Topic: Neurovascular and neuroinflammation mechanisms associated with

Neuroinflammation in bipolar disorders

Georgios D. Kotzalidis, Elisa Ambrosi, Alessio Simonetti, Ilaria Cuomo, Antonio Del Casale, Matteo Caloro, Valeria Savoja, Chiara Rapinesi Department of Psychiatric Rehabilitation, P. Alberto Mileno Onlus Foundation, San Francesco Institute, 66054 Vasto, CH, Italy.

ABSTRACT

Recent literature based on peripheral findings speculated that neuroinflammation, with its connection to microglial activation, is linked to bipolar disorder. The endorsement of the neuroinflammatory hypotheses of bipolar disorder requires the demonstration of causality, which requires longitudinal studies. We aimed to review the evidence for neuroinflammation as a pathogenic mechanism of the bipolar disorder. We carried out a hyper inclusive PubMed search using all appropriate neuroinflammation‑related terms and crossed them with bipolar disorder‑related terms. The search produced 310 articles and the number rose to 350 after adding articles from other search engines and reference lists. Twenty papers were included that appropriately tackled the issue of the presence (but not of its pathophysiological role) of neuroinflammation in bipolar disorder. Of these, 15 were postmortem and 5 were carried out in living humans. Most articles were consistent with the presence of neuroinflammation in bipolar disorder, but factors such as treatment may mask it. All studies were cross‑sectional, preventing causality to be inferred. Thus, no inference can be currently made about the role of neuroinflammation in bipolar disorder, but a link is likely. The issue remains little investigated, despite an excess of reviews on this topic.

Key words: Bipolar disorder, , neuroinflammation, psychoneuroimmunology,

INTRODUCTION the cooperation among the nervous, endocrine, and immune systems,[8] and stress disorders are Among the former mood or affective disorders, bipolar believed to be underpinned by derangements in this disorder is the one that is most intriguing. Its heterogeneity integration.[9] is well‑recognized. While traditionally subdivided into bipolar I and bipolar II according to whether there STRESS AND INFLAMMATION was a history of ,[1] some scholars support the existence of more than 10 subtypes.[2] It is supposed to Inflammation is a process described since antiquity, be pathophysiologically/neurobiologically continuous meaning burning in both Greek (φλόγωσις) and with other psychiatric disorders such as schizophrenia Latin (inflammation), and characterized by according to Griesinger’s model of Einheitspsychose,[3] swelling (tumor), redness (rubor), heating (calor), and with recurrent major depression.[4‑6] pain (dolor), and impaired function (functiolæsa). It is a general reaction to pathogens in a tissue involving Mood disorders along with disorders are an innate response of cells residing within that tissue. considered as “stress disorders.”[7] The organism In the acute phase, the process entails the extravasation responds to stress in an integrated manner, involving of immune cells, permeability changes in blood vessels, and the production of chemical mediators, including Corresponding Author: Dr. Georgios D. Kotzalidis, acute phase proteins, vasoactive amines, eicosanoids, Department of , Mental , and Sensory bradykinin and other tachykinins, and chemical Organs, Sapienza University, School of Medicine and attractors. The nonspecific response usually leads , Sant’Andrea Hospital, UOC Psichiatria, to resolution and repair, with restitution to integrity. Via di Grottarossa 1035‑1039, 00189 Rome, Italy. E‑mail: [email protected] This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License, which allows Access this article online others to remix, tweak, and build upon the work non-commercially, as long as the Quick Response Code: author is credited and the new creations are licensed under the identical terms. Website: For reprints contact: [email protected] www.nnjournal.net

Cite this article as: Kotzalidis GD, Ambrosi E, Simonetti A, Cuomo I, DOI: Del Casale A, Caloro M, et al. Neuroinflammation in bipolar disorders. 10.4103/2347-8659.167309 Neuroimmunol Neuroinflammation 2015;2:252-62. Received: 14-06-2015; Accepted: 31-07-2015

252 © 2015 Neuroimmunology and Neuroinflammation | Published by Hongkong Partner Publishing Co. Limited The inability to restore the previous healthy state The whole process appears, however, to be mediated may ensue in chronic inflammation, characterized through stress,[16] which is a very general term, but it by shifts from the main participating cells toward is essential to understand neuroinflammation as part mononuclear (monocytes, , , of a whole in the pathophysiology of various disorders, and plasma) cells and fibroblasts and from the and disease in general. main participating molecules toward interferon‑γ, interleukins (ILs), growth factors, nitric oxide, and There has been evidence that neuroinflammation hydrolytic enzymes.[10] is reflected in changes in peripheral immunity,[17] but the reverse may not be true, hence, evidence The stress concept was developed from the work of of peripheral immunological alterations cannot be Selye,[11,12] who viewed the body in Cannon’s frame.[13] taken to indicate the presence of neuroinflammation. Stress, like inflammation, was proposed to be the There is no consensus as to whether central and body’s response to “diverse nocuous agents,”[11] a peripheral immunity are specular, and a recent study general syndrome characterized by an that adequately addressed the issue showed that integrated neuroendocrine and immune response immune markers were found to be independent from tending to restore . Selye[12] showed the peripheral activity of the immune system.[18] that the response also involves the immune system, which constitutes another parallel with the Summarizing, stress may lead to the establishment inflammatory response. Currently, the stress response of a chronic inflammatory reaction, which may is considered to be evolutionary and likely to amplify occur in the brain in parallel to the periphery, thus an organism’s resistance to environmental stressors. setting brain function in a sickness mode that may Like inflammation, the inability of an organism to substitute default activity and perpetuate the disorder. adequately address stress may lead to a stress disorder, This does not explain the oscillatory mood activity. which in is represented by posttraumatic The presence of neuroinflammation in the brain in stress disorder, anxiety disorders, and mood disorders patients with bipolar disorder can help to answer such as depression. whether neuroinflammation is a general way by which manic‑depressive symptoms are produced. In recent years, there has been an increasing It is not a sufficient evidence to demonstrate some recognition of altered immunological parameters in of these symptoms in people with autoimmune many psychiatric disorders, including depression, neuroinflammation or other disorders showing both bipolar disorder, autism spectrum disorders, and neuroinflammation and cognitive or mood alterations schizophrenia.[14,15] The rationale is that chronic proper of bipolar disorder. Instead, convincing evidence inflammation, by releasing , may set requires the demonstration of neuroinflammation brain function into a “sickness mode,” thus causing in a population of patients with bipolar disorders psychiatric disorders. However, how this is carried out who have no other comorbidity. For this reason, is not explained, so it remains an interesting paradigm we consider studies that point to the existence of with no demonstration so far. neuroinflammation in bipolar disorder only those studies investigating microglial activation or showing WHAT DO WE INTEND BY increased presence of neuroinflammatory markers “NEUROINFLAMMATION?” in the human brain, in people with bipolar disorder compared to healthy or nonpsychiatric controls. Neuroinflammation is defined as inflammation of These may be postmortem studies or studies in living the . It consists of increased humans that involve the cerebrospinal fluid (CSF) or glial activation, pro‑inflammatory content, brain imaging. blood‑brain‑barrier permeability, and leukocyte extravasation. The process is believed to be driven Having this in mind, we aimed to review the evidence by IL‑1 beta (IL‑1β), a cytokine that has been found for neuroinflammation in the pathogenesis of bipolar to be increased in neurodegenerative disorders such disorder. as Alzheimer’s disease, Parkinson’s disease, and . IL‑1β stimulates the IL‑1 receptor/ OUR SEARCH STRATEGY TO INVESTIGATE IL‑1 accessory protein complex to increase NEUROINFLAMMATION IN BIPOLAR DISORDER nuclear factor kappa B‑dependent transcription of pro‑inflammatory cytokines, such as tumor necrosis We performed a careful PubMed search using the factor (TNF)‑α, IL‑6, and interferons, as well as the following strategy: (neuroinflammation* or glia* or ‑recruiting CXCL1 and CXCL2. * or [(CXCL1 or CXCL2 or IL‑1b* or IL‑6 or

Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 253 interleukin* or interferon* or quantitative and qualitative syntheses are quite weak in or TNF* or NKκB) and (brain or cerebrospinal fluid Prisma, and their definitions fuzzy and because we were or CSF)] and (“bipolar disorder” or mania or manic subsequently faced with an extreme heterogeneity of or “recurrent depression”). There were no restrictions study designs and results that cannot be meta‑analyzed. as to publication date or language. To be included, the paper had to be an original article and carried out WHAT DID THE SEARCH PRODUCE? in humans. We did not restrict our search to human studies using the related PubMed function, because The PubMed search yielded 316 papers as of 11th of July in our experience this practice does not exclude 2015. The search of the reference lists of these papers complete animal studies but is likely to conceal some produced further 40 potentially interesting papers. Of relevant human investigations instead. Hence, animal the total output of 356 papers, 100 were completely studies were subsequently excluded on the basis of unfocused and were captured for the presence of evidence. Furthermore, we excluded reviews and spandrels in the search strategy (however, we chose meta‑analyses, opinion/speculative papers, animal not to modify the strategy by adopting a more specific studies, case reports, as well as papers conducted approach, since this would result in a potential loss of without respect to the 1964 Declaration of Helsinki otherwise eligible material), 98 were excluded because principles of human rights, reviews and meta‑analyses, they were reviews or meta‑analyses, 16 were opinion opinion/speculative papers, animal studies, as well as articles/editorials or speculative with no experimental case reports. However, their references, especially those data, 55 were animal studies, 5 were case studies or case of reviews/meta‑analyses were searched for possible series, 7 did not include patients with bipolar disorder further includible papers. This review followed the separately or did not investigate neuroinflammation bureaucratic Prisma statement[19,20] when appropriate. at all, 50 were investigations of peripheral markers, We inverted the order recommended for screening, 1 was a duplicate, and 4 were carried out in vitro on that is, first exclude duplicates and then the screen. human glial cells. The flow diagram in Figure 1 shows We first classified studies according to their type, the search strategies and papers selected for review then excluded animal studies, reviews and studies according to a modified Prisma algorithm. A total final of peripheral immunity, and included only human number of 20 papers were found to be eligible and studies with data, including clinical and postmortem, were analyzed. The search output spanned from 1982 but not in vitro experiments on cell lines. We also to 2015; recalling that the first occurrence in PubMed excluded and classified as unfocused those postmortem of the term neuroinflammatory was in 1983 that of studies investigating glia in the brain without telling neuroinflammation was in 1995, and that microglial microglia from other types of glial cells. We excluded activation and related terms appeared during 1973- all nonpeer reviewed literature, as it could constitute 1975, we may presume that there was a dearth of a source of bias. We considered as duplicates only focused papers in the 1st year, that is, prior to 2000. studies that reported the same results in different However, one of the includible papers was dated published reports. When different studies by the same 1997. This was the case (19 papers up to 1999 and research group progressively report on increasingly 337 from 2000 on) and the trend has been increasing larger samples and when the past used sample is used through years, witnessing an increasing interest of the and accrued, we adopt the strategy to disregard the first scientific community in the neuroinflammation issue appearing papers, including only the last study with in psychiatric disorders in general, and in bipolar the larger sample, provided its quality of evidence is disorder in particular. Of interest, two of the included high. Papers reporting on the same samples, but on papers were not identified by the PubMed search despite different measures, were considered as different studies the strategy was appropriate for singling them out. The and included if appropriate. Contrary to the distinction included papers are shown in Tables 1 and 2. They made by the Prisma statement between records and fell into two broad categories, postmortem (n = 15) full‑text articles, we considered all papers emerging and in vivo (n = 5), the latter comprising four studies from our research as papers to obtain in full text and of neuroinflammatory mediators in the CSF and one carefully searched for any data that conformed to our was an ingenuous positron emission tomography study aims. Our experience is that you can never say if you of the binding of a neuroinflammatory marker in the only read the abstract, especially for not so recently brain. The only study we excluded as a duplicate published papers. was Rolstad et al.[21] which focused on the correlation between CSF neuroinflammatory markers and cognitive Contrariwise to what Prisma dictates, we did not label performance and admittedly reported data previously our review as “systematic” and did not attempt to reported in Jakobsson et al.,[18] despite differences in carry out a meta‑analysis, both because the concepts of sample size.

254 Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 Records identified through PubMed search: neuroinflammat* or glia* or microglia* or [(CXCL1 or CXCL2 or IL-1b* or IL-6 or interleukin* or interferon* or tumor necrosis Additional records identified through Reference factor or TNF* or NKκB) and (brain or lists, other databases and alternative cerebrospinal fluid or CSF)], and “bipolar search strategies (n = 40) disorder” or mania or manic or “recurrent depression” in July 11, 2015 (n = 316)

Reviews, focused, with reference lists Identification potentially providing further records (n = 42) Total records retrieved and evaluated for inclusion (screened n = 356) Reviews, unfocused (n = 56) Reviews/meta-analyses (n = 98) Case reports/series (n = 5)

Opinion papers, editorials,

Screening speculative articles (n = 16)

Ethics not respected (n = 0) Animal studies (n = 55)

Not relevant, unfocused (n = 100)

Human original investigations (n = 82) Excluded due to inadequate methodology for investigating neuroinflammation in

Eligibility bipolar disorder (n = 61)

Studies included in qualitative Peripheral markers (n = 50) In vitro experiments synthesis (n = 21) (n = 4)

No separate analysis for bipolar disorder or no bipolar disorder patients included or

Included Records after removing studies neuroinflammation not investigated (n = 7) using the same or overlapping samples (n = 20)* Measurements of neuroinflammatory components in the Post-mortem (n = 15) In living humans (n = 5) cerebrospinal fluid (n = 4)

Neuroimaging of neuroinflammatory markers in the brain (n = 1)

Figure 1: Results for search and inclusion and design typology (*Despite the existence of overlapping samples, most studies except one, were not considered duplicates since they reported on different datasets)

BASED ON THE SEARCH, IS THERE ANY to its anti‑neuroinflammatory, anti‑oxidative stress, or EVIDENCE FOR NEUROINFLAMMATION IN anti‑apoptotic effect or to its mitochondrial dysfunction BIPOLAR DISORDER? countering or glutamate/ balancing actions,[44] unless accompanied by evidence of neuroinflammatory This review attempted to answer the question of whether markers moving in the desired direction (and the neuroinflammation plays a role in the pathophysiology demonstration of their alteration at baseline). of bipolar disorder. Furthermore, the evidence of abnormal peripheral inflammatory reactivity cannot be taken as evidence We may not speak of conclusive evidence of of neuroinflammation. neuroinflammatory mechanisms in bipolar disorder when the obtained evidence is too indirect. For Our search yielded a high number of interesting example, when a molecule like N‑acetylcysteine is articles, but few of them suited the purpose of this found to have some therapeutic activity in bipolar review. This was due to the fact that our search strategy disorder, we may not specify whether this is related was over‑inclusive to avoid missing any suitable article.

Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 255 Table 1: Postmortem studies included in this review investigating microglia or neuroinflammation in bipolar disorder Study Origin Source Population Measure(s) Results Observations Ishizuka Japan Kumamoto 11 neuro‑ Northern blot‑ Increased Evidence for et al.[22] (Kumamoto, University, psychiatric patients, immunohistochemistry, expression of MIP increased Fukuoka, Kyushu‑ Department 1 bipolar; 2 surgical hybridization 1a/LD78 in the expression of Maidashi) of Anatomy (brain sampled to assess expression bipolar patient in inflammatory (postmortem during neurosurgery and distribution of MIP white matter glial marker in glia autopsies); for brain tumor) 1a/LD78 in brain cells in the brain of Neurosurgery just one patient School of with bipolar Medicine, disorder; four Fukuoka (brain patients with sampling schizophrenia during brain showed neuronal surgery) as well as glial distribution abnormalities Hamidi USA (Washington Harvard 9 bipolar versus 8 Microglial density No differences Patients with et al.[23] University, St. Brain Tissue major depression (cells/mm3) in the between bipolar bipolar disorder Louis, Mo; NIMH, Resource versus 10 disorder and had been exposed Bethesda, Md) Center nonpsychiatric controls to valproate or controls lithium more often than patients with major depression; higher suicide rates in patients versus controls Frank Germany Stanley 3 bipolar, 4 Microarray‑based HML‑2 family HERV transcription et al.[24] (Mannheim, Foundation schizophrenia, analysis of HERV (HERV‑K10) in brain weakly Heidelberg, Brain 4 nonpsychiatric transcriptional activity significantly correlates with Oberschleissheim) Collection, controls for in the dorsolateral overrepresented schizophrenia Bethesda comparing retroviral . in bipolar and related RNA in various brain DNA chip investigated disorder and disorders, but may areas; 35 bipolar, through Env‑specific schizophrenia, be affected by 35 schizophrenia, QRT‑PCR. An animal compared to individual genetic 35 nonpsychiatric retrovirus‑specific control . background, controls for all microarray was HERV E4‑1 brain‑infiltrating analyses in BA performed to test transcription immune cells, 46 (dorsolateral the hypothesis of overrepresented or medical prefrontal cortex) zoonosis in bipolar disorder treatment; the Env expression higher incidence of HERV‑W, of HERV‑K10 HERV‑FRD, and transcripts in HML unaffected schizophrenia and regardless of the bipolar disorder clinical picture. is a probable No transcripts consequence of of any animal high brain immune retroviruses reactivity detected with pet chip in all 105 human brains Dean Australia Autopsied 8 bipolar versus 20 Prefrontal S100β Decreased BA 9 Postmortem et al.[25] (University of cases at schizophrenia versus levels (indirect and increased BA interval arbitrarily Melbourne, Victorian 20 nonpsychiatric evidence; 40 S100β levels defined in not Parkville, VIC) Institute of controls S100β and microglial in bipolar disorder witnessed deaths; Forensic IL‑1β induce one I versus other no suicide Medicine another) groups Foster England Stanley 15 bipolar, 15 Double Higher levels in Evidence of et al.[26] (King’s‑Maudsley), Foundation schizophrenia, immune‑fluorescence schizophrenia, neuroinflammation Canada Brain Collection 15 major for the lowest in controls, in bipolar (University of depression, and neuroinflammation intermediate in disorder, but not British Columbia, 15 nonpsychiatric marker calprotectin major depression, so strong as in Vancouver), controls and microglia in and bipolar schizophrenia; Norway (Ullevål BA 9 (dorsolateral disorder in post‑mortem University, Oslo) prefrontal cortex) dorsolateral interval, age, and and USA (UCSD, prefrontal sex distribution San Diego, CA) microglia not reported, but said to have not influenced results

Contd..

256 Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 Table 1: Contd... Study Origin Source Population Measure(s) Results Observations Weis USA (Bethesda, Stanley 15 bipolar, 15 Immunohistochemistry In the bipolar This is a very et al.[27] Md), Austria (Linz, Neuropathology schizophrenia, to detect PrPc‑positive group, neuroleptics indirect measure of Oberösterreich) Consortium 15 major cells in the cingulate decreased the neuroinflammation, and Switzerland Collection depression, and gyrus numerical density pointing to the (University of 15 nonpsychiatric of PrPc‑positive possibility that Zürich) controls and drug treatment has increased that of something to do PrPc‑positive white with it. Causes of matter microglia death not provided Rao USA (NIH, Harvard 10 bipolar, 10 Western blot, total Higher protein Excitotoxic et al.[28] Bethesda, Md and Brain Tissue nonpsychiatric RNA isolation‑real and mRNA levels markers were also North Carolina) Resource controls time reverse of IL‑1β, IL‑1 up‑regulated; the Center (McLean transcriptase PCR and receptor, MyD88, authors speculated Hospital, immunohistochemistry NF‑κB subunits, that glutamatergic Belmont, MA, of frontal cortex and astroglial and derangement (as USA) membrane, nuclear, microglial markers shown by and cytoplasmic GFAP, iNOS, c‑fos decreased protein extracts and CD11b in and mRNA for the frontal cortex NMDA receptors of patients with NR‑1 and bipolar disorder NR‑3A) could be responsible for neuroinflammation Kim USA (NIH, Harvard 10 bipolar, 10 Western blot, total Increased protein Deranged et al.[29] Bethesda, Md) Brain Tissue nonpsychiatric RNA isolation‑real and mRNA of neuroinflammatory Resource controls (same time reverse AA‑selective response, which Center (McLean sample as Rao transcriptase PCR and cPLA2 IVA, the authors relate Hospital, et al.[28]) immunohistochemistry secretory to excitotoxicity; Belmont, MA, of frontal cortex sPLA2‑IIA, COX‑2 tendency to repeat USA) membrane, nuclear, and mPGES in the conclusions of and cytoplasmic bipolar disorder; the preceding paper extracts decreased COX‑1 (Rao et al.[28]). and cPGES The arachidonic compared to cascade is not control brains only involved in neuroinflammation but in other processes as well Steiner Germany Magdeburg 12 suicide victims Immunohistochemistry Increased Drug treatment et al.[30] (University of brain bank (D) (5 bipolar, 7 major for the NMDA agonist quinolinic may have Magdeburg) depression) versus quinolinic acid in acid‑staining cells affected quinolinic 10 nonpsychiatric the microglia of the in the anterior acid content of controls anterior cingulate midcingulatecortex microglia (thus gyrus and the masking a subgenual, but not possible difference in the pregenual from controls in cortex, in the bipolar disorder); major depression, admittedly, but not bipolar microglial disorder, suicide immunoreactivity victims may not be attributed to increased synthesis or decreased metabolic breakdown; relevance to neuroinflammation only indirect Rao USA (NIH, Harvard 10 bipolar, 10 Genomic DNA Increased Data compatible et al.[31] Bethesda, Md) Brain Tissue nonpsychiatric isolation, gene‑specific mRNA and with altered frontal Resource controls (same and global DNA protein levels of cortex epigenetic Center (McLean sample as Rao methylation; total RNA neuroinflammatory regulation Hospital, et al.[28]), matched; 10 isolation‑real time markers (IL‑1β related to Belmont, MA, Alzheimer’s disease, reverse transcriptase and TNF‑α) neuroinflammation USA) 10 nonpsychiatric PCR for BDNF, NF‑κB and of markers in bipolar disorder controls, matched p50 and NF‑κB p65, of astrocytic and global histone and microglial acetylation and activation in phosphorylation, all both bipolar and from BA 9 (dorsolateral Alzheimer prefrontal cortex) Contd..

Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 257 Table 1: Contd... Study Origin Source Population Measure(s) Results Observations Dean Australia Victorian Brain 10 bipolar, 10 Western blotting and Transmembrane, Significantly more et al.[32] (University of Bank Network, major depression, PCR for IL‑1β and but not soluble patients with Melbourne, Mental Health 19 schizophrenia, TNF‑related measures TNF‑α transcript bipolar disorder Parkville, VIC) Research 30 age‑ and in the prefrontal cortex increases are had committed Institute, gender‑matched and in the anterior found in the suicide compared Parkville, nonpsychiatric cingulate cingulate (BA 24), to the other groups Australia controls but not prefrontal cortex (BA 46); other groups did not show such increases; soluble TNF‑α and IL‑1β levels did not differ from controls in any mental group; decreased TNFR2 levels in bipolar disorder in BA 46; results not consistent with neuroinflammation Gos Germany Magdeburg 8 bipolar versus 9 Hippocampal S100β Numerical No suicide among et al.[33] (Magdeburg and brain bank (D) major depression levels (indirect density of S100β bipolar disorder Leipzig) and versus 13 evidence; astrocyte immuno‑positive patients, but 7 Poland (Gdańsk) nonpsychiatric S100β and microglial suicides among controls interleukin‑1 β induce bilaterally major depression one another)* decreased in patients; no CA1 pyramidal evidence of layer in both neuroinflammation major depression and bipolar brains compared to controls; decreased density of S100β immuno‑positive oligodendrocytes in left alveus only in bipolar disorder Hercher Canada Stanley 20 bipolar versus 20 Prefrontal Not different Bipolar patients et al.[34] (University of Medical schizophrenia versus microglial clustering from controls had committed British Columbia, Research 20 nonpsychiatric coefficient and and patients with suicide more Vancouver, BC) Institute’s brain controls prefrontal microglial schizophrenia often than collection density (cells/mm2) controls and had more often a heavy drug abuse history Fillman Australia (Sydney, Stanley 34 bipolar, 35 Sample Trend for bipolar 15 suicides among et al.[35] NSW, Australia) Medical schizophrenia, 35 dichotomized to patients to belong bipolar, 7 suicides and USA Research nonpsychiatriccontrols high‑ (n = 32) to the high among patients (University of Institute (Array and low‑ (n = 68) inflammation/ with schizophrenia; Pennsylvania, Cohort) inflammation/stress stress group results partly Philadelphia, clusters according (n = 11), while support Pa and Stanley to differences in patients with neuroinflammation Medical Research inflammatory and schizophrenia in bipolar disorder Institute, stress‑related gene were significantly Rockville, Md) transcripts of more likely than RNA extracted from controls the frontal (n = 15 vs. 6) cortex and assessed through microarray analysis and PCR

Contd..

258 Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 Table 1: Contd... Study Origin Source Population Measure(s) Results Observations de Brazil (São Paulo, Stanley 29 bipolar, 29 RNA extracted Immune Neuroinflammatory Baumont SP and Porto Neuropathology schizophrenia, 30 from frontal cortex response‑and microglia activation et al.[36] Allegre, Rio Consortium nonpsychiatriccontrols to assess gene stress‑related is compatible Grande do Sul) expression profile; genes were with these data; and Portugal cloned DNA for differentially the influence of (Braga and microarray analysis expressed medication has not Guimarães, between the been ruled out Braga) control and the patient group; bipolar patients differed from those with schizophrenia in that the former showed an up‑regulation of a set of 28 genes in bipolar disorder with respect to schizophrenia *S100β induces the expression of IL‑1β in cultured ratmicroglia;[37,38] astrocyte synthesis of S100β in AD may be triggered by microglial‑derived IL‑1.[39] BA: brodmann’s area; TNFR2: tumor necrosis factor receptor, type 2; IL‑1β: interleukin‑1 beta; MIP 1a/LD78: inflammatory protein‑1 alpha/LD78; PrPc: cellular prion protein; QRT‑PCR: quantitative real‑time polymerase chain reaction; MyD88: myeloid differentiation factor 88; NF‑κB: nuclear factor kappa B; GFAP: glial fibrillary acidic protein; iNOS: inducible nitric oxide synthase; NMDA: N‑methyl‑D‑aspartic acid; HERV: human endogenous retrovirus; HERV‑W: human endogenous retrovirus‑W; AA: arachidonic acid; cPLA2: cytosolic phospholipase A2; sPLA2‑IIA: secretory phospholipase A2‑IIA; COX: ; mPGES: membrane prostaglandin E synthase; cPGES: cytosolic prostaglandin E synthase; TNF‑α: tumor necrosis factor a

It proved to be more tiresome to download all papers, individual that eventually ensued in bipolar behavior but this allowed us to identify two papers that would and disorder, cannot be probed. In fact, to demonstrate otherwise have gone undetected. Surprisingly, the causality we need longitudinal study designs, and in majority of suitable articles regarded postmortem the case of neuroinflammation and bipolar disorder all studies. These studies (n = 15) [Table 1] favored the relevant articles were based on cross‑sectional articles. idea of the existence of neuroinflammation in bipolar While this was mandatory for postmortem studies, it disorder in their majority. Two provided indirect was not for studies investigating living humans. Future evidence for microglial activation, while four were not studies should be able to tackle the causality question consistent with the presence of neuroinflammation in by studying the same patients across the various phases bipolar disorder. However, in one of these,[23] it is possible of their illness. However, the fact that the supposed that treatment could have set‑off neuroinflammation. In neuroinflammation was present to some extent also fact, patients were receiving drugs such as lithium and when bipolar disorder was in its euthymic phases valproate, which both interfere with the arachidonic strongly argues against the lack of involvement of the acid cascade,[45] one of the cross‑roads of excitotoxicity brain immune system in bipolar disorder. and neuroinflammation.[46] The evidence stemming from in vivo studies (n = 5) is consistent with the The studies included in our review were methodologically presence of neuroinflammation in bipolar disorder, different, and their sample sizes were much variable. also in consideration of the fact that most patients were Investigations of neuroinflammatory markers in the CSF sampled/tested when euthymic. The demonstration were all but one carried out in Sweden and conducted of peripheral benzodiazepine receptor alterations in by the same Karolinska‑Gothenburg group. One of bipolar I disorder patients may constitute a definitive these studies had to be excluded due to sample and demonstration,[43] but it should also be considered data duplication,[21] but all these studies were quite that such alterations in the brain of people with consistent in their conclusions, supporting the existence bipolar disorder, which are consistent with microglial of neuroinflammation in bipolar disorder. This might activation, may not be specifically related to the have introduced a site bias. The markers tested each pathogenesis of bipolar disorder or any diagnosis, but time in postmortem studies differed, even when the rather to disease activity.[47] same research groups were involved. Hence, we had no population overlap, with the same group first reporting on Causality effects, that is, whether it is bipolar disorder some people and then on a bigger sample that comprised that once established, triggers neuroinflammation, the formerly reported cases. We found no duplicates even either through the adoption of a reckless lifestyle when data referred to the same population in postmortem that is likely to promote a metabolic syndrome that studies. There was a tendency in Bethesda‑based groups facilitates the onset of neuroinflammation, or rather it to support neuroinflammation while the German groups is neuroinflammation that has always existed in a given were more skeptical about it [Table 1].

Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 259 Table 2: Studies in living humans included in this review investigating microglia or neuroinflammation in bipolar disorder Study Origin Population Design Results Observations Soderlund Sweden 15 bipolar I, CSF cytokine Higher IL‑1β and lower Evidence for et al.[40] (Stockholm‑Linköping‑ 15 bipolar II, concentrations IL‑6 levels bipolar than neuroinflammation and Göteborg) all euthymic, assessed through controls. Patients with for IL‑1β involvement; 30 healthy immunoassay‑based recent manic/hypomanic cross‑sectional design volunteers protein array multiplex episodes had significantly is a methodological system; cross‑sectional higher IL‑1β levels than weakness those without Stich Germany 40 bipolar Paired CSF and serum Eight patients with bipolar Cross‑sectional design; et al.[41] (Freiburg, Greifswald) versus 26 samples analyzed through disorder versus 1 control possible that some controls with ELISA to detect the had AI > 1.4; 5 patients bipolar patients have pseudotumor concentration of versus 0 controls had autoimmune disorder; 1 cerebri against Toxoplasma oligoclonal bands of 5 patients with bipolar gondii, HSV types 1 and 2, disorder show evidence of CMV, and EBV. Specific neuroinflammation AI > 1.4 = intrathecal specific synthesis; oligoclonal bands = chronic neuroinflammation Isgren Sweden (Göteborg‑ 21 euthymic Measurements of serum CSF IL‑8 only was higher Cross‑sectional design et al.[42] Stockholm; England bipolar disorder and CSF concentrations in bipolar patients with is a limitation. The study [London]) patients versus of 11 cytokines; IL‑6 respect to controls; favors the presence of 71 age‑ and measured through validation reanalysis neuroinflammation in sex‑matched singleplex assay; IL‑1β, showed measurements to bipolar disorder, but healthy controls IL‑2, IL‑4, IL‑5, IL‑8/ have been valid, but also do not rule out that CXCL8, IL‑10, IL‑12, showed no correlation medication could account IL‑13, TNF‑α, and IFN‑γ between serum and CSF for the results; there was through the MSD 96‑well levels no correlation between multi‑array and multi‑spot central and peripheral human cytokine assay; data validation of IL‑8 measurement re‑analysis: 21 patients and 20 controls with Proseek Multiplex Inflammation I Jakobsson Sweden (Göteborg‑ 221 bipolar MCP‑1, YKL‑40, sCD14, MCP‑1, YKL‑40, and Cross‑sectional design et al.[18] Stockholm; England versus 112 TIMP‑1, and TIMP‑2 TIMP‑2 levels higher is a limitation. The study [London]) healthy controls measured through in patients with bipolar favors the presence for serum ELISAs disorder than controls of peripheral chronic sampling; 125 in CSF; serum and CSF inflammation and bipolar versus MCP‑1, YKL‑40 levels neuroinflammation in 87 controls for correlated, but differences bipolar disorder, but also CSF sampling in CSF levels between stresses the fact that the bipolar patients and two are independent controls were independent from serum levels Haarman The Netherlands 14 bipolar Dynamic 60‑min PET Significantly increased This study provides strong et al.[43] (Groningen) I versus 11 scan after injecting of [11C]‑(R)‑PK11195 evidence for the presence healthy controls [11C]‑(R)‑PK11195, a binding potential in bipolar of neuroinflammation ligand of the peripheral patients versus controls in bipolar I disorder, benzodiazepine receptor in the right ; but the sample was that constitutes a a trend toward the same small. Furthermore, the microglial marker finding was present for the cross‑sectional nature of left hippocampus the design does not allow to establish causality ELISA: enzyme‑linked immunosorbent assay; CSF: cerebrospinal fluid; HSV: herpes simplex virus; CMV: cytomegalovirus; EBV: Epstein‑Barr virus; IL‑1β: interleukin‑1 beta; IFN‑γ: interferon‑γ; MCP‑1: monocyte chemoattractant protein‑1; sCD14: soluble cluster of differentiation 14; TIMP‑1: tissue inhibitor of metalloproteinases‑1; TIMP‑2: tissue inhibitor of metalloproteinases‑2; PET: positron emission tomography; AI: antibody index; TNF‑α: tumor necrosis factor alpha; MSD: meso scale discovery; YKL‑40: human cartilage glycoprotein-39

Despite our care to avoid studies of peripheral immunity talked‑about, but not proved. Another striking result in bipolar disorder, many of these studies appeared of our search is the high number of reviews (n = 96), through our search. This was due to the fact that many and especially of those focusing on this issue (n = 40). of these papers speak about neuroinflammation without Most reviews mix‑up peripheral and brain studies, actually showing it. In many of these fascinating reaching unwarranted conclusions. The activation and fashionable papers, the question of how chronic of microglia must not be taken only as evidence of inflammation and immune derangement would “enter” neuroinflammation, because it might also be consistent the brain and produce neuroinflammation is much with a host of other functions, such as remolding the

260 Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 brain microstructure, contributing to plasticity and Financial support and sponsorship synaptic formation, and responses to environmental Nil. challenges.[48] So, it is possible that neuroglial activation and precisely aberrant neuroglial function, may be Conflicts of interest responsible for some neuronal miswiring that is There are no conflicts of interest. consistent with some psychotic symptoms that are frequently observed in bipolar disorder. REFERENCES

Summarizing, the evidence for the existence of 1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM‑IV). Washington, DC: neuroinflammation in bipolar disorder is more yes than American Psychiatric Association; 1994. no. However, the hypothesis of a preexisting peripheral 2. Akiskal HS, Pinto O. The evolving bipolar spectrum. Prototypes I, inflammation passing in the brain and establishing a II, III, and IV. Psychiatr Clin North Am 1999;22:517‑34, vii. 3. Griesinger W. (Herausgeber). Über Irrenanstalten und deren disease mode function, thus maintaining the disorder, Weiter‑Entwicklung in Deutschland. Arch Psychiatr Nervenkr is a mere supposition that is not based on evidence. 1868;1:8-43. (In German) Whereas neuroinflammation might lead to bipolar 4. Smith DJ, Harrison N, Muir W, Blackwood DH. The high prevalence symptoms, it is not always true that neuroinflammation of bipolar spectrum disorders in young adults with recurrent depression: toward an innovative diagnostic framework. J Affect causes bipolar disorder, and it might be that only Disord 2005;84:167‑78. rarely is so. It is more likely that stress interacts with 5. Guillaume S, Jaussent I, Jollant F, Rihmer Z, Malafosse A, Courtet P. the personal constitution and epigenetic factors to Suicide attempt characteristics may orientate toward a bipolar cause both neuroimmune dysreactivity and bipolar disorder in attempters with recurrent depression. J Affect Disord 2010;122:53‑9. disorder in susceptible individuals. As de Baumont 6. Mosolov S, Ushkalova A, Kostukova E, Shafarenko A, Alfimov P, [36] et al. stated, schizophrenia and bipolar disorder Kostyukova A, Angst J. Bipolar II disorder in patients with may “arise from shared genetic factors, but that the a current diagnosis of recurrent depression. Bipolar Disord resulting clinical phenotype is modulated by additional 2014;16:389‑99. 7. Binder EB, Nemeroff CB. The CRF system, stress, depression alterations mediated by microglia, possibly caused by and anxiety‑insights from human genetic studies. Mol Psychiatry interference of environmental factors at different times 2010;15:574‑88. during neurodevelopment and early life, and/or epistatic 8. McEwen BS. The neurobiology of stress: from serendipity to clinical interactions among groups of genes and environment.” relevance. Brain Res 2000;886:172‑89. 9. Fricchione G, Daly R, Rogers MP, Stefano GB. Neuroimmunologic All this should be borne in mind when projecting influences in neuropsychiatric and psychophysiologic disorders. investigations to explore the relationship between Acta Pharmacol Sin 2001;22:577‑87. neuroinflammation and bipolar disorder. 10. Trowbridge HO, Emling RC, Fornatora M. Inflammation: A Review of the Process. 5th edition. Hanover Park (IL): Quintessence Publishing; 1997. CONCLUSION 11. Selye H. A syndrome produced by diverse nocuous agents. Nature 1936;138:32. This review attempted to answer the question of 12. Selye H. and adrenals in the response of the organism to whether neuroinflammation plays a role in the injuries and intoxications. Br J Exp Pathol 1936;17:234‑48. pathophysiology of bipolar disorder. Direct and 13. Cannon WB. The Wisdom of the Body. New York: W.W. Norton and Company, Inc.; 1932. indirect evidence points to some degree of possibility 14. Brietzke E, Stertz L, Fernandes BS, Kauer‑Sant’anna M, that this is the case, but studies heretofore are much Mascarenhas M, Escosteguy Vargas A, Chies JA, Kapczinski F. heterogeneous in their methodology and conclusions, Comparison of cytokine levels in depressed, manic and euthymic thus suggesting caution. It appears that the topic of patients with bipolar disorder. J Affect Disord 2009;116:214‑7. 15. Gibney SM, Drexhage HA. Evidence for a dysregulated immune neuroinflammation in bipolar disorder is a much system in the etiology of psychiatric disorders. J Neuroimmune under‑investigated but over debated and highly Pharmacol 2013;8:900‑20. reviewed issue. Finally, PubMed should be trusted, 16. Grippo AJ, Scotti MA. Stress and neuroinflammation. In: Halaris A, but alternative search engines should be used, lest Leonard BE, editors. Inflammation in Psychiatry. Modern Trends in Pharmacopsychiatry. Basel (CH): S. Karger; 2013. p. 20‑32. precious articles are lost. 17. Schmitz G, Leuthauser‑Jaschinski K, Orsó E. Are circulating monocytes as microglia orthologues appropriate biomarker Acknowledgments targets for neuronal diseases? Cent Nerv Syst Agents Med Chem The authors wish to thank Ms. Mimma Ariano, 2009;9:307‑30. Ms. Ales Casciaro, Ms. Teresa Prioreschi, and 18. Jakobsson J, Bjerke M, Sahebi S, Isgren A, Ekman CJ, Sellgren C, Olsson B, Zetterberg H, Blennow K, Palsson E, Landen M. Monocyte Ms. Susanna Rospo, Librarians of the Sant’Andrea and microglial activation in patients with mood‑stabilized bipolar Hospital, School of Medicine and Psychology, disorder. J Psychiatry Neurosci 2015;40:250‑8. Sapienza University, Rome, for rendering precious 19. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, bibliographic material accessible, as well as their Ioannidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA statement for reporting systematic reviews and Secretary Ms. Lucilla Martinelli for her assistance meta‑analyses of studies that evaluate healthcare interventions: during the writing of the manuscript. explanation and elaboration. BMJ 2009;339:b2700.

Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015 261 20. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. 34. Hercher C, Chopra V, Beasley CL. Evidence for morphological Preferred reporting items for systematic reviews and meta‑analyses: alterations in prefrontal white matter glia in schizophrenia and the PRISMA statement. Int J Surg 2010;8:336‑41. bipolar disorder. J Psychiatry Neurosci 2014;39:376‑85. 21. Rolstad S, Jakobsson J, Sellgren C, Isgren A, Ekman CJ, 35. Fillman SG, Sinclair D, Fung SJ, Webster MJ, Shannon Weickert C. Bjerke M, Blennow K, Zetterberg H, Palsson E, Landén M. CSF Markers of inflammation and stress distinguish subsets of individuals neuroinflammatory biomarkers in bipolar disorder are associated with schizophrenia and bipolar disorder. Transl Psychiatry with cognitive impairment. Eur Neuropsychopharmacol 2015; 2014;4:e365. 25:1091-8. 36. de Baumont A, Maschietto M, Lima L, Carraro DM, Olivieri EH, 22. Ishizuka K, Igata‑Yi R, Kimura T, Hieshima K, Kukita T, Kin Y, Fiorini A, Barreta LA, Palha JA, Belmonte‑de‑Abreu P, Moreira Misumi Y, Yamamoto M, Nomiyama H, Miura R, Takamatsu J, Filho CA, Brentani H. Innate immune response is differentially Katsuragi S, Miyakawa T. Expression and distribution of CC dysregulated between bipolar disease and schizophrenia. Schizophr macrophage inflammatory protein‑1 alpha/LD78 in the Res 2015;161:215‑21. human brain. Neuroreport 1997;8:1215‑8. 37. Liu L, Li Y, Van Eldik LJ, Griffin WS, Barger SW. S100B‑induced 23. Hamidi M, Drevets WC, Price JL. Glial reduction in amygdala in microglial and neuronal IL‑1 expression is mediated by cell major depressive disorder is due to oligodendrocytes. Biol Psychiatry type‑specific transcription factors. J Neurochem 2005;92:546‑53. 2004;55:563‑9. 38. Griffin WS. Inflammation and neurodegenerative diseases. Am J 24. Frank O, Giehl M, Zheng C, Hehlmann R, Leib‑Mosch C, Seifarth W. Clin Nutr 2006;83:470S‑4S. Human endogenous retrovirus expression profiles in samples from 39. Griffin WS, Stanley LC, Ling C, White L, MacLeod V, Perrot LJ, brains of patients with schizophrenia and bipolar disorders. J Virol White CL 3rd, Araoz C. Brain interleukin 1 and S‑100 2005;79:10890‑901. immunoreactivity are elevated in Down syndrome and Alzheimer 25. Dean B, Gray L, Scarr E. Regionally specific changes in levels of disease. Proc Natl Acad Sci U S A 1989;86:7611‑5. cortical S100beta in bipolar 1 disorder but not schizophrenia. Aust 40. Soderlund J, Olsson SK, Samuelsson M, Walther‑Jallow L, N Z J Psychiatry 2006;40:217‑24. Johansson C, Erhardt S, Landen M, Engberg G. Elevation of 26. Foster R, Kandanearatchi A, Beasley C, Williams B, Khan N, cerebrospinal fluid interleukin‑1ss in bipolar disorder. J Psychiatry Fagerhol MK, Everall IP. Calprotectin in microglia from frontal cortex Neurosci 2011;36:114‑8. is up‑regulated in schizophrenia: evidence for an inflammatory 41. Stich O, Andres TA, Gross CM, Gerber SI, Rauer S, Langosch JM. process? Eur J Neurosci 2006;24:3561‑6. An observational study of inflammation in the central nervous 27. Weis S, Haybaeck J, Dulay JR, Llenos IC. Expression of cellular prion system in patients with bipolar disorder. Bipolar Disord protein (PrP (c)) in schizophrenia, bipolar disorder, and depression. 2015;17:291‑302. J Neural Transm 2008;115:761‑71. 42. Isgren A, Jakobsson J, Palsson E, Ekman CJ, Johansson AG, 28. Rao JS, Harry GJ, Rapoport SI, Kim HW. Increased excitotoxicity Sellgren C, Blennow K, Zetterberg H, Landen M. Increased and neuroinflammatory markers in postmortem frontal cortex from cerebrospinal fluid interleukin‑8 in bipolar disorder patients bipolar disorder patients. Mol Psychiatry 2010;15:384‑92. associated with lithium and antipsychotic treatment. Brain Behav 29. Kim HW, Rapoport SI, Rao JS. Altered arachidonic acid cascade Immun 2015;43:198‑204. enzymes in postmortem brain from bipolar disorder patients. Mol 43. Haarman BC, Riemersma‑Van der Lek RF, de Groot JC, Ruhe HG, Psychiatry 2011;16:419‑28. Klein HC, Zandstra TE, Burger H, Schoevers RA, de Vries EF, 30. Steiner J, Walter M, Gos T, Guillemin GJ, Bernstein HG, Sarnyai Z, Drexhage HA, Nolen WA, Doorduin J. Neuroinflammation in bipolar Mawrin C, Brisch R, Bielau H, Meyer zu Schwabedissen L, disorder – A [(11) C]‑(R)‑PK11195 positron emission tomography Bogerts B, Myint AM. Severe depression is associated with study. Brain Behav Immun 2014;40:219‑25. increased microglial quinolinic acid in subregions of the anterior 44. Deepmala, Slattery J, Kumar N, Delhey L, Berk M, Dean O, cingulate gyrus: evidence for an immune‑modulated glutamatergic Spielholz C, Frye R. Clinical trials of N‑acetylcysteine in psychiatry neurotransmission? J Neuroinflammation 2011;8:94. and : a systematic review. Neurosci Biobehav Rev 31. Rao JS, Keleshian VL, Klein S, Rapoport SI. Epigenetic modifications 2015;55:294‑321. in frontal cortex from Alzheimer’s disease and bipolar disorder 45. Duncan RE, Bazinet RP. Brain arachidonic acid uptake and turnover: patients. Transl Psychiatry 2012;2:e132. implications for signaling and bipolar disorder. Curr Opin Clin Nutr 32. Dean B, Gibbons AS, Tawadros N, Brooks L, Everall IP, Scarr E. Metab Care 2010;13:130‑8. Different changes in cortical tumor necrosis factor‑alpha‑related 46. Rapoport SI. Brain arachidonic and docosahexaenoic acid cascades pathways in schizophrenia and mood disorders. Mol Psychiatry are selectively altered by drugs, diet and disease. Prostaglandins 2013;18:767‑73. Leukot Essent Fatty Acids 2008;79:153‑6. 33. Gos T, Schroeter ML, Lessel W, Bernstein HG, Dobrowolny H, 47. Cagnin A, Kassiou M, Meikle SR, Banati RB. Positron emission Schiltz K, Bogerts B, Steiner J. S100B‑immunopositive astrocytes tomography imaging of neuroinflammation. Neurotherapeutics and oligodendrocytes in the hippocampus are differentially afflicted 2007;4:443‑52. in unipolar and bipolar depression: a postmortem study. J Psychiatr 48. Salter MW, Beggs S. Sublime microglia: expanding roles for the Res 2013;47:1694‑9. guardians of the CNS. Cell 2014;158:15‑24.

262 Neuroimmunol Neuroinflammation | Volume 2 | Issue 4 | October 15, 2015