diversity [256].
Intestinal dysbiosis has been implicated in a number of disorders, including in children
with autism, where symptom severity corresponds to microbiome composition [257, 258]. The
gut brain axis and microbiome in irritable bowel syndrome is also disrupted, as intestinal motility
and secretion is altered, as is the composition of the microbiome in terms of stability and
diversity, and is accompanied by immune changes, including inflammation [259, 260]. Intestinal
dysbiosis in IBS subjects has also been related to post-infectious IBS and psychological distress
[261]. The gut microbiome has been linked to the development of autoimmune disorders,
allergies and asthma [262, 263].
Due to the vast and complex nature of the gut microbiota it is unclear yet whether it could
be established what constitutes a “healthy” microbiota and microbiome. However, emerging
research suggests that specific bacterial species are more closely associated with a healthy
phenotype. This may occur via direct signalling through the host-microbe cross-talk or through
indirect signalling of products derived from the micro-microbe cross-talk. Changes in gut
microbiota composition and function can be attributed to environmental insults (e.g. exposure to
toxic environmental factors, diet, pathogens and antibiotics) as well as to host conditions. Germ
free mice, for example, have increased susceptibility to pathogenic infection [264, 265]. Immune
responses within the gastrointestinal tract can additionally modulate the composition of gut
microbiota, indicating a complex bidirectional relationship between the host and the microbiota.
Although the microbiota tends to go back to “normal” after an insult exposure a drastic
Send Orders for Reprints to
Journal Name, Year, Volume 26
disruption of its delicate balanced could turn our symbiotic microbiota into a part of the problem,
becoming one of the pathogenic actors adversely affecting the host health [266].
An important immune role of the microbiota is to maintain the development of lymphoid
tissue in the gut, including crypt patches, and isolated lymphoid follicles [267]. As well, the
proliferation of Th17 cells, which accumulate within the intestine, can be promoted by
segmented filamentous bacteria. The importance of the microbiota to the differentiation of this
subset of cells can also be observed in germ-free mice, where Th17 cells diminish in number
[268]. The presence of peripheral Treg cells also depends on the presence of the microbiome.
Within the colon, IL-10 producing Treg cells accumulate with colonization of Clostridium
strains, altered Schaedler flora, or Bacteroides fragilis in germ-free mice [269-271].
The gut microbiome shows a degree of plasticity and adaptation in response to the
environment of its host. For example, prolonged cold exposure leads to maximization of caloric
uptake, as maintained by increased gut epithelial surface area. In the short term, this cold
exposure derived microbiota increases insulin sensitivity, and promotes the browning of white
adipose tissue, leading to fat loss and efficient energy use. These effects are contingent on the
microbiome and are transferable with transplants of microbiota [272].
An important implication of the interaction of the gut dysbiosis and the host consists of
behavioural modulation by intestinal microbiota. For instance, stress reactivity of the HPA axis
is modulated by the microbiota, as is expression of behaviours such as anxiety-like symptoms
and behavioural despair [273]. (See Figure 1 for mechanisms of interaction of gut microbiota
with symptoms of depression and metabolic disorders).
The defense of the mucosal barrier is a key protective measure against pathogens, as
many pathogens infiltrate the gut epithelium as a primary infectious mechanism. The mucus
layer, as produced by the epithelium, defends against both pathogenic and commensal bacteria
[274]. Impairment in the mucosal barrier, the epithelium and its tight junctions has been linked to
bacterial translocation, autoimmune diseases, such as celiac disease, type 1 diabetes, multiple
sclerosis, and inflammatory bowel diseases [275, 276].
The gut microbiota interacts with mechanisms involved in depression and metabolic
disorders at several common substrates: gastrointestinal function, central and autonomic nervous
Send Orders for Reprints to
Journal Name, Year, Volume 27
system interaction, and immune interactions. Fat storage and energy balance are regulated by the
gut microbiota, further linking adiposity to functions of the microbiota. Intestinal barrier function
can be disrupted as a result of factors such as dysbiosis, immune activation and inflammatory
response, leading to increased permeability and thus “leaking” of toxic compounds and bacteria
into the bloodstream [32]. Most important is increased bacterial translocation of Gram-negative
bacteria from the gut to the mucosal lymph nodes and the blood stream, including Hafnia alvei,
Pseudomonas aeruginosa, Morganella morganii, Pseudomonas putida, Citrobacter koseri, and
Klebsiella pneumoniae. Thus, the IgA or IgM responses directed against LPS of these bacteria is
increased in patients with depression and these findings together with increased gut-
inflammation in depression indicate leaky gut in a considerable number of depressed patients
[32]. Increased translocation of Gram-negative bacteria is frequently associated with gut-derived
inflammation, that is immune activation and low-grade inflammation as a consequence of
increased LPS levels activating the Toll-Like Receptor 4 complex [277]. This cascade is MyD88
independent, and via activation of transcription factor IRF-3 leads to IFNβ production, which in
turn via Stat 1 induces several IFN-related genes [278]. Moreover, bacterial translocation is
known to cause autoimmune responses via the generation of oxidative and nitrosative stress
modified epitopes (neoepitopes) or mimicry processes [279]. For example, in depression,
increased bacterial translocation of Gram negative commensal bacteria is strongly related to
inflammatory, oxidative and nitrosative and autoimmune processes [279]. Also, in chronic
fatigue syndrome (CFS), an illness that is strongly comorbid with depression similar disorders in
bacterial translocation are found, while there are strong relations between increased LPS load in
serum or mucosal lymph nodes and increased levels of M1 cytokines and autoimmune responses
directed against serotonin which in CFS are associated with mood disorders [280].
Intestinal barrier dysfunction has been noted in T2DM, where this may lead to immune
damage to Β cells and insulin resistance, through the effects of inflammation in response to
bacterial translocation [281]. A high fat diet in mice elevates concentration of LPS, constituting
metabolic endotoxemia, while LPS alone is able to induce obesity and insulin resistance [282].
Furthermore, antibiotic treatment can lower inflammatory markers, decrease glucose intolerance
and caecal LPS levels, alter oxidative stress and reduce adiposity in high-fat diet and ob/ob mice
Send Orders for Reprints to
Journal Name, Year, Volume 28
7. The effects of compositional changes in the microbiome on depressive behaviour, obesity
and diabetes
7.1. Findings from comparative metagenomic analyses
Several landmark studies established a link between obesity and the microbiome.
Initially, changes in microbiome composition and function were proposed to lead to a
characteristic microbiota that was capable of extracting energy from the diet more efficiently
[246], although more complex interactions currently appear to affect the obesity-associated low-
grade inflammation and the neuroendocrine system as well [284] . A large number of human
studies indicate that reductions in the proportion of Bacteroidetes or Bacteroides spp. parallel, in
some cases, to increases in Firmicutes or members of this phylum (e.g. Clostridium subgroups)
are associated with obesity while those proportions tend to normalize on a low-calorie diet and in
parallel to weight loss [38, 285]. Lean phenotypes have been linked to higher functional diversity
within the phylum Bacteroidetes [125, 286]. A possible mechanism is the increased LPS
production with disrupted Bacteroidetes/Firmicutes ratio, with decreased GLP-2, an intestinal
peptide that tightens cell junctions and prevents LPS permeability. If LPS enters plasma, the
consequence is the activation of inflammatory pathway, a core feature in obesity.
Nevertheless, consistency across studies regarding the bacterial taxa associated with
obesity is continuously being strived for. In fact, a large human study reported that low bacterial
gene count in total fecal DNA, interpreted as low bacterial richness, was the main feature that
characterized the microbiota of subjects more prone to adiposity, insulin resistance,
dyslipidaemia and to have an inflammatory phenotype (e.g. having higher levels of highly
sensitive C-reactive protein and an elevated lymphocyte count) when compared with those
individuals with higher bacterial gene count, interpreted as high bacterial richness. Those who
were obese and had a less diverse microbiome were more prone to further weight gain. The
bacterial taxa that discriminate between these two groups of individuals were also to some extent
opposite to previous studies, since Bacteroidetes/Bacteroides and Proteobacteria were more
abundant in subjects with low bacterial gene count, while butyrate producing bacteria like
Faecalibacterium, mucus-degrading bacteria like Akkermansia and bacterial genera classically
used as probiotics (Bifidobacterium, Lactobacillus) were more abundant in subjects with high
bacterial gene count. Individuals were found to be distinguished to be lean or obese by the
relative abundance of 18 species, while high and low diversity of the microbiome can be
Send Orders for Reprints to
Journal Name, Year, Volume 29
distinguished by 4-19 species [287]. However, not all subjects in the former group have
metabolic alterations and vice versa indicating that conclusions could be biased due to the lack of
control of confounder factors, such as diet. Another observational study has suggested that
differences in microbiota composition are more clearly associated with T2DM that with obesity.
Particularly, plasma glucose concentrations and diabetes correlated with increased Β-
proteobacteria, and increased Bacteroidetes:Firmicutes ratio [288]. The microbiota has been
suggested to mediate a signal that promotes inflammation resultant from consuming a high fat
diet to develop obesity, insulin resistance and T2DM [284]. Nevertheless, lack of adequate
control of confounding factors constitutes an important weakness of this study.
Another study compared the intestinal microbiota of subjects with normal glucose
tolerance, prediabetes and newly diagnosed T2DM to understand its possible involvement in
disease development. Individuals with T2DM present intestinal dysbiosis, and differences in
alpha diversity. Further, metabolic factors such as insulin tolerance, fasting plasma glucose and
CRP levels were linked to gut microbiome diversity. Butyrate-producing bacteria such as
Akkermansia muciniphila and Faecalibacterium prausnitzii were more abundant in subjects with
normal glucose tolerance that in pre-diabetics. Also, 28 operational tax units (OTUs) were linked
to T2DM including: Clostridiales and Subdoligranullum, Lachnospiraceae and Ruminococcus, as
well as Eubacterium, with Clostridia and Verrucomicrobiae were increased in T2DM patients,
whereas the abundance of Bacteroides decreased. The microbiota of individuals with T2DM was
also differentiated from individuals with normal glucose tolerance by increased abundance of
Prevotella and Collinsella. Furthermore, Streptococcus abundance lowered with increased
glucose tolerance. Additionally, Prevotella and Megamonas levels in the pre-diabetes group
were higher than in the normal glucose tolerance group [289].
A recent small fecal metagenomic study of monozygotic Korean twins at two time points
separated by approximately 2 years enabled the study of microbial associations of the
development of sub-clinical obesity and T2DM. For instance, BMI, fasting blood sugar and
insulin levels were negatively correlated with the abundance of Akkermansia muciniphila and
positively correlated with riboflavin and NAD biosynthesis. These associations occurred over
both prior to and after the onset of T2D clinical markers, suggesting that the microbiota may
contribute to or react to changes in the host environment (such as oxidative stress) prior to the
onset of disease. In contrast to other studies, a positive correlation between Bacteroides and BMI
Send Orders for Reprints to
Journal Name, Year, Volume 30
was found, which can be attributed to the effects of confounding factors such as demographics,
diet, sample preparation, and analysis [290].
A metagenomic analysis of 345 individuals with T2DM found 60,000 markers to
characterize T2DM. Those with T2DM had moderate dysbiosis, including high levels of
opportunistic pathogens, as well as a lower number of butyrate-producing bacteria, additionally
expressing sulphate reduction and oxidative stress resistance [291]. Further, researchers have
been able to identify the presence of T2DM in women through their metagenomic profiles,
differentiating these women from those with impaired glucose tolerance, regardless of treatment.
Such metagenomic profiles were different between European and Chinese cohorts [292],
revealing that our current understanding is insufficient to enable prediction of an individual’s
disease risk based solely on the gut microbiome [36].
Findings from observational studies human studies described above point to a role of gut
microbiome composition and function in obesity and metabolic disease T2DM; yet, lack of
control of confounder factors (diet, medication, etc.) constitutes a major weakness that partly
explains the conflicting results and precludes their use in practice [36, 38]. In fact, stratifying
previous quantitative gut metagenomic studies of T2DM patients [291, 292] for treatment, and
analysing the effects of the most widely used antidiabetic drug metformin, has led to the
identification of a unified signature of gut microbiome shifts in T2DM patients. This signature is
characterized by depletion of butyrate-producing taxa that also causes functional microbiome
shifts, which can be partially ameliorated by metformin treatment that leads to short-chain fatty
acid production [293].
Furthermore, prospective epidemiological studies that follow the participants’
progression from health to disease, combining multiple-functional -omics technologies are
necessary to provide more sound evidence of the microbiome-related features that constitute risk
factors for these disorders. Big data and machine learning techniques are likely to be needed for
such analyses.
Stress is a well-known risk factor for developing anxiety and, in turn, depression. In
addition to the first-established associations between stress and intestinal dysbiosis in different
rodent models [273, 294], a recent human study has also reported that maternal stress is linked to
alterations in vaginal microbiota, as well as alterations in infant gut microbiota and related
Send Orders for Reprints to
Journal Name, Year, Volume 31
metabolites, which presumably could affect metabolic programming and neurodevelopmental
disorders [295].
A study analyzing the fecal sequences of patients with and without MDD, found
correlations between depression and fecal microbiota, including a higher proportion of the order
Bacteroidales and a lower proportion of the family Lachonspiraceae. Finally, a clade
Oscillobacter, which produces a homolog of γ aminobutyric acid, and one clade in Alistipes
associated with stress, was associated with depression [296]. Further investigation confirmed a
change in diversity in the microbiome in those with current mood episodes, but not in euthymic
individuals. Increased levels of Bacteroidetes, Proteobacteria, and Actinobacteria were observed,
while Firmicutes are reduced in those with MDD, regardless of current mood state. This study
confirmed an increase of Alistipes, as well as Enterobacteriaceae, while Faecalibacterium levels
were low, and correlated with improved depressive symptom severity. Serum BDNF level was
correlated with one Clostridium genus [297].
7.2 The effects of manipulations of the gut microbiome
Germ free mice are often used to study the microbiota-gut-brain axis: these mice are
raised in a sterile environment, with sterile gastrointestinal tracts. This allows for the study of
how a lack of microbiome can affect behaviour. As well, intentional colonization of germ-free
animals with a particular microorganism or human microbiota allows to progress from
associations established in humans between bacterial taxa or a specific microbiota structure and
diverse conditions to cause effects relationships. This strategy also help to understand the
underlying mechanisms of action of the microbiota, for example via signalling through the gut-
brain axis.
Colonizing germ-free mice with a microbiota from obese human volunteers leads to
increased abdominal adiposity, compared to lean volunteers; a similar effect can be observed in
ob/ob mouse microbiota transplant, confirming the direct role of the microbiota associated with
a higher relative abundance of Firmicutes and reduced abundance of Bacteroidetes [246]. A
more recent study also showed that when microbiota from twin subjects discordant for obesity
(uncultured fecal communities and the corresponding fecal bacterial cultures) was transferred to
germ-free mice, these mice develop the corresponding phenotype [298]. Furthermore, cohousing
mice harboring the obese human microbiota with mice containing the lean human microbiota
Send Orders for Reprints to
Journal Name, Year, Volume 32
prevented the development of an obesity-associated metabolic phenotype due to the invasion of
the obese mouse intestine with specific members of the phylum Bacteroidetes (e.g. Bacteroides
uniformis, B. caccae, B. cellulosilyticus, etc.) from the lean microbiota, in which mice were also
fed a chow diet representing lowsaturated fat, high fruit and vegetable [298]. These findings
were also supported by an independent study where the administration of B. uniformis CECT
7771 isolated from a breast-fed infant to a mouse model of diet-induced obesity ameliorated
metabolic and immune dysfunction [299].
Roux-en-Y gastric bypass leads to changes in the gut microbiota, potentially exerting
some of the beneficial effects of this surgery through changed SCFA production. Furthermore,
gut microbiota transplants from mice that had undergone gastric bypass surgery to germ-free
mice led to changes in SCFA production, and decreased adiposity in the mice. Increased
Γproteobacteria abundance is characteristic of Roux-en-Y gastric bypass surgery [300].
Mice given microbiota from donors on a high fat diet showed a number of behavioural
changes, with impairments in exploratory and cognitive behaviour. These changes were induced
by distinct microbiota groups for the high-fat diet-fed mice, with decreased diversity, and
changes to metabolically active taxa, correlating with disruptions in the epithelial barrier, higher
levels of NOS in the intestines [301].
While anxiety-like behaviour has not clearly shown an association with absence of
microbiota, as shown by contrasting findings in behavioural tests such as the elevated plus maze,
the light-dark box preference tests [302-304], versus performance on the open field test [302,
305], reconstitution of germ-free mice with physiologic microbiota colonization normalizes
anxious behaviour in those who had anxious behavioral symptoms, and partially normalizes
HPA response, tryptophan metabolism, and hippocampal BDNF levels [302, 303, 306, 307] in
animal models. Additionally, memory is impaired for germ-free mice, in terms of T-maze and
novel object test performance [305]. HPA axis hyperactivation has been observed in these
animals through elevated peripheral corticosterone [303-305, 307] as well as plasma
adrenocorticotropic hormone levels [307]. Levels of monamine turnover are higher in germ free
mice, where tryptophan is metabolized quicker, and levels of serotonin are higher [302, 303,
308]. Receptor expression for the 5HT1A receptor decreases in the dentate gyrus of the
hippocampus, while D1 dopamine receptor expression is elevated in the hippocampi, and
decreased in the striatum and nuclei accumbens [302, 308].
Send Orders for Reprints to
Journal Name, Year, Volume 33
In germ free mice, BDNF levels are decreased in male rat cortices, including
hippocampal, amygdala, and cingulate cortex expression [302, 307]. Levels of N-methyl-D-
aspartate (NMDA) receptor NR2B mRNA expression in the amygdala of female mice are
elevated, and NR2A receptor expression is lower in the hippocampus while NR1 and NR2A
receptor expression is increased in the cortex [304, 307].
Using germ-free and specific pathogen-free mice, it has also been recently demonstrated
that early-life stress induced by maternal separation alters the hypothalamic-pituitary-adrenal
axis and colonic cholinergic neural regulation in a microbiota-independent fashion and that
microbiota are required for the induction of anxiety-like behaviour and behavioural despair
8.1 Use of antioxidants to target the leaky gut: implications for depression
In patients with Chronic Fatigue Syndrome (CFS), administration of high doses of
glutamine, zinc and N-acetyl-L-cysteine has been shown to reduce bacterial translocation as
measured by means of IgA responses to the LPS of Gram negative bacteria [309]. This
improvement in bacterial translocation is accompanied by a clinical improvement in CFS and
accompanying symptoms, including mood disorders. There is preclinical and clinical evidence
that this combination of improves signs of leaky gut and may improve tight junction functions in
different paradigms [309]. We have also evidence that the same treatment may improve leaky
gut in depression (Maes et al., personal data).
8.2 Probiotics: linking depression, obesity and T2DM treatment
Several recent streams of evidence have pointed towards the efficacy of using probiotics
as a treatment for depression. Administration of a strain of Lactobacillus helveticus for example,
has been reported to improve memory deficits and anxiety in mice fed a high-fat diet [310].
Further, anxiety-like behaviours have been alleviated in healthy mice, though such evidence for
depression-like behavior has not been consistent [311-313]. Treatment with some bacterial
strains has also been linked to cognitive improvement [314]. HPA axis activity is attenuated by
probiotics such as Bifidobacteria infantis 35624, and commercially available L. rhamnosus
R0011 and L. helveticus R0052 treatment, as evidenced by changes in corticosterone levels [305,
Send Orders for Reprints to
Journal Name, Year, Volume 34
311]. The attenuation of anxiety-like behavior in mice by Lactobacillus rhamnosus JB-1 is
vagus-nerve dependent [315]. Finally, L. rhamnosus R0011 and L. helveticus R0052 or L.
rhamnosus JB-1 treatment reduces levels of IL-6 and TNF-α, relevant to both MDD and
metabolic disorders [305, 313]. The role of bifidobacteria in the regulation of the HPA axis and
anxiety has been recently reviewed [316]. The exaggerated response of germ-free mice to mild
restraint stress, reflected in an increased release of corticosterone and ACTH, has been reversed
by mono-colonization with a strain of B. infantis. In a maternal separation model of
psychological stress, a strain of B. infantis also reduced corticosterone levels. Potential
mechanisms by which some bifidobacteria may modulate the HPA axis and anxiety, were
illustrated by a study where B. infantis 35624 was shown to elevate plasma tryptophan levels, a
precursor of serotonin, by modulating the expression of enzymes (e.g., indoleamine-2,3-
dioxygenase) involved in the tryptophan degradation pathway in Sprague-Dawley rats. A strain
of B. longum reversed infection-induced behavioural changes associated with decreased
hippocampal BDNF mRNA expression, without affecting cytokine or tryptophan metabolism
and independently of vagus nerve activation [316]. A blend of L. helveticus R0052 and B.longum
R0175 has also been reported to reduce anxiety in both rats and human subjects [317].
Some of the first randomized control trials of probiotics targeting depressive symptoms
have been conducted. An 8-week administration of strains of L. acidophilus, L. casei and B.
bifidum reduced depression scores in MDD patients, lowered insulin levels and insulin
resistance, as well as serum hs-CRP levels. As well, plasma glutathione levels rose, though
antioxidant capacity levels and lipid profiles did not change [318]. Chronic fatigue syndrome
patients exhibited improved anxiety scores after 8-week L. casei Shirota treatment [319], while
2-week administration of a strain of Clostridium butyricum to pre-laryngeal cancer surgery
patients relieved anxiety scores [320].
Several studies have looked at probiotic treatments of individuals without mood, anxiety
or gastrointestinal disorders. A 4-week probiotic treatment consisting of strains of B. bifidum
W23, B. lactis W52, L. acidophilus W37, L. brevis W63, L. casei W56, L. salivarius W24, and
Lactococcus lactis W19 and W58 was administered to healthy participants. Here, treatment was
found to have lower reactivity to sad mood, characterized by a diminishment of aggression and
rumination [321]. Other studies of healthy volunteers have found probiotic strains such as L.
helveticus R0052, B. longum R0175; B. animalis lactis I-2494, Streptococcus thermophilus I-
Send Orders for Reprints to
Journal Name, Year, Volume 35
1630, L. bulgaricus I-1632 and I-1519, Lactococcus lactis lactis I-1631; to exert positive benefits
on stress, relieve anxiety and depression scores, and changes in task-related activity in affective,
interoceptive, and somatosensory regions [317, 322-324].
A few human intervention trials with different potential probiotic strains have been
designed to evaluate effect on obesity or related biochemical and inflammatory markers, but
evidence of efficacy is still limited [34, 38]. One such study aims to determine the effects of
Lactobacillus gasseri SBT2055 on obesity and reported decreased body fat mass and BMI in the
probiotic group compared to placebo with the control group at 12 weeks of intervention [325].
An 8-week probiotic supplement, containing strains of Lactobacillus acidophilus, L. casei, L.
rhamnosus, L. bulgaricus, Streptococcus thermopilus, Bifidobacterium breve, and B. longum
combined with prebiotic fructo-olisaccharides administered to T2DM patients led to decreased
fasting plasma glucose, and lower insulin resistance compared to the placebo group. Meanwhile,
decreased hs-CRP and increased plasma GSH levels were also observed, indicating antioxidant
and anti-inflammatory effects [326]. Intake of prebiotics, i.e. oligofructose that aids the
proliferation of Bifidobacterium spp., linked to better insulin sensitivity, glucose tolerance,
reduction of adiposity and normalization of IL-1α and IL-6 levels, which had been increased in
high fat diet fed mice [282]. In type 2 diabetes, a six week intervention with L. acidophilus La5
and B. lactis Bb12 led to improved lipid profile in terms of cholesterol and LDL-C levels [327].
In rats, treatment with L. acidophilus W70, L.casei W56, L. salivarius W24, Lactococcus lactis
W58, B. bifidum W23 and B. lactis W52 can also reduce lipid peroxidation, as marked by
malondialdehyde, and increased glutathione synthesis [328]. Finally, in human participants, an
intestinal infusion of lean donor microbiota to those with MetS led to improvements in insulin
sensitivity, as well as elevated levels of butyrate-producing bacteria. Post-treatment, Escherichia
coli levels and NO-producing Alcaligenes facealis differed between treatment groups, though no
difference in post-treatment bacterial diversity was found [253].
Further well-designed randomized placebo-controlled human trials of sufficient duration
are still necessary to firmly prove that intervention in the gut ecosystem could be part of the
solution for anxiety or mood disorders [34]. Future longitudinal studies intend to disentangle the
role of the gut microbiota and its genome in the progression from health to disease should also
take into consideration the possible commonalities between metabolic and mood disorders [329].
This would also help to identify common solutions, whose effectiveness will also have to be
Send Orders for Reprints to
Journal Name, Year, Volume 36
verified by appropriate intervention trials in subjects with or at risk of developing comorbid
mood and metabolic disorders. Further microbiome-based studies could also contribute to
progressing into personalized medicine approaches that address both metabolic and mood
disorders As well, the interactions of current pharmaceutical treatment with the host microbiome
should be considered when evaluating the response to current antidepressant or diabetic
treatments, as has been done with metformin use.
Metabolic disorders and depression share many immune-inflammatory and O&NS and
neuroendocrine pathways. Intestinal dysbiosis, leaky gut and increased bacterial translocation are
common pathophysiological mechanism underlying the activated immune-inflammatory and
O&NS pathways associated with these disorders thereby possibly explaining their high
comorbidity. Further evidence confirming this new pathway could set the basis of future gut-
related intervention strategies to better prevent and treat comorbid mood and metabolic
YS contribution is supported by the European Union’s Seventh Framework Program under the
grant agreement no 613979 (MyNewGut) and the Spanish Ministry of Economy and
Competitiveness (MINECO, Spain) grant AGL2014-52101-P. MB is supported by a NHMRC
Senior Principal Research Fellowship 1059660. AFC is supported by a research fellowship
award from Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq; Brazil).
Send Orders for Reprints to
Journal Name, Year, Volume 37
AS, MM, RMV, GA, AS, YS, MS, CAK and AFC have no conflicts of interest to declare. MB
has received a grant/research support from Bristol Myers Squibb, Eli Lilly, Glaxo SmithKline,
Meat and Livestock Board, Organon, Novartis, Mayne Pharma, Servier and Woolworths, and has
been a speaker for Astra Zeneca, Bristol Myers Squibb, Eli Lilly, Glaxo SmithKline, Janssen
Cilag, Lundbeck, Merck, Pfizer, Sanofi Synthelabo, Servier, Solvay, and Wyeth as well as
serving as a consultant to Allergan, Astra Zeneca, Bioadvantex, Eli Lilly, Glaxo SmithKline,
Janssen Cilag, Lundbeck Merck, Pfizer and Servier.
Send Orders for Reprints to
Journal Name, Year, Volume 38
Table 1: Classifications of Inflammatory Biomarkers Implicated in MDD, Obesity and/or T2DM
Leptin, adiponectin, ghrelin
Hormones derived from fatty
tissues; linked to modulation
of inflammatory effects,
energy use.
Cell-mediated immune
IFN-γ, IL-2, sIL-2R, sTNFR
Th-1 like
IFN-γ, IL-2
Participate in activating
extracellular antigen response
Th-2 like
IL-4, IL-10
Participate in activating
intracellular antigen response
Th-3 like
Participate in the immune
response of mucosal tissues
Th-17 like
IL-17A, IL-17F, IL-21, IL-22
Act in conjunction with
regulatory T (Treg) cells, and
appear to be involved in
autoimmune disease
T-reg like
IL-10, TGF-β
Are the major regulatory cells
involved in immune response.
Anti-inflammatory Cytokines
IL-4, IL-6, IL-10
Part of the anti-inflammatory
response, which aims to
compensate for and reverse
Pro-inflammatory Cytokines
IFN-γ, TNF-α, IL-1Β, IL-2,
IL-6, IL-8
Part of the inflammatory
response, which aims to
activate body response against
various antigens
Acute phase proteins
CRP, haptoglobin, transferrin
von Willebrand factor,
Produced by liver in response
to cytokine levels in
Send Orders for Reprints to
Journal Name, Year, Volume 39
fibrinogen, serum amyloid A,
mannan-binding lectin,
plasminogen activator
inhibitor PAI-1
CRP- C-reactive protein, IFN interferon, ILInterleukin, MDD Major Depressive disorder,
sIL-2R soluble Interleukin 2 Receptor, sTNFR soluble tumour necrosis factor receptor,
T2DM- type 2 Diabetes Mellitus, TGF transforming growth factor, Th T helper, TNF
tumor necrosis factor,
Send Orders for Reprints to
Journal Name, Year, Volume 40
Send Orders for Reprints to
Journal Name, Year, Volume 41
Figure 1: A number of comorbid mechanisms are at play in the mutual comorbidity of major
depressive disorder, obesity and type 2 diabetes mellitus, including low-grade inflammation,
cell-mediated immunity, endocrine disturbances, oxidative and nitrosative stress, and changes in
activity of the hypothalamic-pituitary-adrenal axis. Many of these pathways interact
bidirectionally with the microbiome, making the microbiome an important, novel target for
these comorbid disorders. CCL CC motif ligand; CRP C-reactive protein; HPA
Hypothalamic Pituitary Adrenal; IDO indoleamine 2,3 dioxygenase; IL interleukin; ROS
Reactive Oxygen Species; RNS Reactive Nitrogen Species; sIL-2R soluble interleukin 2
receptor; Th T helper; TNF - Tumor necrosis factor; TRYCAT tryptophan catabolite
Send Orders for Reprints to
Journal Name, Year, Volume 42
Abbreviation List
Adrenocorticotrophic Hormone
Brain-derived Neurotrophic Factor
C-reactive Protein
Hypothalamic-Pituitary-Adrenal Axis
Indoleamine 2,3-dioxygenase
Major Depressive Disorder
Metabolic Syndrome
Oxidative and Nitrosative Stress
Short-chain Fatty Acids
Soluble Interleukin-2 Receptor
Selective Serotonin Reuptake Inhibitor
Tryptophan 2,3-dioxygenase
Type II Diabetes Mellitus
T helper cell
Regulatory T cell
Tumor Necrosis Factor
Tryptophan Catabolite
Send Orders for Reprints to
Journal Name, Year, Volume 43
[1] Wilson PW, D'Agostino RB, Sullivan L, Parise H, Kannel WB. Overweight and obesity
as determinants of cardiovascular risk: the Framingham experience. Archives of
internal medicine 2002; 162: 1867-1872.
[2] Sheth SG, Gordon FD, Chopra S. Nonalcoholic steatohepatitis. Ann Intern Med 1997;
126: 137-45.
[3] Simon GE, Von Korff M, Saunders K, Miglioretti DL, Crane PK, van Belle G, Kessler RC.
Association between obesity and psychiatric disorders in the US adult population.
Archives of general psychiatry 2006; 63: 824-830.
[4] Anderson RJ, Freedland KE, Clouse RE, Lustman PJ. The prevalence of comorbid
depression in adults with diabetes a meta-analysis. Diabetes care 2001; 24: 1069-
[5] McEvoy JP, Meyer JM, Goff DC, Nasrallah HA, Davis SM, Sullivan L, Meltzer HY, Hsiao
J, Stroup TS, Lieberman JA. Prevalence of the metabolic syndrome in patients with
schizophrenia: baseline results from the Clinical Antipsychotic Trials of Intervention
Effectiveness (CATIE) schizophrenia trial and comparison with national estimates
from NHANES III. Schizophrenia research 2005; 80: 19-32.
[6] Onyike CU, Crum RM, Lee HB, Lyketsos CG, Eaton WW. Is obesity associated with
major depression? Results from the Third National Health and Nutrition
Examination Survey. American journal of epidemiology 2003; 158: 1139-1147.
[7] Ali S, Stone M, Peters J, Davies M, Khunti K. The prevalence of comorbid
depression in adults with Type 2 diabetes: a systematic review and metaanalysis.
Diabetic Medicine 2006; 23: 1165-1173.
[8] Dennis C-L, Ross LE, McIntyre RS, Konarski JZ, Wilkins K, Soczynska JK, Kennedy SH,
Fleury M-J, Grenier G, Lesage A. Obesity in bipolar disorder and major depressive
disorder: results from a national community health survey on mental health and
well-being. Canadian Journal of Psychiatry 2006; 51: 274-280.
[9] Moussavi S, Chatterji S, Verdes E, Tandon A, Patel V, Ustun B. Depression, chronic
diseases, and decrements in health: results from the World Health Surveys. The
Lancet 2007; 370: 851-858.
[10] Sturm R. The effects of obesity, smoking, and drinking on medical problems and
costs. Health Affairs 2002; 21: 245-253.
[11] Görlitz N, Keller M, Ziegler A. [Prevalence of complications and comorbidities in type
2 diabetes: a cross-sectional study of disease management program participants in
Bavaria]. Deutsche medizinische Wochenschrift (1946) 2008; 133: 1667-1672.
[12] Katon W, Ciechanowski P. Impact of major depression on chronic medical illness.
Journal of psychosomatic research 2002; 53: 859-863.
[13] Dixon JB. The effect of obesity on health outcomes. Molecular and cellular
endocrinology 2010; 316: 104-108.
[14] Association AD. Economic costs of diabetes in the US in 2007. Diabetes care 2008;
31: 596-615.
[15] Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The disease burden
associated with overweight and obesity. Jama 1999; 282: 1523-1529.
Send Orders for Reprints to
Journal Name, Year, Volume 44
[16] Von Korff M, Katon W, Lin EH, Simon G, Ciechanowski P, Ludman E, Oliver M, Rutter
C, Young B. Work disability among individuals with diabetes. Diabetes care 2005;
28: 1326-1332.
[17] Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US
adults: findings from the third National Health and Nutrition Examination Survey.
Jama 2002; 287: 356-359.
[18] Dandona P, Aljada A, Bandyopadhyay A. Inflammation: the link between insulin
resistance, obesity and diabetes. Trends in immunology 2004; 25: 4-7.
[19] Miller AH, Maletic V, Raison CL. Inflammation and its discontents: the role of
cytokines in the pathophysiology of major depression. Biological psychiatry 2009;
65: 732-741.
[20] Pickup JC. Inflammation and activated innate immunity in the pathogenesis of type 2
diabetes. Diabetes care 2004; 27: 813-823.
[21] Lowell BB, Shulman GI. Mitochondrial dysfunction and type 2 diabetes. Science
2005; 307: 384-387.
[22] Bournat JC, Brown CW. Mitochondrial dysfunction in obesity. Current Opinion in
Endocrinology, Diabetes and Obesity 2010; 17: 446-452.
[23] Maes M, Ruckoanich P, Chang YS, Mahanonda N, Berk M. Multiple aberrations in
shared inflammatory and oxidative & nitrosative stress (IO&NS) pathways explain
the co-association of depression and cardiovascular disorder (CVD), and the
increased risk for CVD and due mortality in depressed patients. Progress in Neuro-
Psychopharmacology and Biological Psychiatry 2011; 35: 769-783.
[24] Vargas HO, Nunes SOV, Barbosa DS, Vargas MM, Cestari A, Dodd S, Venugopal K,
Maes M, Berk M. Castelli risk indexes 1 and 2 are higher in major depression but
other characteristics of the metabolic syndrome are not specific to mood disorders.
Life sciences 2014; 102: 65-71.
[25] Bortolasci CC, Vargas HO, Nunes SOV, de Melo LGP, de Castro MRP, Moreira EG,
Dodd S, Barbosa DS, Berk M, Maes M. Factors influencing insulin resistance in
relation to atherogenicity in mood disorders, the metabolic syndrome and tobacco
use disorder. Journal of affective disorders 2015; 179: 148-155.
[26] Kleinridders A, Cai W, Cappellucci L, Ghazarian A, Collins WR, Vienberg SG, Pothos
EN, Kahn CR. Insulin resistance in brain alters dopamine turnover and causes
behavioral disorders. Proceedings of the National Academy of Sciences 2015; 112:
[27] Black PH. The inflammatory response is an integral part of the stress response:
Implications for atherosclerosis, insulin resistance, type II diabetes and metabolic
syndrome X. Brain, behavior, and immunity 2003; 17: 350-364.
[28] Kan C, Silva N, Golden SH, Rajala U, Timonen M, Stahl D, Ismail K. A systematic
review and meta-analysis of the association between depression and insulin
resistance. Diabetes care 2013; 36: 480-489.
[29] Holt RI, de Groot M, Golden SH. Diabetes and depression. Curr Diab Rep 2014; 14:
[30] Kan C, Pedersen NL, Christensen K, Bornstein SR, Licinio J, MacCabe JH, Ismail K,
Rijsdijk F. Genetic overlap between type 2 diabetes and depression in Swedish and
Danish twin registries. Mol Psychiatry 2016.
Send Orders for Reprints to
Journal Name, Year, Volume 45
[31] O'Neil A, Jacka FN, Quirk SE, Cocker F, Taylor CB, Oldenburg B, Berk M. A shared
framework for the common mental disorders and Non-Communicable Disease: key
considerations for disease prevention and control. BMC Psychiatry 2015; 15: 15.
[32] Maes M, Kubera M, Leunis J-C. The gut-brain barrier in major depression: intestinal
mucosal dysfunction with an increased translocation of LPS from gram negative
enterobacteria (leaky gut) plays a role in the inflammatory pathophysiology of
depression. Neuroendocrinology Letters 2008; 29: 117-124.
[33] Anderson G, Maes M. The gutbrain axis: The role of melatonin in linking
psychiatric, inflammatory and neurodegenerative conditions. Advances in
Integrative Medicine 2015; 2: 31-37.
[34] Slyepchenko A, F Carvalho A, S Cha D, Kasper S, S McIntyre R. Gut emotions-
mechanisms of action of probiotics as novel therapeutic targets for depression and
anxiety disorders. CNS & Neurological Disorders-Drug Targets (Formerly Current
Drug Targets-CNS & Neurological Disorders) 2014; 13: 1770-1786.
[35] Tilg H, Kaser A. Gut microbiome, obesity, and metabolic dysfunction. The Journal of
clinical investigation 2011; 121: 2126.
[36] Sanz Y, Olivares M, Moya-Pérez Á, Agostoni C. Understanding the role of gut
microbiome in metabolic disease risk. Pediatric research 2014; 77: 236-244.
[37] Sanz Y, Moya-Pérez A. Microbiota, inflammation and obesity. In: ed.^eds., Microbial
Endocrinology: The Microbiota-Gut-Brain Axis in Health and Disease. Springer,
2014; pp. 291-317.
[38] Sanz Y, Rastmanesh R, Agostonic C. Understanding the role of gut microbes and
probiotics in obesity: how far are we? Pharmacological Research 2013; 69: 144-155.
[39] Scott KM, McGee MA, Wells JE, Browne MAO. Obesity and mental disorders in the
adult general population. Journal of psychosomatic research 2008; 64: 97-105.
[40] Blaine B. Does depression cause obesity? A meta-analysis of longitudinal studies of
depression and weight control. Journal of health psychology 2008; 13: 1190-1197.
[41] Luppino FS, de Wit LM, Bouvy PF, Stijnen T, Cuijpers P, Penninx BW, Zitman FG.
Overweight, obesity, and depression: a systematic review and meta-analysis of
longitudinal studies. Archives of general psychiatry 2010; 67: 220-229.
[42] Anderson SE, Cohen P, Naumova EN, Jacques PF, Must A. Adolescent obesity and risk
for subsequent major depressive disorder and anxiety disorder: prospective
evidence. Psychosomatic Medicine 2007; 69: 740-747.
[43] Egede LE, Zheng D. Independent factors associated with major depressive disorder
in a national sample of individuals with diabetes. Diabetes care 2003; 26: 104-111.
[44] Mezuk B, Eaton WW, Albrecht S, Golden SH. Depression and type 2 diabetes over the
lifespan a meta-analysis. Diabetes care 2008; 31: 2383-2390.
[45] Knol M, Twisk J, Beekman A, Heine R, Snoek F, Pouwer F. Depression as a risk factor
for the onset of type 2 diabetes mellitus. A meta-analysis. Diabetologia 2006; 49:
[46] Everson SA, Maty SC, Lynch JW, Kaplan GA. Epidemiologic evidence for the relation
between socioeconomic status and depression, obesity, and diabetes. Journal of
psychosomatic research 2002; 53: 891-895.
[47] Vaiserman A. Epigenetic programming by earlylife stress: Evidence from human
populations. Developmental Dynamics 2015; 244: 254-265.
Send Orders for Reprints to
Journal Name, Year, Volume 46
[48] Bohacek J, Mansuy IM. Epigenetic inheritance of disease and disease risk.
Neuropsychopharmacology 2013; 38: 220-236.
[49] Lamers F, de Jonge P, Nolen WA, Smit JH, Zitman FG, Beekman A, Penninx B.
Identifying depressive subtypes in a large cohort study: results from the
Netherlands Study of Depression and Anxiety (NESDA). The Journal of clinical
psychiatry 2010; 71: 1582-1589.
[50] Cizza G, Ronsaville DS, Kleitz H, Eskandari F, Mistry S, Torvik S, Sonbolian N,
Reynolds JC, Blackman MR, Gold PW. Clinical subtypes of depression are associated
with specific metabolic parameters and circadian endocrine profiles in women: the
power study. PloS one 2012; 7: e28912.
[51] Vogelzangs N, Beekman A, Boelhouwer IG, Bandinelli S, Milaneschi Y, Ferrucci L,
Penninx B. Metabolic depression: a chronic depressive subtype? Findings from the
InCHIANTI study of older persons. The Journal of clinical psychiatry 2011; 72: 598-
[52] Mansur RB, Brietzke E, McIntyre RS. Is there a “metabolic-mood syndrome”? A
review of the relationship between obesity and mood disorders. Neuroscience &
Biobehavioral Reviews 2015; 52: 89-104.
[53] Byers AL, Vittinghoff E, Lui L-Y, Hoang T, Blazer DG, Covinsky KE, Ensrud KE, Cauley
JA, Hillier TA, Fredman L. Twenty-year depressive trajectories among older women.
Archives of general psychiatry 2012; 69: 1073-1079.
[54] Vogelzangs N, Kritchevsky SB, Beekman AT, Brenes GA, Newman AB, Satterfield S,
Yaffe K, Harris TB, Penninx BW. Obesity and onset of significant depressive
symptoms: Results from a community-based cohort of older men and women. The
Journal of clinical psychiatry 2010; 71: 391.
[55] Ciechanowski PS, Katon WJ, Russo JE. Depression and diabetes: impact of depressive
symptoms on adherence, function, and costs. Archives of internal medicine 2000;
160: 3278-3285.
[56] Simon GE, Katon WJ, Lin EH, Ludman E, VonKorff M, Ciechanowski P, Young BA.
Diabetes complications and depression as predictors of health service costs. General
hospital psychiatry 2005; 27: 344-351.
[57] Egede LE, Zheng D, Simpson K. Comorbid depression is associated with increased
health care use and expenditures in individuals with diabetes. Diabetes care 2002;
25: 464-470.
[58] Gonzalez JS, Peyrot M, McCarl LA, Collins EM, Serpa L, Mimiaga MJ, Safren SA.
Depression and diabetes treatment nonadherence: a meta-analysis. Diabetes care
2008; 31: 2398-2403.
[59] Egede LE. Diabetes, major depression, and functional disability among US adults.
Diabetes care 2004; 27: 421-428.
[60] Vamos EP, Mucsi I, Keszei A, Kopp MS, Novak M. Comorbid depression is associated
with increased healthcare utilization and lost productivity in persons with diabetes:
a large nationally representative Hungarian population survey. Psychosomatic
Medicine 2009; 71: 501-507.
[61] De Groot M, Anderson R, Freedland KE, Clouse RE, Lustman PJ. Association of
depression and diabetes complications: a meta-analysis. Psychosomatic medicine
2001; 63: 619-630.
Send Orders for Reprints to
Journal Name, Year, Volume 47
[62] Labad J, Price J, Strachan M, Fowkes F, Ding J, Deary I, Lee A, Frier B, Seckl J, Walker
B. Symptoms of depression but not anxiety are associated with central obesity and
cardiovascular disease in people with type 2 diabetes: the Edinburgh Type 2
Diabetes Study. Diabetologia 2010; 53: 467-471.
[63] Kimbro LB, Steers WN, Mangione CM, Duru OK, Ettner SL. The association of
depression and the cardiovascular risk factors of blood pressure, HbA1c, and body
mass index among patients with diabetes: results from the translating research into
action for diabetes study. International journal of endocrinology 2012; 2012.
[64] De la Roca-Chiapas JM, Hernández-González M, Candelario M, de la Luz Villafaña M,
Hernández E, Solorio S, Rivera AE, Sosa M, Jasso JA. Association between depression
and higher glucose levels in middle-aged Mexican patients with diabetes. Rev Invest
Clin 2013; 65: 209-213.
[65] Lustman PJ, Anderson RJ, Freedland KE, De Groot M, Carney RM, Clouse RE.
Depression and poor glycemic control: a meta-analytic review of the literature.
Diabetes care 2000; 23: 934-942.
[66] Lasserre AM, Glaus J, Vandeleur CL, Marques-Vidal P, Vaucher J, Bastardot F, Waeber
G, Vollenweider P, Preisig M. Depression with atypical features and increase in
obesity, body mass index, waist circumference, and fat mass: a prospective,
population-based study. JAMA psychiatry 2014; 71: 880-888.
[67] Carpenter KM, Hasin DS, Allison DB, Faith MS. Relationships between obesity and
DSM-IV major depressive disorder, suicide ideation, and suicide attempts: results
from a general population study. American journal of public health 2000; 90: 251.
[68] Penninx BW, Beekman AT, Honig A, Deeg DJ, Schoevers RA, van Eijk JT, van Tilburg
W. Depression and cardiac mortality: results from a community-based longitudinal
study. Archives of General Psychiatry 2001; 58: 221-227.
[69] Jiang J, Ahn J, Huang W-Y, Hayes RB. Association of obesity with cardiovascular
disease mortality in the PLCO trial. Preventive medicine 2013; 57: 60-64.
[70] Kivimäki M, Hamer M, Batty GD, Geddes JR, Tabak AG, Pentti J, Virtanen M, Vahtera J.
Antidepressant Medication Use, Weight Gain, and Risk of Type 2 Diabetes A
population-based study. Diabetes care 2010; 33: 2611-2616.
[71] Clark NG. Consensus development conference on antipsychotic drugs and obesity
and diabetes. Diabetes care 2004; 27: 596.
[72] Goodnick PJ, Henry JH, Buki V. Treatment of depression in patients with diabetes
mellitus. Journal of Clinical Psychiatry 1995.
[73] Trento M, Kucich C, Tibaldi P, Gennari S, Tedesco S, Balbo M, Arvat E, Cavallo F,
Ghigo E, Porta M. A study of central serotoninergic activity in healthy subjects and
patients with Type 2 diabetes treated by traditional one-to-one care or Group Care.
Journal of endocrinological investigation 2010; 33: 624-628.
[74] Callaly E, Walder K, Morris G, Maes M, Debnath M, Berk M. Mitochondrial
dysfunction in the pathophysiology of bipolar disorder: effects of pharmacotherapy.
Mini Rev Med Chem 2015; 15: 355-65.
[75] Kloiber S, Ising M, Reppermund S, Horstmann S, Dose T, Majer M, Zihl J, Pfister H,
Unschuld PG, Holsboer F. Overweight and obesity affect treatment response in
major depression. Biological psychiatry 2007; 62: 321-326.
Send Orders for Reprints to
Journal Name, Year, Volume 48
[76] Marcus MD, Wing RR, Guare J, Blair EH, Jawad A. Lifetime prevalence of major
depression and its effect on treatment outcome in obese type II diabetic patients.
Diabetes care 1992; 15: 253-255.
[77] Legenbauer T, Petrak F, de Zwaan M, Herpertz S. Influence of depressive and eating
disorders on short-and long-term course of weight after surgical and nonsurgical
weight loss treatment. Comprehensive psychiatry 2011; 52: 301-311.
[78] Ohsiek S, Williams M. Psychological factors influencing weight loss maintenance: an
integrative literature review. Journal of the American Academy of Nurse
Practitioners 2011; 23: 592-601.
[79] Simon GE, Rohde P, Ludman EJ, Jeffery RW, Linde JA, Operskalski BH, Arterburn D.
Association between change in depression and change in weight among women
enrolled in weight loss treatment. General hospital psychiatry 2010; 32: 583-589.
[80] Busch AM, Whited MC, Appelhans BM, Schneider KL, Waring ME, DeBiasse MA,
Oleski JL, Crawford SL, Pagoto SL. Reliable change in depression during behavioral
weight loss treatment among women with major depression. Obesity 2013; 21:
[81] Sullivan PF, Neale MC, Kendler KS. Genetic epidemiology of major depression:
review and meta-analysis. American Journal of Psychiatry 2014.
[82] Almgren P, Lehtovirta M, Isomaa B, Sarelin L, Taskinen M, Lyssenko V, Tuomi T,
Groop L, Group BS. Heritability and familiality of type 2 diabetes and related
quantitative traits in the Botnia Study. Diabetologia 2011; 54: 2811-2819.
[83] Poulsen P, Kyvik KO, Vaag A, Beck-Nielsen H. Heritability of type II (non-insulin-
dependent) diabetes mellitus and abnormal glucose tolerancea population-based
twin study. Diabetologia 1999; 42: 139-145.
[84] Bell CG, Walley AJ, Froguel P. The genetics of human obesity. Nature Reviews
Genetics 2005; 6: 221-234.
[85] Hernandez LM, Blazer DG, Faith MS, Kral TV. Social Environmental and Genetic
Influences on Obesity and Obesity-Promoting Behaviors: Fostering Research
Integration. 2006.
[86] Rosmond R, Björntorp P. Occupational Status, Cortisol Secretory Pattern, and
Visceral Obesity in Middleaged Men. Obesity Research 2000; 8: 445-450.
[87] Williams LJ, Brennan SL, Henry MJ, Berk M, Jacka FN, Nicholson GC, Kotowicz MA,
Pasco JA. Area-based socioeconomic status and mood disorders: cross-sectional
evidence from a cohort of randomly selected adult women. Maturitas 2011; 69: 173-
[88] Mujica-Parodi L, Renelique R, Taylor M. Higher body fat percentage is associated
with increased cortisol reactivity and impaired cognitive resilience in response to
acute emotional stress. International Journal of Obesity 2009; 33: 157-165.
[89] Jones A, McMillan MR, Jones RW, Kowalik GT, Steeden JA, Deanfield JE, Pruessner JC,
Taylor AM, Muthurangu V. Adiposity is associated with blunted cardiovascular,
neuroendocrine and cognitive responses to acute mental stress. PloS one 2012; 7:
[90] Brown ES, Varghese FP, McEwen BS. Association of depression with medical illness:
does cortisol play a role? Biological psychiatry 2004; 55: 1-9.