Immune system and glucose metabolism interaction in schizophrenia: A chicken–egg dilemma

Progress in Neuro-Psychopharmacology & Biological Psychiatry xxx (2012) xxx–xxx

PNP-08259; No of Pages 8

Contents lists available at SciVerse ScienceDirect

Progress in Neuro-Psychopharmacology & BiologicalPsychiatry

j ourna l homepage: www.e lsev ie r .com/ locate /pnp

Immune system and glucose metabolism interaction in schizophrenia:A chicken–egg dilemma

Johann Steiner a,b,⁎, Hans-Gert Bernstein a, Kolja Schiltz a,b, Ulf J. Müller a, Sabine Westphal c,Hemmo A. Drexhage d, Bernhard Bogerts a,b

a Department of Psychiatry, University of Magdeburg, Magdeburg, Germanyb Center for Behavioral Brain Sciences, Magdeburg, Germanyc Institute of Clinical Chemistry/Lipid Outpatient Clinic, University of Magdeburg, Magdeburg, Germanyd Department of Immunology, Erasmus Medical Center, Rotterdam, The Netherlands

Abbreviations: 5HT2C receptor, a subtype of 5-hydroxthe neurotransmitter serotonin; AgRP, agouti-related pepdisease; Akt1, a serine–threonine protein kinase; ARC,BMI, Body-Mass-Index; CAPON, C-terminal PDZ domainsynthase; CART, cocaine- and amphetamine-regulated trCD, cluster of differentiation; CRH, corticotropin-releasingtein; CSF, cerebrospinal fluid; DCs, dendritic cells; FDG-PEemission tomography; fMRI, functional magnetic resonalike peptide-1; GLUT4, insulin-sensitive glucose transportIGF-1, insulin-like growth factor-1; IL, interleukin; LMNMCH, melanin-concentrating hormone; MHC, Major Hismononuclear phagocyte system; NPY, neuropeptide Y;PBMCs, peripheral blood mononuclear cells; PolyI:C, poacid; POMC, pro-opiomelanocortin; PYY, peptide YY; RGSing 4; S100B, a calcium-binding member of a family of pammonium sulfate at neutral pH; sIL-2R, soluble IL-2 recfor advanced glycation products; TGF-β, transforming gronecrosis factor-alpha; TRH, thyrotropin releasing hormon⁎ Corresponding author at: Department of Psychiatry, Un

Str. 44, D-39120 Magdeburg, Germany. Tel.: +49 391 67 1E-mail address: [email protected] (J. Stei

0278-5846/$ – see front matter © 2012 Elsevier Inc. Allhttp://dx.doi.org/10.1016/j.pnpbp.2012.09.016

Please cite this article as: Steiner J, et al, ImmNeuro-Psychopharmacol Biol Psychiatry (2

a b s t r a c t
a r t i c l e i n f o
Article history:Received 28 July 2012Received in revised form 15 September 2012Accepted 22 September 2012Available online xxxx

Keywords:Glucose metabolismImmune systemLymphocytesMicrogliaMononuclear phagocyte systemSchizophrenia

Impaired glucose metabolism and the development of metabolic syndrome contribute to a reduction in theaverage life expectancy of individuals with schizophrenia. It is unclear whether this association simply reflectsan unhealthy lifestyle or whether weight gain and impaired glucose tolerance in patients with schizophreniaare directly attributable to the side effects of atypical antipsychotic medications or disease-inherent derange-ments. In addition, numerous previous studies have highlighted alterations in the immune system of patientswith schizophrenia. Increased concentrations of interleukin (IL)-1, IL-6, and transforming growth factor-beta(TGF-β) appear to be state markers, whereas IL-12, interferon-gamma (IFN-γ), tumor necrosis factor-alpha(TNF-α), and soluble IL-2 receptor (sIL-2R) appear to be trait markers of schizophrenia. Moreover, the mononu-clear phagocyte system (MPS) and microglial activation are involved in the early course of the disease. Thisreview illustrates a “chicken–egg dilemma”, as it is currently unclear whether impaired cerebral glucose utiliza-tion leads to secondary disturbances in peripheral glucose metabolism, an increased risk of cardiovascularcomplications, and accompanying pro-inflammatory changes in patients with schizophrenia or whetherimmune mechanisms may be involved in the initial pathogenesis of schizophrenia, which leads to disturbancesin glucosemetabolism such asmetabolic syndrome. Alternatively, shared underlying factors may be responsiblefor the co-occurrence of immune system and glucose metabolism disturbances in schizophrenia.

© 2012 Elsevier Inc. All rights reserved.

ytryptamine receptor that bindstide; AITD, autoimmune thyroidhypothalamic arcuate nucleus;ligand of neuronal nitric oxideanscript; CCK, cholecystokinin;hormone; CRP, C-reactive pro-T, fluorodeoxyglucose positronnce imaging; GLP-1, glucagon-er 4; IFN-γ, interferon-gamma;A, gene encoding lamin A/C;

tocompatibility Complex; MPS,NTS, nucleus tractus solitarii;lyriboinosinic–polyribocytidilic4, regulator of G-protein signal-roteins that are 100% soluble ineptor; sRAGE, soluble receptorwth factor-beta; TNF-α, tumore.iversity ofMagdeburg, Leipziger5019; fax: +49 391 67 15223.ner).

rights reserved.

une system and glucose me012), http://dx.doi.org/10.10

1. Introduction

1.1. The influence of cardiovascular risk factors, diabetes and metabolicsyndrome on the life expectancy of schizophrenic patients

The average life expectancy of individuals with schizophrenia is 12to 15 years lower than that of the general population (van Os andKapur, 2009). In addition to the higher suicide rate (approximately5%), increased cardiovascular risk factors, such as types 1/2 diabetesand metabolic syndrome (defined by the American Heart Associationas the presence of three or more of the following components: abdom-inal obesity, atherogenic dyslipidemia, elevated blood pressure, insulinresistance, prothrombotic state, and pro-inflammatory state (Grundyet al., 2004)) are important factors contributing to the high mortalityrate observed for patients with schizophrenia (Mitchell et al., inpress). It is unclear whether this association is simply reflective of anunhealthy lifestyle (e.g., smoking, physical inactivity, poor diet, andobesity) or whether weight gain and impaired glucose tolerance inschizophrenic patients is due to the side effects of atypical antipsychoticmedications, e.g., clozapine and olanzapine (Buchholz et al., 2008;Newcomer, 2007; Scheen and De Hert, 2007; Stahl et al., 2009).

tabolism interaction in schizophrenia: A chicken–egg dilemma, Prog16/j.pnpbp.2012.09.016

http://dx.doi.org/10.1016/j.pnpbp.2012.09.016

mailto:[email protected]


http://www.sciencedirect.com/science/journal/02785846


Table 1Important orexigenic (appetite stimulating) and anorexi-genic (appetite suppressing) factors that regulate foodintake and energy metabolism. These factors are pro-duced in the pancreas (insulin), adipose tissue (leptin),gastrointestinal tract (ghrelin, cholecystokinin/CCK,glucagon-like peptide-1/GLP-1, and peptide YY/PYY)and in the brain (neuropeptide Y/NPY, Agouti-relatedpeptide/AgRP, orexin-A, melanin concentrating hor-mone/MCH, pro-opiomelanocortin/POMC, cocaine- andamphetamine-regulated transcript/CART, thyrotropin re-leasing hormone/TRH, corticotropin releasing hormone/CRH, and oxytocin).

Orexigenicfactors

Anorexigenicfactors

InsulinLeptin

Ghrelin CCKGLP-1PYY

NPY POMCAgRP CARTOrexin-A TRHMCH CRH

Oxytocin

2 J. Steiner et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry xxx (2012) xxx–xxx

However, impaired fasting glucose tolerance has also been reported indrug-naïve patients with schizophrenia (Chiu et al., 2009; Guest et al.,2010; Kirkpatrick et al., 2009; Ryan et al., 2003; Saddichha et al.,2008; Spelman et al., 2007) and their unaffected siblings(Fernandez-Egea et al., 2008), suggesting disease-inherent abnormali-ties in glucose metabolism. Moreover, as summarized by McIntyre etal. (2005), schizophrenia-related insulin resistance has already beenobserved in several studies that were performed in thepre-neuroleptic era. For instance, increased fasting glucose levels or ab-normal oral glucose tolerance tests have been documented.

Interestingly, numerous previous studies have highlighted alter-ations in the immune system of patients with schizophrenia. Recentsystematic reviews and meta-analyses have provided evidence of al-tered cytokine and inflammation-related kynurenine pathway me-tabolite levels in the peripheral blood of schizophrenic patients(Miller et al., 2011; Myint, 2012; Potvin et al., 2008; Schwarcz et al.,2012). Increased concentrations of interleukin (IL)-1, IL-6, andtransforming growth factor-beta (TGF-β) appear to be state markers,whereas IL-12, interferon-gamma (IFN-γ), tumor necrosis factor-alpha (TNF-α), and soluble IL-2 receptor (sIL-2R) appear to be traitmarkers, as these protein levels remained elevated during acute exac-erbations and following antipsychotic treatment (Miller et al., 2011).Moreover, it has been proposed that the mononuclear phagocyte sys-tem (MPS)/microglial activation is involved in the pathogenesis ofearly acute disease phases (Busse et al., 2012; Doorduin et al., 2009;Drexhage et al., 2010; Nikkila et al., 1995; Steiner et al., 2006, 2008;van Berckel et al., 2008). Recently, it has been identified that centralnervous system (auto)antibodies directed against neurotransmitterreceptors, such as NMDA glutamate receptors or acetylcholine recep-tors, are produced in a subpopulation of patients with a clinical diag-nosis of schizophrenia (Borda et al., 2002; Steiner et al., in press;Tanaka et al., 2003; Tsutsui et al., 2012). This finding could help tobridge the gap between current immune and neurotransmitter hy-potheses of schizophrenia (Steiner et al., 2012a).

This review illustrates the potential interactions between the im-mune system and glucose metabolism disturbances in schizophrenia.We present three different perspectives to explain these findings:1) impaired cerebral glucose utilization is an important startingpoint in the pathogenesis of schizophrenia and leads to secondarychanges in peripheral glucose metabolism, an increased risk of car-diovascular complications, and accompanying pro-inflammatorychanges; 2) immune mechanisms may be involved from the begin-ning in the pathogenesis of schizophrenia, leading to disturbancesin glucose metabolism, giving rise to metabolic syndrome as aresulting secondary phenomenon; and 3) shared underlying factorsmay be responsible for the co-occurrence of immune system and glu-cose metabolism disturbances in schizophrenia.

1.2. Physiologic regulation of adipose tissue mass, food intake, andenergy expenditure

The various afferent inputs used by the brain to adjust food intakeand energy metabolism can be broadly subdivided into two groups,which are as follows: orexigenic (appetite stimulating) and anorexi-genic (appetite suppressing) factors (see Table 1).

Of the afferent signals reflecting the size of body adipose mass,insulin and leptin are the best studied and understood, and bothhormones appear to be required for the control of food intake,body weight, and metabolic homeostasis (Porte et al., 2005).Although leptin is secreted primarily from adipocytes and insulin isreleased from the endocrine pancreas, both hormones circulate at levelsproportionate to body fatmass and exert relatively long-term inhibitoryeffects on food intake. This signal is transduced via the inhibition ofneuropeptide Y (NPY) as well as Agouti-related peptide (AgRP) neu-rons, and the activation of pro-opiomelanocortin (POMC), cocaine-

Please cite this article as: Steiner J, et al, Immune system and glucose meNeuro-Psychopharmacol Biol Psychiatry (2012), http://dx.doi.org/10.10

and amphetamine-regulated transcript (CART) neurons of the hypotha-lamic arcuate nucleus (ARC) (Porte et al., 2005).

Satiety signals responding to recently ingested nutrients are morevaried and function primarily on a meal-to-meal basis to controlgastric emptying and the timing of meal initiation and termination.Unlike adiposity signals, these meal-related signals collectively regu-late the amount of energy consumed during individual meals but arenot generated in proportion to body energy stores. These short-termsignals, including cholecystokinin (CCK), glucagon-like peptide-1(GLP-1), Ghrelin and peptide YY (PYY), originate from the gastroin-testinal tract during a meal and reach the nucleus tractus solitarii(NTS) in the caudal brainstem via the vagus nerve (Grill and Kaplan,2002; Valassi et al., 2008). Ghrelin, an acylated peptide secreted bycells in the gastric mucosa, has orexigenic effects, stimulates foodintake and is implicated in meal initiation (Cummings and Foster,2003). In contrast, PYY, a close relative of NPY, is secreted primarilyfrom the distal small intestine and colon and appears to inhibitfeeding (anorexigenic effect) (Batterham et al., 2002). Afferent fibersproject from the NTS to the ARC, where satiety signals are integratedwith adiposity signals, namely leptin and insulin, and with severalhypothalamic and supra-hypothalamic inputs.

As previously mentioned, ARC neurons secrete orexigenic sub-stances, such as NPY and AgRP, and anorexigenic peptides, such asPOMC and CART. Other brain areas involved in the control of food intakeare located downstream from the ARC. These areas include theparaventricular nucleus, which produces anorexigenic peptides suchas thyrotropin releasing hormone (TRH), corticotropin-releasinghormone (CRH) and oxytocin; the lateral hypothalamus; and theperifornical area, which secrets the orexigenic substances orexin-Aand melanin-concentrating hormone (MCH) (Valassi et al., 2008).

Evidence suggests that many of these hormones and hypothalamicsystems also regulate energy expenditure. The importance of insulinfor thermogenesis was originally inferred from pharmacologicalstudies in rats that demonstrated increased body temperature and en-ergy expenditure, as well as reduced food intake when insulin wasinjected into the hypothalamic ventromedial and paraventricular nuclei(McGowan et al., 1992; Menendez and Atrens, 1991). These pharmaco-logical studies suggest that the action of insulin in the hypothalamussimultaneously reduces food intake while increasing sympathetic ner-vous system outflow to brown adipose tissue to produce heat fromfatty acid oxidation and increase energy expenditure.



1 Under certain conditions, protons can re-enter the mitochondrial matrix withoutcontributing to ATP synthesis. This process is known as a proton leak, or mitochondrialuncoupling, and is due to the facilitated diffusion of protons into the matrix. The pro-cess results in the production of heat and is mediated by a proton channel calledthermogenin, or uncoupling protein 1 (UCP1).

3J. Steiner et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry xxx (2012) xxx–xxx

2. The “selfish-brain” theory and its implications for schizophrenia:impaired cerebral glucose utilization in schizophrenia leads tosecondaryperipheralmetabolic changes and increased cardiovascularrisk, including pro-inflammatory changes

2.1. Physiology of the cerebral glucose metabolism

The brain has the highest metabolic activity of all organs. Its uptakerates during physical rest are approximately 15% of the heart minutevolume, 20% of the body's oxygen supply, and 50% of the body's glucosesupply, while the brain-to-body mass ratio is only approximately 2–3%(Magistretti et al., 1995). Other important glucose “consumers” includeskeletal and heart muscles (particularly during physical activity), aswell as red blood cells. Due to its high need, continuous cerebral glucosesupply is important for intact brain function, as the brain has very lim-ited energy storage compounds (i.e., it has a small glycogen reservoir).Alternatively, lactate, which is produced by anaerobic glycolysis duringphysical activity inmuscle cells, and ketone bodies, which are producedfrom the degradation of fatty acids during fasting, can be used as energyfor neurons (Magistretti et al., 1995).

Active insulin transportation at the blood–brain barrier links thebrain's glucose metabolism with pancreatic insulin production(Banks, 2006). Insulin and insulin receptors are not only expressedin the hypothalamus, as previously assumed, but are also distributedthroughout the central nervous system (Frölich et al., 1998; Wanget al., 2009). Findings by Steen et al. (2005) showed that patientswith Alzheimer's disease have impairments in insulin signal transduc-tion cascade, including insulin, insulin receptors, and Akt1, particularlyin the brain. The term “type 3” diabetes was coined to describe distur-bances in insulin signaling with a strong focus on the brain (Steenet al., 2005). Similar mechanisms may be relevant in schizophrenia(see below). Physiologically, insulin facilitates cellular glucose uptakeinto brain tissue, while systemically it regulates body weight, energydisposal, food intake and locomotion (Bruning et al., 2000). Thus,impaired cerebral insulin receptor signaling may cause disturbances inthe uptake of glucose into brain tissue, resulting in cellular glucosedeprivation.

2.2. Disturbances of cerebral glucose metabolism in schizophrenia

Several researchers have aimed to assess the glucose metabolismin the brain of patients with schizophrenia. For example, glucose me-tabolism has been studied via biochemical analyses of post-mortemtissues, measurements of glucose levels in the cerebrospinal fluid(CSF), or in vivo imaging studies. Cerebral insulin signaling is quiteaffected by schizophrenia (see the above-mentioned notion of“type 3 diabetes”), as evidenced by the decreased expression of cere-bral insulin receptors (β-subunit), the reduced activity of the signaltransduction protein Akt1, and the diminished neuronal expressionof insulin-degrading enzyme in the dorsolateral prefrontal cortex(Bernstein et al., 2009; Emamian et al., 2004; Zhao et al., 2006).These changes may cause disturbances in neural glucose uptake andutilization, as suggested by measurements of elevated CSF glucoselevels (Holmes et al., 2006). In vivo fluorodeoxyglucose positronemission tomography (FDG-PET) and functional magnetic resonanceimaging (fMRI) studies also demonstrate impaired cerebral glucoseutilization in brain areas that are important in the pathogenesis ofschizophrenia, resulting in the “metabolic disconnection” of thedorsolateral prefrontal cortex and mediodorsal thalamus with thelimbic system (Buchsbaum et al., 2007; Mitelman et al., 2005). Themediotemporal region and the hippocampus in particular appear tobe of specific importance for the cognitive impairments noted inschizophrenia. Studies addressing neuronal activation using fMRIhave demonstrated that schizophrenic subjects show impairedpatterns of hippocampal activity in novelty detection, declarativelearning, and memory tasks (Ragland et al., 2006; Weiss et al., 2004;


Zierhut et al., 2010). Notably, it has been demonstrated thathippocampus-dependent memory performance can be improved bythe administration of glucose in rodents and humans (Gold et al.,1986; Sunram-Lea et al., 2002). In humans, it has been shown thatthis effect is more pronounced when the task is cognitively demanding(Sunram-Lea et al., 2002) or when the cognitive resources that can beapplied for it are limited, as is the case in the elderly and in schizophren-ic patients (Fucetola et al., 1999; Newcomer et al., 1999). Nevertheless,the defective glucose utilization noted in functional studies may resultfrom either defective cellular mechanisms of glucose utilization or itmay be a secondary effect of lower neuronal activation.

2.3. The “selfish brain” theory and schizophrenia

An interpretation of the aforementioned findings in the context ofthe “selfish brain” theory is intriguing. This theory describes the tenden-cy of the human brain to cover its own, comparably high energyrequirements with the utmost priority when regulating energy fluxeswithin an organism. In this respect, the brain is thought to behaveselfishly (Peters et al., 2004, 2007). Given the assumption of disruptedcerebral energy supply in schizophrenia due to cerebral insulin resis-tance, the “selfish brain” theory may provide a novel explanation forthe preferred intake of high carbohydrate, high fat foods and dimin-ished physical activity and exercise in schizophrenic patients as a wayto improve the brain's energy supply. Thus, one could speculate thatincreased blood glucose levels following treatment with atypicalantipycotic drugs are not an “unwanted side effect” after all but mightbe therapeutic to improve cerebral glucose supply.

2.3.1. Adaptive behavioral changes and atypical antipsychotic drugs mayinduce counter-regulatory mechanisms to overcome impaired cerebralglucose metabolism: a short-term perspective

According to the “selfish brain” theory, impaired glucose supply tothe cerebral cortex, which could be caused by cerebral disturbances ininsulin signaling, may lead to counter-regulatory adaptivemechanisms,such as the above-mentioned changes in food intake and physical activ-ity. The most effective drugs, clozapine and olanzapine (Ebenbichleret al., 2003; Lieberman et al., 2005; Newcomer, 2007), may improveenergy supplies via several different mechanisms, which are as follows:1) appetite stimulation, possibly via serotonergic mechanisms (5HT2C-antagonism), could lead to increased food intake; 2) the antihistaminicproperties of these drugs cause sedation, daytime sleepiness orprolonged nighttime sleep and result in an overall reduction in physicalactivity, which subsequently decreases energy expenditure; 3) previousanimal experiments suggest that antipsychotic drugs may amelioratethe action of insulin on brain tissue, e.g., via the increased phosphoryla-tion of Akt1 (Emamian et al., 2004; Sutton and Rushlow, 2011) — thismechanism could improve cerebral glucose uptake and utilization;and 4) clozapine increases the span of mitochondrial respiration thatcan be used for ATP synthesis (unpublished data, J. Steiner) via a de-crease in leak respiration,1 which contributes to thermogenesis (Mozoet al., 2005; Rolfe and Brown, 1997). This mechanism may promotereduced energy expenditure. Consistent with this assumption, reducedbody temperature has been observed in clozapine- and olanzapine-treated patients (olanzapine is pharmacologically similar to clozapine)and may contribute to the side-effect of increased weight gain, whilethe increased efficacy of mitochondrial respiration may be beneficialwith respect to supplying the brain with energy (Evers et al., 2010;Salmi et al., 1994; Stefanidis et al., 2009). In summary, several




mechanisms may lead to improved cerebral glucose supply, includingthe more efficient utilization of glucose.

2.3.2. Adaptive behavioral changes and atypical antipsychotic drugs mayinduce metabolic syndrome, type 2 diabetes and a systemicpro-inflammatory state: a long-term perspective

According to the “selfish brain” theory, the above-mentionedadaptive behavioral changes and the effects of atypical antipsychoticdrugs on food intake and energy expenditure are “selfish”, as thebrain's energy supply is improved. However, the long-term systemicside effects of atypical antipsychotics include obesity, metabolicsyndrome, and the development of type 2 diabetes. Indeed, the mosteffective drugs, clozapine and olanzapine, are those with the highestdiabetogenic risk and lead to a mean weight gain of approximately10 kg within the first year of treatment (Ebenbichler et al., 2003;Lieberman et al., 2005; Newcomer, 2007).

In the long term, the risk of developing overweight and particularlyvisceral obesity is increased in schizophrenia due to disease-inherentand drug-induced mechanisms (see above). Visceral obesity is a chron-ic, low-grade inflammatory disease that predisposes people tometabol-ic syndrome, type 2 diabetes and its cardiovascular complications.Adipose tissue is not a passive storehouse for fat; rather, it is an endo-crine organ that synthesizes and releases a variety of bioactive mole-cules, some of which are produced by local inflammatory cells,including macrophages (Bories et al., 2012). MPS cells located in thevisceral fat are an important site for IL-1β, IL-6, and TNF-α secretion,and they provide a potential link between abdominal obesity and sys-temic inflammation (Fontana et al., 2007; Meshkani and Adeli, 2009;Xu et al., 2003). In particular, IL-6 seems to induce insulin resistanceand an increased production of C-reactive protein (CRP) in the liver(Meshkani and Adeli, 2009; Xu et al., 2003). TNF-αmay also induce in-sulin resistance by suppressing the expression of the insulin-sensitiveglucose transporter 4 (GLUT4) in peripheral tissues (Hauner et al.,1995; Lofgren et al., 2000). Shim et al. (2006) observed a higher whiteblood cell count, including increased numbers of monocytes in theblood of patients with metabolic syndrome and type 2 diabetes. MPSactivation along with increased levels of IL-1β, IL-6, and CRP have alsobeen detected in schizophrenia; however, many of these studies didnot control for the important confounding factor of visceral obesity(Drexhage et al., 2010; Miller et al., 2011; Potvin et al., 2008; Sicras-Mainar et al., in press). Thus, it remains unclear whether such cytokinechanges reflect schizophrenia-specific pathophysiological changes orwhether these changes are related to visceral obesity and the develop-ment of metabolic syndrome. While elevated levels of CRP havebeen considered to be cardiovascular risk markers in schizophrenia(Sicras-Mainar et al., in press), Beumer et al. (in press) recentlyperformed a study that focused on the potential link betweenpro-inflammatory cytokines, schizophrenia and obesity. Compared tohealthy controls, IL-1β and IL-6 levelswere increased in the schizophre-nia cohort, even when considering Body-Mass-Index (BMI) as a poten-tial confounder. However, unfortunately, BMI values were onlyavailable in some of the tested healthy controls, and waist circumfer-ence data were not presented in this study. The authors concludedthat increased MPS activation is a key component in the pathophysiol-ogy of schizophrenia. Additional studies are necessary to validate thisfinding.

Several studies have observed increased levels of S100B in theperipheral blood and cerebrospinal fluid of schizophrenic patients.This finding has been attributed primarily to astrocyte damage ordysfunction (Rothermundt et al., 2001, 2004; Schroeter and Steiner,2009). However, meanwhile, it is known that peripheral S100Bblood levels correlate with body mass and the development of insulinresistance in schizophrenic patients (Steiner et al., 2010a,b). According-ly, we observed a negative correlation between body mass and the“counter player” and competitive inhibitor of S100B, namely, the solu-ble receptor for advanced glycation products (sRAGE) (Donato, 2007;


Steiner et al., 2009). Notably, S100B is widely expressed in adipocytes(Goncalves et al., 2010; Netto et al., 2006) and may provide a linkbetween disturbances in glucose metabolism and pro-inflammatorychanges, as S100B is capable of activating the MPS, including microglialcells (Adami et al., 2004; Bianchi et al., 2007; Steiner et al., 2011, 2012b).

The historic use of insulin coma therapy, which was introduced in1933 by Manfred Sakel, has been widely used for the treatment ofmajor psychiatric disorders, including schizophrenia (Crammer,2000). The mechanism of action has not yet been elucidated; however,this therapy could be helpful in overcoming cerebral insulin resistanceby improving the cellular glucose supply for nervous tissue. Similarly,studies are currently underway to determine if other drugs may behelpful in improving insulin signaling using less harmful interventions.Several studies have shown that treatment with the antidiabetic drugmetformin reduces the risk of systemic metabolic side effects inolanzapine-treated patients (Bushe et al., 2009). Moreover, the“glitazones”, a group of insulin sensitizing drugs, have been tested inthis context and have shown promising results (Baptista et al., 2009;Edlinger et al., 2007; Henderson et al., 2009). These and other com-pounds may improve the efficacy of treatment and reduce the systemiccardiovascular side effects of atypical antipsychotic drugs when used infuture combination therapies.

3. Immune mechanisms may be involved as primary factors in thepathogenesis of schizophrenia, leading to disturbances in glucosemetabolism as a secondary phenomenon

In contrast to the above-mentioned view that immune systemchanges andmetabolic dysfunction represent a secondary phenomenonin schizophrenia as a result of disturbances in cerebral glucose utiliza-tion, recently published animal experiments point to a different per-spective. In this section, we summarize data from animal experimentsthat suggest a role for prenatal immune activation in the triggering ofschizophrenia-related MPS/microglial activation and the developmentof metabolic syndrome. Stimulated by the work of Barker et al.(Dover, 2009), the concept of “early-life priming of adult disease” isnowwidely accepted. This concept refers to the phenomenon that envi-ronmental acting during sensitive embryonic developmental periodscan induce lifelong changes in physiologic, metabolic, and behavioralfunctions, as well as in psychiatric disorders (Bale et al., 2010; Luoet al., 2010).

3.1. Prenatal immune activation is associated with an increased risk ofdeveloping schizophrenia

Accordingly, the neurodevelopmental hypothesis of schizophreniapostulates that the occurrence of disturbances in brain maturation inthe second trimester may result in limbic dysfunction and structuralchanges, such as decreased hippocampal and cortical volume due toneuropil reduction (Weinberger, 1987). Genetic vulnerability isbelieved to interact with inflammatory and immunological reactionsthat interfere with brain development during pregnancy. In severalstudies, the children of mothers who suffered from influenza, cytomeg-alovirus, or herpes virus infections during pregnancy appeared to havean increased risk of developing schizophrenia (Brown, 2008; Torrey etal., 2006; Yolken and Torrey, 1995). Polyriboinosinic–polyribocytidilicacid (PolyI:C) is an analog of double-stranded RNA that stimulates acytokine-associated acute phase response, as it is a virus-like mimetic(Fatemi et al., 2002, 2008). Its administration inmice during vulnerableperiods of pregnancy induces changes in brain morphology, physiologyand neurochemistry, as well as inducing behavioral changes that arepartly reminiscent of human schizophrenia after puberty in the off-spring of female mice (Meyer and Feldon, 2009; Winter et al., 2009).This induced immune activation results in enhanced dopamineturnover in the prefrontal cortex and a decrease in serotonin levels inseveral subcortical regions of the rat brain. The authors concluded




that in utero inflammation predisposes the offspring to persistent neu-rotransmitter changes and thus to the development of schizophrenialater in life. Of note, many of these changes may be prevented bytreating rats with risperidone during adolescence (Piontkewitz et al.,2012).

Post-mortem and PET analysis showed an increased number ofactivated microglial cells in patients with schizophrenia (Bayer et al.,1999; Busse et al., 2012; Doorduin et al., 2009; Radewicz et al., 2000;Steiner et al., 2008; van Berckel et al., 2008). Therefore, it has been hy-pothesized that these cells contribute to disease pathogenesis andmay actively be involved in the gray matter loss observed in schizo-phrenia. Juckel et al. analyzed the number and shape of microglia inthe offspring of mice exposed to PolyI:C during pregnancy and ob-served higher microglial densities in the hippocampus and striatum,but not in the frontal cortex at postnatal day 30, which is similar toadolescence in humans (Juckel et al., 2011). The microglia of the off-spring of PolyI:C-treated mothers were morphologically character-ized by reduced arborization, which is indicative of a state ofhigher activation compared to the offspring of vehicle-treated mice(Juckel et al., 2011). This study supports the hypothesis of “early-lifepriming of adult disease” in schizophrenia, in which maternal infec-tion during embryogenesis contributes to MPS/microglial activationduring adulthood.

4. Prenatal immune activation is associated with an increased riskof developing schizophrenia-related metabolic alterations and anincreased pro-inflammatory cytokine production

Furthermore, U. Meyer's group used the same animal model to testwhether schizophrenia-related metabolic alterations and increasedpro-inflammatory cytokine production can be primed byprenatal expo-sure to PolyI:C (Pacheco-Lopez et al., in press). They usedhigh-resolution computed tomography of fat distribution and indirectcalorimetry along with an oral glucose tolerance test and performedplasma cytokine and cortisol measurements. Prenatal immune activa-tion caused altered glycemic regulation and abnormal ingestive behav-ior, which led to adult onset of excess visceral and subcutaneous fatdeposition. These effects were accompanied by age-dependent changesin the peripheral secretion of pro-inflammatory (IL-6 and TNF-α) and Tcell-related (IL-2 and IFN-γ) cytokines and stress-axis activation.

Thus, schizophrenia-related metabolic abnormalities and in-creased pro-inflammatory cytokine production may be primed byprenatal virus-like immune activation. However, the exact mecha-nisms are still unclear. It will be quite difficult to substantiate thesefindings in schizophrenia patients, as such studies in humans are pri-marily restricted to an epidemiological approach due to ethicalrestrictions.

5. Sharedunderlying factorsmaybe responsible for the co-occurrenceof immune system and glucose metabolism disturbancesin schizophrenia

5.1. Clustering of schizophrenia, thyroid autoimmune disease and type 1diabetes: an aberrant immune activation set point for local mononuclearphagocytes and microglial cells

Autoimmune thyroid disease (AITD) and autoimmune type 1 diabe-tes co-occur frequently and the co-occurrence of AITD and type 1 diabe-tes is known as subtype 3A of a syndrome called autoimmunepolyendocrine syndrome (Betterle and Zanchetta, 2003). While AITDmay present as a schizophrenia-like disorder, i.e., as Hashimoto'sencephalitis (Lin and Liao, 2009), a large Danish national study showedthat patients with schizophrenia had a 45% higher chance of developingan autoimmune disease, including thyroid autoimmune disease (Eatonet al., 2010). Moreover, these autoimmune diseases were more preva-lent in first-degree relatives of patients with schizophrenia.


The finding of a relatively high familial co-occurrence of psychiatricdisorders and endocrine and other organ-specific autoimmune diseasesrefutes the concept that psychiatric disorders and endocrine autoim-mune disease are the cause or consequence of each other and implythe existence of shared immune pathogenic factors for schizophreniaand endocrine autoimmune diseases. One of these shared factors isthought to be an intrinsically high activation set point for the MPS.Microglia are the representatives of theMPS in the brain, while dendrit-ic cells andmacrophages are the representatives of this system in endo-crine organs and monocytes are the representatives of this system inthe circulation.

There is mounting evidence that dendritic cells and macrophagesin the thyroid and pancreatic islets of patients with AITD and autoim-mune diabetes exist in an aberrant pro-inflammatory state. Thisabnormal set point of cells most likely plays a role in breaking T celltolerance and inducing an aberrant autoimmune response towardsthyrocytes and β cells. An intrinsic defect in T regulatory cell functionadds to this problem, leading to a further imbalance between T effectorand T regulator forces (unpublished data, H.A. Drexhage).

Under steady state conditions, immature forms of dendritic cells(iDCs), due to their strong endocytic capability, also take up compoundspresent in the vicinity, such as thyroglobulin when in the thyroid andinsulin when in the pancreatic islets. After the uptake of these“auto-antigens”, the iDCs mature partly under the influence of localcytokines (such as TGF-β) to semi-mature DCs and travel via the lym-phatics to the draining lymph nodes (known there as interdigitatingcells) while carrying the auto-antigens. In the draining lymph nodes,the superb antigen-presenting capability of the semi-mature DCsbecomes evident, and they begin to trigger and expand in particularsubsets of auto-reactive natural T regulatory cells (Sakaguchi et al.,2008; Steinman et al., 2003). In triggering this subpopulation ofCD4+ T cells, DCs build up a strong non-reactivity (=tolerance)towards self under steady-state conditions.

Tissue macrophages in the brain, known as microglia, have similarregulatory functions as dendritic cells and macrophages in endocrinetissues. Animal studies have shown that microglia do not differentiatefrom circulating monocytes, as originally thought, but differ insteadfrom primitive myeloid progenitors that emigrate from the yolk sacinto the brain parenchyma (Alliot et al., 1999; Ginhoux et al., 2010).Microglia participates in various aspects of brain development,including developmental cell death, axon remodeling, synaptogenesis,and synaptic pruning (Pont-Lezica et al., 2011; Schlegelmilch et al.,2011; Tremblay and Majewska, 2011).

A switch to a high immune activation set point of steady-stateglandular dendritic cells and microglia may alter their role in the reg-ulation of growth and the function of neighboring parenchymal cells,while reduced auto-antigen tolerance could trigger the co-occurrenceof schizophrenia, thyroid autoimmune disease and type 1 diabetes(unpublished data, H.A. Drexhage). As stated before, histologic andPET studies observed signs of microglial activation in schizophrenia(Bayer et al., 1999; Busse et al., 2012; Doorduin et al., 2009;Radewicz et al., 2000; Steiner et al., 2006, 2008; van Berckel et al.,2008) and an accumulation of monocytes and macrophages in thecerebrospinal fluid of patients with schizophrenia during acute psy-chotic episodes. These findings support the MPS activation hypothesisin schizophrenia (Nikkilä et al., 1999). Consistent with the idea of animpaired auto-antigen tolerance, more T cells have been observed inthe cerebrospinal fluid and in the hippocampal brain tissue of schizo-phrenia patients during later disease stages (Busse et al., 2012;Maxeiner et al., 2009). With regard to the peripheral circulation,there is evidence for increased numbers of pro-inflammatory IL-17-producing cells (and, to a lesser extent, IL-4-producing cells), predomi-nantly in younger and active schizophrenia cases (unpublished data,H.A. Drexhage). Similarly, the activation of dendritic cells, macrophagesand various subtypes of T cells have been observed in the thyroid andpancreatic islets of patients affected by AITD and autoimmune type 1




diabetes (unpublished data, H.A. Drexhage). Elimination of the dendrit-ic cells and T cells from the thyroid or pancreatic islets in animalmodelsof endocrine autoimmune disease, e.g., via knock-outs or treatmentwith clodronate or T cell-depleting therapies, results in a halting ofthe autoimmune process. Therefore, it may be assumed that theseimmune cells are the local force of the autoimmune reaction. It hasalso been assumed that it is not a numerical deficit, but a defect in thefunction of T regulator cells (possibly an intrinsically reduced IL-2signaling), which fails to keep the locally activated dendritic cells, Teffector cells and macrophages in check (unpublished data, H.A.Drexhage).

5.2. Shared genetic predisposition for schizophrenia and a dysfunctionalimmune system

Perhaps the most compelling evidence for the link betweenschizophrenia and a dysfunctional immune system is provided bygenetic studies (Leonard et al., 2012). For example, Badenhoop et al.(1996), Lindholm et al. (1999) and Stefansson et al. (2009) demon-strated that a gene locus on chromosome 6p22 was linked to bothschizophrenia and to the genes of the major histocompatibilitycomplex (MHC) system, a system that is involved in one of themost im-portant functions of the immune system, i.e. antigen presentation.There is also a genetic link between schizophrenia and type 1 diabetes(Lindholm et al., 1999).

5.3. Shared genetic predisposition for schizophrenia and type 2 diabetes

It is now well established that individuals with schizophrenia are atan increased risk for the development of type 2 diabetes, with estimatessuggesting a prevalence of between 15% and 20% in this population(Gough andO'Donovan, 2005).While shared environmental risk factors(see above) may explain the link between the two disorders, a sharedgenetic link is an alternative explanation.

Many chromosomal regions that have been associated withschizophrenia show an overlapping linkage with type 2 diabetes(Gough and O'Donovan, 2005). The most promising region of interestin both schizophrenia and type 2 diabetes is located on chromosome1q21–25. These genes include LMNA (which encodes lamin A/C)(Hegele et al., 2000), mutations of which can cause lipodystrophyand diabetes. The insulin receptor-related gene (IRR) also maps tothis region (Hirayama et al., 1999) and encodes a protein involvedin insulin signaling. Association analysis of chromosome 1q21–22fine mapping has identified the CAPON (C-terminal PDZ domainligand of neuronal nitric oxide synthase) gene (Brzustowicz et al.,2004). The encoded protein functions as an adapter protein targetingneuronal nitric oxide synthase. The regulator of G-protein signaling 4(RGS4) gene is another good positional candidate in this region forschizophrenia (Chowdari et al., 2002). This gene is a negative regula-tor of G protein-coupled receptors, and evidence also suggests thatRGS4 may modulate activity at certain serotonergic and metabotropicglutamatergic receptors (Gough and O'Donovan, 2005).

5.4. Disturbances in Akt1 signaling in schizophrenia: possibleconsequences for the brain and for the immune system

The serine–threonine protein kinase Akt1 is important for the signaltransduction of a variety of growth factors and insulin. It has beensuggested that cerebral and peripheral insulin signaling in schizophre-nia may be disrupted by an altered expression and phosphorylation ofAkt1 (Emamian et al., 2004; van Beveren et al., 2012; Zhao et al., 2006).

Akt1 plays an important role in intact prefrontal cortex function (Laiet al., 2006). Moreover, insulin and insulin-like growth factor-1 (IGF-1)stimulate the maturation of oligodendrocytes in culture (van der Palet al., 1988). Accordingly, Akt1 signaling plays an important role inmyelination processes (Narayanan et al., 2009) and is involved in


synaptic plasticity and myelination. Thus, the reduced expression andphosphorylation of Akt1 in the dorsolateral prefrontal cortex may con-tribute to the impaired connectivity of different brain regions, whichhas been observed in schizophrenia. Indeed, an approximately 75% re-duction in Akt1 content and activity in the postmortem brain tissue ofschizophrenia patients was observed by Zhao et al. (2006). This findingis consistent with a previous study by Emamian et al. (2004), whichreported a significant association between schizophrenia and an Akt1haplotype that is associatedwith lowerAkt1 protein levels and a greatersensitivity to the sensorimotor gating-disruptive effect of amphet-amine. These data support the proposal that alterations in Akt1 signal-ing contribute to the pathogenesis of schizophrenia and identify Akt1as a potential schizophrenia susceptibility gene. Consistent with thisproposal, Emamian et al. (2004) and Sutton et al. (Sutton andRushlow, 2011) also showed improved phosphorylation of Akt1 in thebrains of rodents after treatment with clozapine or haloperidol, whichcould compensate for the impaired functioning of this signaling path-way in schizophrenia. Interestingly, Emamian et al. (2004) found simi-lar alterations in Akt1 signaling in the lymphocytes of schizophrenicpatients. This finding was replicated by van Beveren et al. (2012) inperipheral blood mononuclear cells (PBMCs) of patients with schizo-phrenia. Based on these findings, we speculate that Akt1 may providea potential link for the immune system pathology and glucose metabo-lism alterations noted in patient with schizophrenia. The consequencesof impaired Akt1 on the immune system of patients with schizophreniaare currently unknown. Recent studies on the physiological role of Akt1suggests that it not only regulates neutrophil and monocyte migrationinto inflamed tissues (Di Lorenzo et al., 2009), thymic development(Fayard et al., 2007; Juntilla et al., 2007), and peripheral B-cell matura-tion and survival (Calamito et al., 2010), but that Akt1 also plays animportant role in the differentiation of cells in the MPS and the regula-tion of their inflammatory/tolerogenic state (Van den Bossche et al.,2012).

6. Summary and conclusion

Impaired glucose metabolism and the development of metabolicsyndrome contribute to a reduction in the average life expectancy ofindividuals with schizophrenia. It has been questioned whether thisassociation is simply reflective of an unhealthy lifestyle or whetherweight gain and impaired glucose tolerance in patients with schizo-phrenia are attributable to the side effects of atypical antipsychoticmedications or disease-inherent abnormalities. In addition, numerousprevious studies have highlighted the alterations in immune systemfunctioning among patients with schizophrenia. Increased concentra-tions of IL-1, IL-6, and TGF-β appear to be state markers, whereasIL-12, IFN-γ, TNF-α, and sIL-2R appear to be trait markers of thedisease. Moreover, mononuclear phagocyte system (MPS)/microglialactivation are involved in the pathogenesis of the early acute phasesof the disease.

It is currently unclear whether impaired cerebral glucose utilizationleads to secondary disturbances in peripheral glucose metabolism, anincreased risk of cardiovascular complications and accompanyingpro-inflammatory changes in schizophrenia. Based on the currentlyavailable data, it is more likely that immune system dysfunction andaberrations in glucose metabolism interact as co-occurring factorsin patients with schizophrenia. Longitudinal studies focusing onimmune- and glucose metabolism-related parameters are needed toclarify the time course of the above-mentioned alterations. RepeatedPET scans, indirect calorimetry, the analysis of peripheral blood andCSF, and provocation maneuvers, such as oral glucose tolerance tests,may be helpful in revealing pathophysiological changes that are in-volved in the pathogenesis of schizophrenia. In this context, future anal-yses should also focus on prodromal patients and first-degree relativesto allow for the discrimination of risk from changes that are associatedwith full-blown psychosis.




References

Adami C, Bianchi R, Pula G, Donato R. S100B-stimulated NO production by BV-2 mi-croglia is independent of RAGE transducing activity but dependent on RAGE extra-cellular domain. Biochim Biophys Acta 2004;1742:169–77.

Alliot F, Godin I, Pessac B. Microglia derive from progenitors, originating from the yolksac, and which proliferate in the brain. Brain Res Dev Brain Res 1999;117:145–52.

Badenhoop K, Tonjes RR, Rau H, Donner H, Rieker W, Braun J, et al. Endogenous retro-viral long terminal repeats of the HLA-DQ region are associated with susceptibilityto insulin-dependent diabetes mellitus. Hum Immunol 1996;50:103–10.

Bale TL, Baram TZ, Brown AS, Goldstein JM, Insel TR, McCarthy MM, et al. Early life pro-gramming and neurodevelopmental disorders. Biol Psychiatry 2010;68:314–9.

Banks WA. Blood–brain barrier and energy balance. Obesity (Silver Spring)2006;14(Suppl. 5):234S–7S.

Baptista T, Rangel N, El Fakih Y, Uzcategui E, Galeazzi T, Beaulieu S, et al. Rosiglitazone inthe assistance of metabolic control during olanzapine administration in schizophrenia:a pilot double-blind, placebo-controlled, 12-week trial. Pharmacopsychiatry 2009;42:14–9.

Batterham RL, Cowley MA, Small CJ, Herzog H, Cohen MA, Dakin CL, et al. Gut hormonePYY(3–36) physiologically inhibits food intake. Nature 2002;418:650–4.

Bayer TA, Buslei R, Havas L, Falkai P. Evidence for activation of microglia in patientswith psychiatric illnesses. Neurosci Lett 1999;271:126–8.

Bernstein HG, Ernst T, Lendeckel U, Bukowska A, Ansorge S, Stauch R, et al. Reducedneuronal expression of insulin-degrading enzyme in the dorsolateral prefrontalcortex of patients with haloperidol-treated, chronic schizophrenia. J PsychiatrRes 2009;43:1095–105.

Betterle C, Zanchetta R. Update on autoimmune polyendocrine syndromes (APS). Actabio-medica: Atenei Parmensis 2003;74:9-33.

Beumer W, Drexhage RC, DeWit H, Versnel MA, Drexhage HA, Cohen D. Increased levelof serum cytokines, chemokines and adipokines in patients with schizophrenia isassociated with disease and metabolic syndrome. Psychoneuroendocrinology inpress.

Bianchi R, Adami C, Giambanco I, Donato R. S100B binding to RAGE in microglia stim-ulates COX-2 expression. J Leukoc Biol 2007;81:108–18.

Borda T, Perez Rivera R, Joensen L, Gomez RM, Sterin-Borda L. Antibodies against cere-bral M1 cholinergic muscarinic receptor from schizophrenic patients: molecularinteraction. J Immunol 2002;168:3667–74.

Bories G, Caiazzo R, Derudas B, Copin C, Raverdy V, Pigeyre M, et al. Impaired alterna-tive macrophage differentiation of peripheral blood mononuclear cells from obesesubjects. Diabetes & vascular disease research: Official Journal of the InternationalSociety of Diabetes and Vascular Disease 2012;9:189–95.

Brown AS. The risk for schizophrenia from childhood and adult infections. Am J Psychi-atry 2008;165:7-10.

Bruning JC, GautamD, Burks DJ, Gillette J, SchubertM, Orban PC, et al. Role of brain insulinreceptor in control of body weight and reproduction. Science 2000;289:2122–5.

Brzustowicz LM, Simone J, Mohseni P, Hayter JE, Hodgkinson KA, Chow EW, et al. Link-age disequilibriummapping of schizophrenia susceptibility to the CAPON region ofchromosome 1q22. Am J Hum Genet 2004;74:1057–63.

Buchholz S, Morrow AF, Coleman PL. Atypical antipsychotic-induced diabetes mellitus:an update on epidemiology and postulated mechanisms. Intern Med J 2008;38:602–6.

Buchsbaum MS, Buchsbaum BR, Hazlett EA, Haznedar MM, Newmark R, Tang CY, et al.Relative glucose metabolic rate higher in white matter in patients with schizophre-nia. Am J Psychiatry 2007;164:1072–81.

Bushe CJ, Bradley AJ, Doshi S, Karagianis J. Changes in weight and metabolic parametersduring treatment with antipsychotics and metformin: do the data inform as to po-tential guideline development? A systematic review of clinical studies. Int J ClinPract 2009;63:1743–61.

Busse S, Busse M, Schiltz K, Bielau H, Gos T, Brisch R, et al. Different distribution pat-terns of lymphocytes and microglia in the hippocampus of patients with residualversus paranoid schizophrenia: further evidence for disease course-related im-mune alterations? Brain Behav Immun 2012;26:1273–9.

Calamito M, Juntilla MM, Thomas M, Northrup DL, Rathmell J, Birnbaum MJ, et al. Akt1and Akt2 promote peripheral B-cell maturation and survival. Blood 2010;115:4043–50.

Chiu HJ, Loh el W, Lan TH, Sou JH, Hsieh HC, Hu TM. Increase of incidence of impairedfasting glucose in young schizophrenia patients. Psychiatry Clin Neurosci 2009;63:127–8.

Chowdari KV, Mirnics K, Semwal P, Wood J, Lawrence E, Bhatia T, et al. Association andlinkage analyses of RGS4 polymorphisms in schizophrenia. Hum Mol Genet2002;11:1373–80.

Crammer JL. Insulin coma therapy for schizophrenia. J R Soc Med 2000;93:332–3.Cummings DE, Foster KE. Ghrelin-leptin tango in body-weight regulation. Gastroenterology

2003;124:1532–5.Di Lorenzo A, Fernandez-Hernando C, Cirino G, Sessa WC, et al. Akt1 is critical for acute

inflammation and histamine-mediated vascular leakage. Proc Natl Acad Sci U S A2009;106:14552–7.

Donato R. RAGE: a single receptor for several ligands and different cellular responses:the case of certain S100 proteins. Curr Mol Med 2007;7:711–24.

Doorduin J, et al. Neuroinflammation in schizophrenia-related psychosis: a PET study.J Nucl Med 2009;50:1801–7.

Dover GJ. The Barker hypothesis: how pediatricians will diagnose and prevent commonadult-onset diseases. Trans Am Clin Climatol Assoc 2009;120:199–207.

Drexhage RC, et al. The mononuclear phagocyte system and its cytokine inflammatorynetworks in schizophrenia and bipolar disorder. Expert Rev Neurother 2010;10:59–76.


Eaton WW, et al. Autoimmune diseases, bipolar disorder, and non-affective psychosis.Bipolar Disord 2010;12:638–46.

Ebenbichler CF, et al. Olanzapine induces insulin resistance: results from a prospectivestudy. J Clin Psychiatry 2003;64:1436–9.

Edlinger M, et al. Treatment of antipsychotic-associated hyperglycemia withpioglitazone: a case series. J Clin Psychopharmacol 2007;27:403–4.

Emamian ES, et al. Convergent evidence for impaired AKT1–GSK3beta signaling inschizophrenia. Nat Genet 2004;36:131–7.

Evers SS, et al. Olanzapine causes hypothermia, inactivity, a deranged feeding patternand weight gain in female Wistar rats. Pharmacol Biochem Behav 2010;97:163–9.

Fatemi SH, et al. Human influenza viral infection in utero alters glial fibrillary acidic pro-tein immunoreactivity in the developing brains of neonatal mice. Mol Psychiatry2002;7:633–40.

Fatemi SH, et al. Maternal infection leads to abnormal gene regulation and brain atro-phy in mouse offspring: implications for genesis of neurodevelopmental disorders.Schizophr Res 2008;99:56–70.

Fayard E, et al. Deletion of PKBalpha/Akt1 affects thymic development. PLoS One2007;2:e992.

Fernandez-Egea E, et al. Glucose abnormalities in the siblings of people with schizo-phrenia. Schizophr Res 2008;103:110–3.

Fontana L, et al. Visceral fat adipokine secretion is associated with systemic inflamma-tion in obese humans. Diabetes 2007;56:1010–3.

Frölich L, et al. Brain insulin and insulin receptors in aging and sporadic Alzheimer'sdisease. J Neural Transm 1998;105:423–38.

Fucetola R, et al. Age- and dose-dependent glucose-induced increases in memory andattention in schizophrenia. Psychiatry Res 1999;88:1-13.

Ginhoux F, et al. Fate mapping analysis reveals that adult microglia derive from prim-itive macrophages. Science 2010;330:841–5.

Gold PE, et al. Glucose effects on memory: behavioral and pharmacological character-istics. Behav Neural Biol 1986;46:145–55.

Goncalves CA, et al. Adipocytes as an important source of serum S100B and possibleroles of this protein in adipose tissue. Cardiovasc Psychiatry Neurol 2010;7. http://dx.doi.org/10.1155/2010/790431. (Article ID 790431:7 pages).

Gough SC, O'Donovan MC. Clustering of metabolic comorbidity in schizophrenia: agenetic contribution? J Psychopharmacol 2005;19:47–55.

Grill HJ, Kaplan JM. The neuroanatomical axis for control of energy balance. FrontNeuroendocrinol 2002;23:2-40.

Grundy SM, et al. Definition of metabolic syndrome: report of the National Heart, Lung,and Blood Institute/American Heart Association conference on scientific issuesrelated to definition. Arterioscler Thromb Vasc Biol 2004;24:e13–8.

Guest PC, et al. Increased levels of circulating insulin-related peptides in first onset,antipsychotic naive schizophrenia patients. Mol Psychiatry 2010;15:118–9.

Hauner H, et al. Effects of tumour necrosis factor alpha (TNF alpha) on glucose trans-port and lipid metabolism of newly-differentiated human fat cells in cell culture.Diabetologia 1995;38:764–71.

Hegele RA, et al. Lamin A/C mutation in a woman and her two daughters with Dunnigan-type partial lipodystrophy and insulin resistance. Diabetes Care 2000;23:258–9.

Henderson DC, et al. A double-blind, placebo-controlled trial of rosiglitazone forclozapine-induced glucose metabolism impairment in patients with schizophrenia.Acta Psychiatr Scand 2009;119:457–65.

Hirayama I, et al. Insulin receptor-related receptor is expressed in pancreatic beta-cellsand stimulates tyrosine phosphorylation of insulin receptor substrate-1 and -2.Diabetes 1999;48:1237–44.

Holmes E, et al. Metabolic profiling of CSF: evidence that early intervention may impacton disease progression and outcome in schizophrenia. PLoS Med 2006;3:e327.

Juckel G, et al. Microglial activation in a neuroinflammational animal model ofschizophrenia — a pilot study. Schizophr Res 2011;131:96-100.

Juntilla MM, et al. Akt1 and Akt2 are required for alphabeta thymocyte survival anddifferentiation. Proc Natl Acad Sci U S A 2007;104:12105–10.

Kirkpatrick B, et al. Differences in glucose tolerance between deficit and nondeficitschizophrenia. Schizophr Res 2009;107:122–7.

Lai WS, et al. Akt1 deficiency affects neuronal morphology and predisposes to abnormal-ities in prefrontal cortex functioning. Proc Natl Acad Sci U S A 2006;103:16906–11.

Leonard BE, et al. The metabolic syndrome in schizophrenia: is inflammation a contrib-uting cause? J Psychopharmacol 2012;26:33–41.

Lieberman JA, et al. Effectiveness of antipsychotic drugs in patients with chronicschizophrenia. N Engl J Med 2005;353:1209–23.

Lin YT, Liao SC. Hashimoto encephalopathy presenting as schizophrenia-like disorder.Cogn Behav Neurol 2009;22:197–201.

Lindholm E, et al. Linkage analysis of a large Swedish kindred provides further supportfor a susceptibility locus for schizophrenia on chromosome 6p23. Am J Med Genet1999;88:369–77.

Lofgren P, et al. Secretion of tumor necrosis factor-alpha shows a strong relationship toinsulin-stimulated glucose transport in human adipose tissue. Diabetes 2000;49:688–92.

Luo ZC, et al. Mechanisms of developmental programming of the metabolic syndromeand related disorders. World J Diab 2010;1:89–98.

Magistretti PJ, et al. Brain energy metabolism: an integrated cellular perspective. In:Bloom FE, Kupfer DJ, editors. Psychopharmacology: the fourth generation of prog-ress. New York: Raven Press; 1995. p. 657–67.

Maxeiner HG, et al. Flow cytometric analysis of T cell subsets in paired samples of ce-rebrospinal fluid and peripheral blood from patients with neurological and psychi-atric disorders. Brain Behav Immun 2009;23:134–42.

McGowan MK, et al. Chronic intrahypothalamic infusions of insulin or insulin anti-bodies alter body weight and food intake in the rat. Physiol Behav 1992;51:753–66.


http://dx.doi.org/10.1155/2010/790431

http://dx.doi.org/10.1155/2010/790431



McIntyre RS, et al. Mood and psychotic disorders and type 2 diabetes: a metabolic triad.Can J Diab 2005;29:122–32.

Menendez JA, Atrens DM. Insulin and the paraventricular hypothalamus: modulationof energy balance. Brain Res 1991;555:193–201.

Meshkani R, Adeli K. Hepatic insulin resistance, metabolic syndrome and cardiovascu-lar disease. Clin Biochem 2009;42:1331–46.

Meyer U, Feldon J. Neural basis of psychosis-related behaviour in the infection model ofschizophrenia. Behav Brain Res 2009;204:322–34.

Miller BJ, et al. Meta-analysis of cytokine alterations in schizophrenia: clinical statusand antipsychotic effects. Biol Psychiatry 2011;70:663–71.

Mitchell AJ, et al. Prevalence of metabolic syndrome and metabolic abnormalities inschizophrenia and related disorders — a systematic review and meta-analysis.Schizophr Bull in press.

Mitelman SA, et al. Metabolic disconnection between the mediodorsal nucleus of thethalamus and cortical Brodmann's areas of the left hemisphere in schizophrenia.Am J Psychiatry 2005;162:1733–5.

Mozo J, et al. Thermoregulation: what role for UCPs in mammals and birds? Biosci Rep2005;25:227–49.

Myint AM. Kynurenines: from the perspective of major psychiatric disorders. FEBS J2012;279:1375–85.

Narayanan SP, et al. Akt signals through the mammalian target of rapamycin pathwayto regulate CNS myelination. J Neurosci 2009;29:6860–70.

Netto CB, et al. Serum S100B protein is increased in fasting rats. Arch Med Res 2006;37:683–6.

Newcomer JW. Antipsychotic medications: metabolic and cardiovascular risk. J ClinPsychiatry 2007;68(Suppl. 4):8-13.

Newcomer JW, et al. Glucose-induced increase in memory performance in patientswith schizophrenia. Schizophr Bull 1999;25:321–35.

Nikkila H, et al. Abnormal distributions of T-lymphocyte subsets in the cerebrospinalfluid of patients with acute schizophrenia. Schizophr Res 1995;14:215–21.

Nikkilä HV, et al. Accumulation of macrophages in the CSF of schizophrenic patientsduring acute psychotic episodes. Am J Psychiatry 1999;156:1725–9.

Pacheco-Lopez G, et al. Priming ofmetabolic dysfunctions by prenatal immune activa-tion in mice: relevance to schizophrenia. Schizophr Bull in press.

Peters A, et al. The selfish brain: competition for energy resources. Neurosci BiobehavRev 2004;28:143–80.

Peters A, et al. Causes of obesity: looking beyond the hypothalamus. Prog Neurobiol2007;81:61–88.

Piontkewitz Y, et al. Effects of risperidone treatment in adolescence on hippocampalneurogenesis, parvalbumin expression, and vascularization following prenatalimmune activation in rats. Brain Behav Immun 2012;26:353–63.

Pont-Lezica L, et al. Physiological roles of microglia during development. J Neurochem2011;119:901–8.

Porte Jr D, et al. Insulin signaling in the central nervous system: a critical role inmetabolichomeostasis and disease from C. elegans to humans. Diabetes 2005;54:1264–76.

Potvin S, et al. Inflammatory cytokine alterations in schizophrenia: a systematic quan-titative review. Biol Psychiatry 2008;63:801–8.

Radewicz K, et al. Increase in HLA-DR immunoreactive microglia in frontal and tempo-ral cortex of chronic schizophrenics. J Neuropathol Exp Neurol 2000;59:137–50.

Ragland JD, et al. Functional magnetic resonance imaging of internal source monitoringin schizophrenia: recognition with and without recollection. Schizophr Res2006;87:160–71.

Rolfe DF, Brown GC. Cellular energy utilization and molecular origin of standard meta-bolic rate in mammals. Physiol Rev 1997;77:731–58.

Rothermundt M, et al. Increased S100B blood levels in unmedicated and treated schizo-phrenic patients are correlated with negative symptomatology. Mol Psychiatry2001;6:445–9.

Rothermundt M, et al. Glial cell dysfunction in schizophrenia indicated by increasedS100B in the CSF. Mol Psychiatry 2004;9:897–9.

Ryan MC, et al. Impaired fasting glucose tolerance in first-episode, drug-naive patientswith schizophrenia. Am J Psychiatry 2003;160:284–9.

Saddichha S, et al. Diabetes and schizophrenia— effect of disease or drug? Results froma randomized, double-blind, controlled prospective study in first-episode schizo-phrenia. Acta Psychiatr Scand 2008;117:342–7.

Sakaguchi S, et al. Regulatory T cells and immune tolerance. Cell 2008;133:775–87.Salmi P, et al. Antagonism by SCH 23390 of clozapine-induced hypothermia in the rat.

Eur J Pharmacol 1994;253:67–73.Scheen AJ, De Hert MA. Abnormal glucose metabolism in patients treated with antipsy-

chotics. Diabetes Metab 2007;33:169–75.Schlegelmilch T, et al. Microglia in the developing brain: from immunity to behaviour.

Curr Opin Neurobiol 2011;21:5-10.Schroeter ML, Steiner J. Elevated serum levels of the glial marker protein S100B are not

specific for schizophrenia or mood disorders. Mol Psychiatry 2009;14:235–7.Schwarcz R, et al. Kynurenines in the mammalian brain: when physiology meets

pathology. Nat Rev Neurosci 2012;13:465–77.Shim WS, et al. The association of total and differential white blood cell count with

metabolic syndrome in type 2 diabetic patients. Diabetes Res Clin Pract 2006;73:284–91.

Sicras-Mainar A, et al. C-reactive protein as a marker of cardiovascular disease in pa-tients with a schizophrenia spectrum disorder treated in routine medical practice.Eur Psychiatry in press.

Spelman LM, et al. Impaired glucose tolerance in first-episode drug-naive patients withschizophrenia. Diabet Med 2007;24:481–5.


Stahl SM, et al. Which comes first: atypical antipsychotic treatment or cardiometabolicrisk? Acta Psychiatr Scand 2009;119:171–9.

Steen E, et al. Impaired insulin and insulin-like growth factor expression and signalingmechanisms in Alzheimer's disease — is this type 3 diabetes? J Alzheimers Dis2005;7:63–80.

Stefanidis A, et al. The role of thermogenesis in antipsychotic drug-induced weightgain. Obesity (Silver Spring) 2009;17:16–24.

Stefansson H, et al. Common variants conferring risk of schizophrenia. Nature2009;460:744–7.

Steiner J, et al. Distribution of HLA-DR-positive microglia in schizophrenia reflects im-paired cerebral lateralization. Acta Neuropathol 2006;112:305–16.

Steiner J, et al. Immunological aspects in the neurobiology of suicide: elevatedmicroglial density in schizophrenia and depression is associated with suicide.J Psychiatr Res 2008;42:151–7.

Steiner J, et al. A new pathophysiological aspect of S100B in schizophrenia: potentialregulation of S100B by its scavenger soluble RAGE. Biol Psychiatry 2009;65:1107–10.

Steiner J, et al. S100B serum levels are closely correlated with body mass index: an im-portant caveat in neuropsychiatric research. Psychoneuroendocrinology 2010a;35:321–4.

Steiner J, et al. Elevated S100B levels in schizophrenia are associated with insulin resis-tance. Mol Psychiatry 2010b;15:3–4.

Steiner J, et al. Human CD8(+) T cells and NK cells express and secrete S100B uponstimulation. Brain Behav Immun 2011;25:1233–41.

Steiner J, et al. Bridging the gap between the immune and glutamate hypotheses ofschizophrenia and major depression: potential role of glial NMDA receptor modu-lators and impaired blood-brain barrier integrity. World J Biol Psychiatry2012a;13:482–92.

Steiner J, et al. Increased S100B+ NK cell counts in acutely ill schizophrenia patientsare correlated with the free cortisol index, but not with S100B serum levels.Brain Behav Immun 2012b;26:564–7.

Steiner J, et al. Increased prevalence of diverse NMDA receptor antibodies in patientswith an initial diagnosis of schizophrenia: specific relevance of IgG NR1a anti-bodies for distinction from NMDA receptor encephalitis. Arch Gen Psychiatry. inpress.

Steinman RM, et al. Dendritic cell function in vivo during the steady state: a role in pe-ripheral tolerance. Ann N Y Acad Sci 2003;987:15–25.

Sunram-Lea SI, et al. The effect of retrograde and anterograde glucose administrationon memory performance in healthy young adults. Behav Brain Res 2002;134:505–16.

Sutton LP, Rushlow WJ. The effects of neuropsychiatric drugs on glycogen synthasekinase-3 signaling. Neuroscience 2011;199:116–24.

Tanaka S, et al. Autoantibodies against four kinds of neurotransmitter receptors inpsychiatric disorders. J Neuroimmunol 2003;141:155–64.

Torrey EF, et al. Cytomegalovirus and schizophrenia. CNS Drugs 2006;20:879–85.Tremblay ME, Majewska AK. A role for microglia in synaptic plasticity? Commun Integr

Biol 2011;4:220–2.Tsutsui K, et al. Anti-NMDA-receptor antibody detected in encephalitis, schizophrenia,

and narcolepsy with psychotic features. BMC Psychiatry 2012;12:37.Valassi E, et al. Neuroendocrine control of food intake. Nutr Metab Cardiovasc Dis

2008;18:158–68.van Berckel BN, et al. Microglia activation in recent-onset schizophrenia: a quantitative

(R)-[11C]PK11195 positron emission tomography study. Biol Psychiatry 2008;64:820–2.

van Beveren NJ, et al. Marked reduction of AKT1 expression and deregulation ofAKT1-associated pathways in peripheral blood mononuclear cells of schizophreniapatients. PLoS One 2012;7:e32618.

Van den Bossche J, et al. Regulation and function of the E-cadherin/catenin complex incells of the monocyte–macrophage lineage and DCs. Blood 2012;119:1623–33.

van der Pal RH, et al. Effects of insulin and insulin-like growth factor (IGF-I) onoligodendrocyte-enriched glial cultures. J Neurosci Res 1988;19:483–90.

van Os J, Kapur S. Schizophrenia. Lancet 2009;374:635–45.Wang H, et al. Coexistences of insulin signaling-related proteins and choline

acetyltransferase in neurons. Brain Res 2009;1249:237–43.Weinberger DR. Implications of normal brain development for the pathogenesis of

schizophrenia. Arch Gen Psychiatry 1987;44:660–9.Weiss AP, et al. Impaired hippocampal function during the detection of novel words in

schizophrenia. Biol Psychiatry 2004;55:668–75.Winter C, et al. Prenatal immune activation leads to multiple changes in basal neuro-

transmitter levels in the adult brain: implications for brain disorders ofneurodevelopmental origin such as schizophrenia. Int J Neuropsychopharmacol2009;12:513–24.

Xu H, et al. Chronic inflammation in fat plays a crucial role in the development ofobesity-related insulin resistance. J Clin Invest 2003;112:1821–30.

Yolken RH, Torrey EF. Viruses, schizophrenia, and bipolar disorder. Clin Microbiol Rev1995;8:131–45.

Zhao Z, et al. Insulin receptor deficits in schizophrenia and in cellular and animalmodels of insulin receptor dysfunction. Schizophr Res 2006;84:1-14.

Zierhut K, et al. The role of hippocampus dysfunction in deficient memory encodingand positive symptoms in schizophrenia. Psychiatry Res Neuroimaging 2010;183:187–94.



Documents

Immune system and glucose metabolism interaction in schizophrenia: A chicken–egg dilemma