The Pennsylvania State University

The Graduate School

Intercollege Program of Neuroscience

THE INTERACTION OF INFLAMMATION

AND GENDER/SEX IN BIPOLAR DISORDER

A Dissertation in

Neuroscience

by

Caitlin E. Millett

Ó 2018 Caitlin E. Millett

Submitted in Partial Fulfillment

of the Requirements

for the Degree of

Doctor of Philosophy

May 2018

The dissertation of Caitlin E. Millett was reviewed and approved* by the following:

Erika FH Saunders Chair, Department of Psychiatry Director, Mood Disorders Program Dissertation Advisor Chair of Committee

Patricia (Sue) Grigson Professor, Neural and Behavioral Science

James Connor Distinguished Professor of Neurosurgery, Neural and Behavioral Sciences and Pediatrics Vice Chair of Neurosurgery Research Director, Center for Aging and Neurodegenerative Diseases

Julio Fernandez-Mendoza Associate Professor, Department of Psychiatry

Colin Barnstable Professor and Chair, Neural & Behavioral Sciences Co-Director of the Graduate Program in Neuroscience

*Signatures are on file in the Graduate School

iii ABSTRACT

In recent decades, inflammatory/immune activation has emerged as an important factor in the onset and progression of mood disorders. Increased pro-inflammatory markers have been observed in the peripheral blood, cerebrospinal fluid, and postmortem brain tissue of individuals with Bipolar Disorder (BD) and Major Depressive Disorder (MDD). Further, pro- inflammatory cytokine concentrations may decline during periods of normal mood

(euthymia). There is evidence for a causal relationship between peripheral concentrations of the cytokine IL-6 in childhood, which can predict the onset of depressive symptoms in adulthood.

Here we have proposed that gender and/or sex moderates the relationship between inflammation and mood in BD. Women and men are known to have different symptomatology and comorbidities associated with a diagnosis of BD; women have higher rates of post-traumatic stress disorder (PTSD) and eating disorders, whereas men have higher rates of alcohol and cannabis abuse and conduct disorder. Further, women with BD are known to experience depressive symptoms at the onset of the illness, whereas men are more likely to experience mania at onset. Moreover, there are known sex-based differences in inflammatory/immune activation. In general females have more robust cell- mediated and humoral immune responses to antigenic challenges compared with males.

Women had higher self-reported feelings of depression and social isolation after endotoxin injection compared to men in a recent report. In contrast, it is known that the female sex hormone 17-β estradiol (E2) is broadly anti-inflammatory and confers protection against sepsis – a life-threatening complication of infection.

iv Our research aims to address gaps in knowledge related to inflammation and gender/sex in BD. In our first study (Chapter 2) we examined peripheral biomarkers from human participants with symptomatic BD. The biomarkers of interest were (1) neopterin – a marker of the acute phase immune response that is released from activated macrophages, and (2) zinc – an essential biometal and antioxidant in the human body. We postulated that neopterin would be increased in symptomatic BD and zinc would be decreased. Further, we hypothesized that gender/sex would moderate the relationship between biomarker concentration and mania and depression severity. Our results indicated that peripheral zinc concentrations were in fact reduced in symptomatic BD compared to a healthy control population. Further, we found that zinc concentrations were associated with severity of depressive symptoms in women with BD. In contrast, neopterin concentrations were associated with mania severity in men with BD. In this way, our findings underscore the importance of analyzing gender differences in biomarker studies. This research will help to advance the field of personalized medicine for BD.

In our second study (Chapter 3), we postulated that male and female mice would respond differently to an injection of the endotoxin lipopolysaccharide (LPS) – a component of the cell wall of gram-negative bacteria. Further, we expected that there would be a sex- based dichotomy in regard to the levels of oxidative stress and apoptosis in the brain after

LPS injection. We found that males had higher levels of depressive-like behavior after LPS, compared to saline treated males. Further, we found increased protein levels of superoxide dismutase in the male hippocampus. 28 days after the LPS injection, males showed signs of increased apoptosis in the hippocampus – higher cleaved caspase 3 protein levels and increased TUNEL staining. Females overall appeared more resilient against LPS administration – they did not show signs of apoptosis or oxidative stress in the

v hippocampus and did not increase depressive-like behavior. However, females appeared more ‘sick’ than males, indicated by decreased mobility in the open field test. Overall our study indicated that (1) inflammation may underlie hippocampal atrophy in some cases in males, (2) females appear to have a different behavioral response to inflammation compared to males, (3) females are more resilient against deleterious effects on LPS on the hippocampus compared to males.

vi TABLE OF CONTENTS

LIST OF FIGURES ...... viii

LIST OF TABLES ...... xi

ACKNOWLEDGEMENTS...... xii

Chapter 1 Introduction ...... 1

1.1 The History of Melancholy ...... 1 1.1.1 Hysteria ...... 2 1.2 Mood Disorders Defined ...... 4 1.3 Global Impact ...... 7 1.4 Burden...... 9 1.5 Gender/sex Influences Psychopathology and Prevalence ...... 11 1.6 Inflammation in Mood Disorders ...... 13 1.6.1 Inflammation Mechanisms to Depressive Symptoms ...... 16 1.6.2 Inflammation and Gender/Sex...... 22 1.7 Animal Models ...... 24 1.8 The Antioxidant Zinc ...... 25 1.9 Summary and Specific Aims ...... 30 1.10 References……………………………………………………………………….34

Chapter 2 Peripheral zinc and neopterin concentrations are associated with mood severity in bipolar disorder in a gender-specific manner ...... 43

2.1 Abstract ...... 43 2.2 Introduction ...... 44 2.3 Methods ...... 46 2.3.1 Participants ...... 46 2.3.2 Rating Scales ...... 46 2.3.3 Sample Collection and Biomarker Analysis ...... 47 2.3.4 Statistical Analysis ...... 48 2.4 Results ...... 48 2.4.1 Demographics and Clinical Data ...... 48 2.4.2 Peripheral Biomarkerd in BD and HC ...... 50 2.4.3 Differential Association of Zinc, Neopterin, and Mood by Gender ...... 53 2.4.4 Follow-up Zinc and Mood State Analysis ...... 56 2.5 Discussion ...... 57 2.6 References ...... 63

Chapter 3 The sex-specific effects of LPS on depressive-like behavior and oxidative stress in the hippocampus of the mouse...... 68

vii 3.1 Abstract ...... 68 3.2 Introduction ...... 70 3.3 Methods ...... 72 3.3.1 Animals ...... 72 3.3.2 Experimental Design ...... 72 3.3.3 Behavioral Analysis...... 73 3.3.4 Plasma Corticosterone and Zinc ...... 74 3.3.5 Brain Dissection and Preparation of Homogenates ...... 74 3.3.6 Western Blot ...... 74 3.3.7 RNA Isolation and qPCR...... 75 3.3.8 Immunofluorescence ...... 76 3.3.9 Microscopy...... 77 3.3.10 Quantification of Labeled Cells...... 77 3.3.11 Statistical Analysis ...... 77 3.4 Results ...... 77 3.4.1 General Parameters ...... 77 3.4.2 Experiment 1: Anxiety and Depression Behavioral Tests ...... 78 3.4.3 Oxidative Stress in the Hippocampus...... 78 3.4.4 Labile Zinc in the Hippocampus ...... 79 3.4.5 Markers of Plasticity ...... 79 3.4.6 Correlations ...... 79 3.4.7 Longitudinal Effects on Behavior ...... 85 3.4.8 Longitudinal Effects on Apoptosis and Plasticity ...... 85 3.4.9 Longitudinal Effects on TUNEL Staining ...... 85 3.5 Discussion ...... 89 3.6 References ...... 95

Chapter 4 Discussion ...... 99 4.1 Summary of Main Findings of the Dissertation ...... 99 4.2 Why Are the Sexes/Genders Different? ...... 101 4.3 How Good is the Model? ...... 108 4.4 Can We Treat Mood Disorders with Anti-Inflammatory Drugs? ...... 110 4.5 Is Zinc a Good Biomarker? ...... 113 4.6 Female Research Subjects ...... 114 4.7 Conclusions ...... 115 4.8 References ...... 117

viii LIST OF FIGURES

Figure 1-1: Mania and depression are comprised of a constellation of symptoms. Manic and depressive mood episodes are two components that make up a diagnosis of Bipolar Disorder. Each category is made up of a constellation of mood symptoms defined by the Diagnostic Statistical Manual of Mental Disorders (DSM) ...... 7

Figure 1-2: Routes of Communication Between the Immune System and Brain. There are various mechanisms by which the central nervous system (CNS) and peripheral immune system communicate. Pro-inflammatory cytokines such as IL-1β can gain access to the brain via Circumventricular organs, where the blood-brain-barrier (BBB) is “leaky”. There are also transporters for the cytokine IL-1β on the BBB. Two other modes of communication include vagal afferent fibers and monocyte infiltration into the CNS. Adapted from: Dilger and Johnson, 2008...... 17

Figure 1-3: The Metabolism of Tryptophan into Excitotoxic Molecules. The kynurenine pathway involves the conversion of tryptophan into kynurenine. INF-γ, interferon-gamma; TNF-α, tumor necrosis factor-alpha; TPH IDO, indolamine-2,3-dioxygenase; KAT, kynurenine aminotransferase; KMO, kynurenine 3-monooxygenase; KYNU, kynureninase; TDO, tryptophan-2,3- dioxygenase; TPH, tryptophan hydroxylase Adapted from: Mukherjee et al., 2018 ...... 19

Figure 1-4: Zinc is an Antioxidant. The biometal zinc has a wide array of functions in the human body and is found in every cell, in every organ. Zinc has an important role in maintaining cellular oxidative homeostasis and plays a part as an antioxidant by associating with antioxidant proteins such as metallothionein (MT), superoxide dismutase-1 (SOD1). Zinc can also inhibit the action of certain transporters and enzymes that generate oxidative radicals, like the enzyme NADPH oxidase, which produces superoxides, and the NMDA receptor. Adapted from: Marreiro et al., 2017 ...... 26

Figure 2-1: Serum Zinc in Symptomatic Bipolar Disorder. Box and whisker plot for peripheral zinc in bipolar disorder (BD) and healthy control (HC) groups at baseline. The range of zinc concentrations was larger for the bipolar population (Range=129.25), compared to healthy controls (Range=63.5). Women with bipolar disorder had two outliers, and men had one. Middle bar in each box plot is the median value...... 51

Figure 2-2: Lower Serum Zinc in BD Compared to HC. Two-way Analysis of Covariance of peripheral zinc concentrations by diagnosis and gender. A main effect showed that participants with bipolar disorder (BD) had significantly lower zinc concentrations on average (p=.04), as compared to healthy controls (HC). Bipolar disorder N=24 (men N=13 and women N=11) and healthy controls

ix N=31 (men N=13 and women N=18). Data represents marginal means ±SE. Marginal means are means adjusted for the covariate BMI...... 52

Figure 2-3: Zinc Correlates with Deprssion Severity in Women. There is a significant interaction between zinc and gender in prediction of depression severity (HDRS-21 +AT). Women have a positive correlation between zinc and depression severity (Pearson’s R=0.61), and men have a negative correlation (Pearson’s R=-0.33) ...... 54

Figure 2-4: Neopterin is Associated with Mania Severity in Men. Neopterin predicts mania severity in men. There is a significant interaction between gender and neopterin in prediction of mania severity (YMRS). The correlation was stronger for men (Pearson’s R=.74) than women (Pearson’s R=.38)...... 55

Figure 2-5: Baseline and Follow-up Values for Participants with Bipolar Disorder. Zinc concentration significantly increased at follow up. N=12 (men N=4 and women N=8). Data represented as mean ± SE. ***p<.001, **p<.01, *p<.05.)...... 56

Figure 3-1: General Parameters. Data is represented as mean ±SEM, n=12 mice/group for (a) weight, (b) CORT (n=6-8) and (c) pooled plasma zinc samples...... 80

Figure 3-2: The Effect of LPS on Male and Female Behavior. (a) Timeline of behavioral testing. (b) The forced swim test determined total immobility time for males and females. (c) Locomotor activity. (d) Total distance traveled. (e) Center square entries. ‘*’ = significantly different from control, p<0.05. Data is represented as mean ±SEM, n=12 mice/group...... 81

Figure 3-3: Figure 3.3. Examination of transcript and protein in the male and female hippocampus. (a) qPCR in males and (b) females, all set relative to saline (as one). (c) Males showed significantly increased SODI after LPS. (*) significantly different from saline p<0.05. (#) p<0.09. Data is represented as mean ±SEM...... 82

Figure 3-4: Quantification of Labile Zinc in the Hippocampus. TSQ quantification goes down in females given LPS (c) but not in males (b). (*) = significantly different from control, p<.05. Data is represented as means ±SD, n=4 mice/group. Bar is 100μm...... 83

Figure 3-5: Longitudinal Examination of Protein Levels in the Hippocampus. (a) Timeline of the longitudinal study. (b) Forced swim test behavior at three timepoints in males and females. (c) Males increased cleaved caspase-3 and decreased BDNF protein levels after LPS. (d) males had increased TUNEL (+) cells after LPS. (*) significantly different from control, p<.05. Data is represented as means ±SEM, n=3 mice/group...... 86

x Figure 3-6: Male and Female Mice Respond Differently to Lipopolysaccharide (LPS). In our model, female mice decreased their level of general mobility, but did not develop depressive-like behaviors in the forced swim test (FST). However, male mice given LPS increased immobility in the FST, indicating a depressive-like phenotype. We found that superoxide dismutase was increased in the male hippocampus 24 hours after LPS, indicating increased oxidative stress. By 28 days, male mice had increased levels of hippocampal apoptosis and decreased brain-derived neurotrophic factor, indicating that the male hippocampus is less resilient that the female hippocampus in response to endotoxin challenge. In future, we hope to elucidate the level of hippocampal atrophy that may occur in this model in males, and the effect this has on memory tasks and cognitive decline...... 88

xi LIST OF TABLES

Table 2-1: Demographic and Clinical Data of the Participants with Bipolar Disorder (BD) and Healthy Controls (HC) at Baseline. Data represented as Mean (± SD) or N (%). P values represent comparisons between ‘BD all’ and ‘HC all’ groups. *Neopterin N values (BD=27 men=14, women=13; HC=31 men=13, female=18)...... 49

Table 3-1: Correlational Analysis of Transcript Levels and Behavioral Parameters in LPS Treated Animals. (*) significantly different from control, p<0.05. Data is represented as mean ±SEM. N= 4 (females) and 8 (males) mice/group. MT (metallothionein); SOD (superoxide dismutase); GSR (glutathione reductase); GFAP (glial fibrillary acidic protein); IBA1 (ionized calcium-binding adaptor molecule 1); SYN1 (synapsin-1); BDNF (brain derived neurotrophic factor) ...... 84

xii ACKNOWLEDGEMENTS

First, I would like to thank Dr. Erika Saunders, my thesis advisor and mentor, for giving me the freedom to pursue my own projects and develop as an independent scientist. I am grateful to have been your first graduate student. You have truly led by example, with intelligence, grace, and authenticity.

In addition to our group, I was fortunate enough to receive support from many wonderful faculty in the department. I would like to thank Professor Sue Grigson for being a consistent source of support and encouragement over the last five years. Through my ups-and-downs you have always seen the best in me. I will always remember you for your wisdom, kindness, and humor. Thank you to Professor Julio Fernandez-Mendoza for your unique perspective and advice. I am also thankful for my rotation advisors, including

Professors James Connor and Ian Simpson, from whom I learned many basic skills.

Thanks to the late Professor Thomas Pritchard for grading my candidacy exam – writing (and re-writing…) that essay helped me grow into a tougher academic. Thanks also to the late Professor Robert Bonneau for lending me your old Janeway Immunology textbook – I think of you whenever I open it.

I am deeply grateful for the hands-on advice and support I received from excellent postdocs, graduate students, and research associates. Dr. Brett Phillips was a scientific lifeline throughout my PhD (and he made excellent raspberry cheesecake). Dr. Dahlia

Mukherjee was a supportive presence and encouraged me with her kind words. We have also had our share of laughs and margaritas. Thanks to Dr. Samina Alam for her limitless kindness, patience, and baked goods – I am honored to be your ‘blonde bombshell’. Many

xiii thanks to Dr. Jenny Lee for teaching me how to do PCR, even when you were a busy graduate student!

I would also like to thank the Penn State Hersey Commission for Women for nurturing my leadership potential. Thanks to Catherine Caruso for being my mentor and friend over the last few years – I will miss our dates to the Hershey Recreational Center for swimming and conversation.

I am grateful to all my friends in Hershey. Thank you to Ashley McFalls for your wit and wisdom – I will truly miss our lingering walks around the cornfields. Thank you to Shirley

Li for being one of my first friends in Hershey and a great roommate – your hot pot and beef ribs kept me alive our second year. Finally, thanks to Raz Abdulqadir for inspiring me to be bolder. I will miss our long chats in lab and lunch dates at Sorrento’s.

I would like to dedicate this thesis to the people who have supported me in reaching my academic and personal goals – my parents.

I am grateful to my father, Robert Millett, for pushing me to be brave and take risks in life, and for making time to visit me every semester. Our adventures walking the rail trail, at the TROP range, and the Tröegs brewery are memories that I treasure.

Thank you to my mother, Katherine Millett, for emotional support and great advice and perspective throughout the hard times as well as the good ones. I think back fondly on our adventures in Gettysburg, the Pennsylvania Renaissance Faire, and listening to country music as we drove through cornfields. Thank you, mom, for telling me I can ‘have it all’ and that ‘the world is my oyster’ – you have no idea how much those words motivate me.

xiv Finally, I dedicate this thesis to my wonderful boyfriend Anton Mazurenko. We have had so many adventures together. You are a positive force in my life and I love you more each day. This thesis is also dedicated to our cat Chloe.

- 1 -

Chapter 1

Introduction

“When I lie waking all alone, Recounting what I have ill done, My thoughts on me then tyrannise, Fear and sorrow me surprise, Whether I tarry still or go, Methinks the time moves very slow. All my griefs to this are jolly, Naught so mad as melancholy.”

-Robert Burton, The Anatomy of Melancholy (1621)

1.1 The History of Melancholy

Writing in the fifth century BCE provided the first known description of melancholy – the

Greek name for psychological states of depression (Horwitz et al., 2016). Melancholy is one of but a few disorders with an ancient written, recorded history (Horwitz et al., 2016). The term melancholia first appeared in the Corpus Hippocratium – a collection of old Grecian medical writings, in which Hippocrates wrote a succinct definition of melancholy – “if a fright or despondency lasts for a long time, it is a melancholic affection” (Hippocrates 2002a,

2002b). In addition to fear and sadness, Hippocratic writers also made mention of melancholic symptoms including “aversion to food, despondency, sleeplessness, irritability, and restlessness” (Hippocrates 2002a, 2002b). Interestingly, these symptoms are similar to the contemporary definition of a Major Depressive Episode (MDE) in the Diagnostic

Statistical Manual of Mental Disorders – the DSM-V (APA, 2013).

Ancient Greeks viewed mental and physical disease regarding four bodily humors – liquid-like substances which, when unbalanced, led to various conditions (Horwitz et al.,

- 2 - 2016). The four bodily humors were: blood, phlegm, yellow bile, and black bile. The term

“melancholy” derives from the ancient Greek word “melania chole,” which meant “black bile”

(Hippocrates 2002a, 2002b). Further, each humor possessed two of four properties – hot, cold, moist or dry (Horwitz et al., 2016). Melancholic humors were characterized by coldness and dryness (Horwitz et al., 2016).

The Hippocratic ideologies relating to the humoral theory of melancholy dominated medical thinking for subsequent eras until the end of the 18th century. Robert Burton, whose work The Anatomy of Melancholy (1621) – considered a masterwork of all serious discussions on melancholy – was situated within the Hippocratic traditions (Horwitz et al.,

2016). Burton believed there were three main components to melancholy – physical symptoms, cognition, and mood (Horwitz et al., 2016). Burton defined melancholy this way

– “Fear and sorrow are the true characters and inseparable companions of melancholy.” He also upheld the contextual prerequisites for a melancholic diagnosis – "a kind of dotage without a fever, having for his ordinary companions, fear and sadness, without any apparent occasion" (Horwitz et al., 2016).

1.1.1 Hysteria

The first mental disorder attributable disproportionately to women since the second millennium BCE is hysteria (Tasca et al., 2012). Hysteria has 4000 years of history – up until

1980 when the concept of “hysterical neuroses” was removed from the DSM-III (Tasca et al., 2012). Hysteria is derived from the Greek “hystera”, meaning uterus (Tasca et al., 2012).

Ancient physicians attributed a number of female maladies to a starved or misplaced womb

(Tasca et al., 2012). Ancient Egyptians also made mentioned to hysteria-like symptoms – the Eber papyrus (1600 BC) contains references to depressive symptoms and traditional symptoms of hysteria, described as seizures and an overall sense of suffocation and

- 3 - imminent death (Tasca et al., 2012).

There were many proclaimed causes for hysteria. In Timaeus, Plato argued that the uterus becomes "sad and unfortunate when it does not join with the male and does not give rise to new birth" (Tasca et al., 2012). Hippocrates expanded on this idea; he said that especially in virgins, widows, single or sterile women, the “bad” uterus produces toxic fumes and takes to wandering around the body (Tasca et al., 2012). A "bad" uterus was thought to cause various disorders like anxiety, a sense of suffocation, tremors, sometimes convulsions and paralysis (Cosmacini, 1997a, 1997b; Sterpellone, 2002). Ancient Greek mythology also contains references to hysteria (Tasca et al., 2012). The Argonaut

Melampus, a physician, believed that women’s madness derived from a uterus poisoned by venomous humors due to “uterine melancholy” (Cosmacini, 1997a, 1997b; Sterpellone,

2002).

Over the millennia, treatments for hysteria included “moving” an upwardly displaced uterus back to its original location by placing malodorous or acrid substances near a woman’s face and scented ones near her vagina (Tasca et al., 2012). Hippocrates suggested women get married and have a satisfactory sexual life to cure them of hysteria

(Tasca et al., 2012). Further, he advised women to fumigate themselves with acrid or fragrant scents near their face and genitals to push the uterus back into place (Tasca et al.,

2012). The Roman physician Galen treated hysteria with various herbs: hellebore and mint, as well as laudanum, belladonna extract and valerian (Tasca et al., 2012). He also suggested that young women avoid new stimuli (Cosmacini, 1997a, 1997b). Another Greek physician (from the 2nd century AD), Soranus, wrote a treatise on women’s diseases and postulated that they arose from the labors of procreation (Tasca et al., 2012). In contrast to his predecessors, Soranus suggested that sexual abstinence and perpetual virginity in women is the ideal condition, and the “hysterical body” should be treated with care: hot

- 4 - baths, massages, and exercise (Penso, 2002).

1.2 Mood Disorders Defined

“Dearest, I feel certain that I am going mad again. I feel we can't go through another of those terrible times. And I shan't recover this time. I begin to hear voices, and I can't concentrate. So, I am doing what seems the best thing to do.”

-Virginia Woolf, 1941

In the present day, mood disorders are characterized by the Diagnostic Statistical Manual of Mental Disorders (DSM), which is currently in its fifth iteration (APA, 2013). Mood disorders are defined as mood disturbances such as severe depression or depression altering with mania, as depicted in Figure 1-1 (APA, 2013). Mood disorders are a broad class of psychiatric disorder that subsumes two general types of mental illness: Bipolar

Disorder (BD) and Major Depressive Disorder (MDD). Both BD and MDD have a set of diagnostic criteria and symptomatology that are at times overlapping and distinct. Moreover, both are recurrent and chronic mental illnesses associated with functional decline, mortality and significant health care costs (Ferrari et al., 2013).

BD and MDD include the experience of major depressive episodes – in BD it is called

‘bipolar depression’ – that are defined as states of depression lasting two weeks or more

(APA, 2013). BD differentiating symptom from MDD is the experience of mania, which is a component of BD characterized by the experience of elation, unbounded euphoria without sufficient reason, and grandiose thoughts or feelings about personal abilities (APA, 2013).

The extreme poles of the mood spectrum for both BD and MDD include the experience of psychosis (APA, 2013). Psychosis entails some loss of contact with reality

(APA, 2013). Psychotic episodes are more common in BD than in MDD and may include the experience of auditory or visual hallucinations (e.g., hearing voices) and delusions –

- 5 - fixed, false beliefs that are "maintained despite evidence to the contrary" (APA, 2013).

BD is itself a broad category that includes a few different variations: BD type-I, BD type-II, cyclothymia, and BD- not otherwise specified (NOS). The BD-I criteria in the DSM-

V do not require the person to have experienced psychosis or a major depressive episode

(APA, 2013). However, most people whose symptoms meet the criteria for a fully syndromal manic episode also experience depressive events during their lives (Merikangas et al.,

2007). BD-II requires the lifetime experience of at least one bout of major depression and at least one hypomanic episode (APA, 2013). Cyclothymia is a diagnosis in adults that experience two years of both hypomanic and depressive periods without ever fulfilling the criteria for an episode (APA, 2013).

BD and MDD are made up of a constellation of mood symptoms. Both consist of states of depression – MDEs – and are diagnosed using the same criteria in the DSM-V

(APA, 2013). A diagnosis of an MDE requires the presence of five or more symptoms during the same two-week period (APA, 2013). Mood symptoms must represent a change from previous functioning (APA, 2013). Also, at least one of the signs must be: (1) depressed mood most of the day (e.g., sad, empty, and hopeless), or (2) diminished interest or pleasure in all or most activities (anhedonia). Besides these two main symptoms, a diagnosis of a depressive episode may also involve the presence of (3) significant weight change or a change in appetite nearly every day; (4) insomnia or hypersomnia, psychomotor agitation

(restlessness) or retardation (slowness); (5) fatigue; (6) excessive or inappropriate guilt; (7) diminished ability to concentrate; (8) recurrent thoughts of death or suicidal ideation (APA,

2013). These symptoms may cause clinically significant distress and impairment in social, occupational and other areas of functioning (APA, 2013).

A diagnosis of BD-I requires the presence of a “distinct period of abnormally and persistently elevated exposure or irritable mood and abnormally and persistently increased

- 6 - activity or energy, lasting at least one week and present most of the day, or any duration if hospitalized (APA, 2013).” During the period of mood elevation, three or four of the following symptoms are present to a significant degree: (1) inflated self-esteem (grandiosity); (2) reduced need for sleep; (3) more talkative than usual; (4) racing thoughts; (5) attention is quickly drawn away to unimportant or irrelevant external stimuli; (6) increased goal directed activity (social, sexual, at work or school) or agitation (not goal-directed); (7) risky behaviors, e.g., sexual promiscuity, buying sprees, and foolish business ventures (APA, 2013).

Additionally, the mood disturbance must be sufficiently severe to cause marked impairment in social or occupational functioning or necessitates hospitalization, or there are psychotic features (APA, 2013).

The definition of a hypomanic episode is like the preceding description, but the presenting mood symptoms are only required to last four consecutive days instead of seven

(APA, 2013). However, the episodes must be an unequivocal change in functioning, and the mood disturbance must be noticeable by others (APA, 2013). In contrast to a full manic episode, hypomania may not be severe enough to impair social or occupational functioning

(APA, 2013). Mixed features involve the presence of mania and depression concurrently

(APA, 2013).

- 7 -

Figure 1-1: Mania and depression are comprised of a constellation of symptoms.

Figure 1.1- Manic and depressive mood episodes are two components that make up a diagnosis of Bipolar Disorder. Each category is made up of a constellation of mood symptoms defined by the Diagnostic Statistical Manual of Mental Disorders (DSM).

1.3 Global Impact

Mood disorders are common mental disorders that occur in people of all ages across all world regions (Ferrari et al., 2013). The Global Burden of Disease (GBD) study, a global research conducted every ten years, estimated that the burden of depressive disorders increased by 37.5% between 1990 and 2010 (Ferrari et al., 2013). In 1990 depressive disorders – which include MDE – were the fourth leading cause of burden worldwide, after lower respiratory infections, diarrheal diseases, and conditions arising during the perinatal period (Ferrari et al., 2013).

- 8 - In the year 2000, depressive disorders were the third leading cause of burden after lower respiratory infections and diarrheal diseases (Ferrari et al., 2013). Depressive disorders were also the leading cause of disability in 2000, responsible for 13.4% of disability in women and 8.3% in men (Ferrari et al., 2013). In 2010, depressive disorders accounted for 8.2% of the total global disability, making it the second leading cause of disability around the world, second to lower back pain (Ferrari et al., 2013). Depressive disorders are estimated to affect over 300 million people in the world today (Ferrari et al., 2013).

Similarly, there has been a global increase in BD burden since 1990 (Ferrari et al.,

2016). In 2013, BD was the 54th leading cause of burden, and in 1990 it was the 76th leading cause of global burden (Ferrari et al., 2016). Further, in 1990 there were 32 million cases of BD globally, and 48 million in 2013, equivalent to a 49% increase in cases (Ferrari et al., 2016). The lifetime prevalence of BD-I disorder is estimated to be 0.6%, BD-II is 0.4%, sub-threshold BD was 1.4%, and Bipolar Spectrum (BPS) is 2.4% (Merikangas et al., 2011).

Some estimates indicate that BD, and subthreshold cases of BD, could affect as much as

4-6% of the world’s population (Angst et al., 2002a, 2002b; Judd et al., 2003; Merikangas et al., 2007; Zimmermann et al., 2009). The high disability weight for BD – primarily because of its early onset and chronicity across the lifespan – meant that the global rate of BD disability was like more prevalent conditions such as asthma and Alzheimer’s disease

(Ferrari et al., 2016; Merikangas et al., 2011).

Inappropriate treatment of BD is a severe problem in the U.S. (Merikangas et al.,

2007). Misdiagnosis of BD is common and results in the delay of appropriate treatment

(Bowden, 2005a). The most common misdiagnosis for BD is MDD, which risks the inappropriate use of antidepressants; this may result in rapid cycling and manic episodes

(Bowden, 2005a; Hirschfeld and Vornik, 2004). However, overall, treatment rates are low for BD, especially in low and middle-income countries (Merikangas et al., 2011).

- 9 - People with BD are also at a high risk for suicide attempt – about 15 times greater than the general population (Ferrari et al., 2016). Untreated BD can lead to suicide in up to

20% of cases (Ferrari et al., 2016). Overall an estimated 800,000 people die due to suicide every year around the world, and suicide is the second leading cause of death for 15-29- year-olds globally (Phillips et al., 2014). Most suicides occur in low and middle-income countries (~78%) (Phillips et al., 2014). The most common methods for suicide are ingestion of pesticides (30%), hanging, and firearms (Phillips et al., 2014). The global male-to-female ratio of suicide rates was 1.7, indicating that there were more male suicides than female suicides (Phillips et al., 2014). However, more women attempt suicide than men (Tsirigotis et al., 2011).

1.4 Burden

BD is a chronic condition with an average age of onset of 21 years (Kessler et al., 1997).

More than 90% of people who have one manic episode will have recurrent mood episodes

(APA, 2013). Patients with BD-I were shown to be symptomatically ill 47.3% weeks of their lives throughout a 12.8-year follow up period (Judd et al., 2002). Further, it was shown that recurrent depressive symptoms (31.9% of weeks) predominated over manic and hypomanic symptoms (8.9% of weeks) or cycling/mixed symptoms (5.9% of weeks) (Judd et al., 2002).

It is well known that individuals with BD suffer from substantial functional impairment when they experience manic, depressive or mixed episodes (Miklowitz and Johnson, 2006).

Moreover, approximately 30% of people with BD continue to suffer impairment in work role function when euthymic (Miklowitz and Johnson, 2006; Rosa et al., 2010). Sub-syndromal depressive symptoms were associated with the level of impairment measured by working ability, home management, private leisure, social leisure and family relationships (Rosa et al., 2010). BD diagnosis is inversely correlated with age and education, elevated among the

- 10 - previously married, compared to the currently married and the unemployed and disabled compared to the employed (Merikangas et al., 2011).

Mental illness comorbidities are common in people with BD. Co-occurring anxiety affects about 75% of people with BD (Krishnan et al., 2005; Kilbourne et al., 2004). Over

97% of patients with BD-I meet criteria for a concurrent psychiatric illness, and co- occurrence of three or more disorders is higher than co-morbidity with only one disorder across the bipolar spectrum (Krishnan et al., 2005; Kilbourne et al., 2004). Also, there are high rates of substance abuse disorder (alcohol, cocaine and opiate use), which is seen in about 50% of people with BD (Krishnan et al., 2005; Kilbourne et al., 2004). There is also a higher rate of sexually transmitted diseases in populations with BD (Krishnan et al., 2005;

Kilbourne et al., 2004).

General medical conditions also cluster densely with BD populations and result in earlier and increased mortality from cardiovascular, respiratory and endocrine-related issues. Cardiovascular disease and diabetes contribute to a significant proportion of the burden to people with BD (Krishnan et al., 2005; Kilbourne et al., 2004). Also, pulmonary conditions (e.g., chronic obstructive pulmonary disease) are more prevalent in BD compared to national samples (Krishnan et al., 2005; Kilbourne et al., 2004).

Kemp and colleagues (2010) found that metabolic illness – a group of risk factors predominantly associated with the development of heart disease, stroke, and type II diabetes – appears to be associated with higher depressive symptom severity and poorer treatment outcomes among patients with rapid cycling BD receiving lithium and valproate

(Kemp et al., 2010). Importantly, clinical and epidemiological studies reveal that more than half of BD patients are either overweight or obese Kemp et al., 2010). Obese people experience an increased number of depressive and manic episodes, but they also relapse more quickly following stabilization, primarily into depressive episodes Kemp et al., 2010).

- 11 - Further, cardio-metabolic risk factors (such as hypertension) appear to be associated with higher psychiatric symptom severity in BD (Kemp et al., 2010).

It has been proposed that the underlying pathophysiology BD furthers the development of medical disorders like metabolic syndrome (Kemp et al., 2010; Leboyer et al., 2012). There are standard underlying biological mechanisms in both disorders, including abnormal autonomic function, oxidative stress, glucocorticoid signaling, and altered energy biosynthesis (Kemp et al., 2010). Therefore, it has proposed that medical comorbidity might be a core feature of BD, and it can, therefore, be conceptualized in a whole-body framework, as an inflammatory condition (Leboyer et al., 2012).

1.5 Gender Influences on Psychopathology and Prevalence

When considering biological factors that influence the expression of mood disorders, we must consider the effect of sex and gender. First, a note to clarify the use of terms ‘sex’ and

‘gender’ in clinical studies – these terms are defined in the American Psychological

Association Dictionary of Psychology (APA, 2015) as follows:

“Sex (n): the traits that distinguish between males and females. Sex refers especially

to physical and biological traits, whereas GENDER refers especially to social or

cultural traits, although the distinction between the two terms is not regularly

observed.”

The terms ‘sex’ and ‘gender’ are often used interchangeably in clinical studies, unlike in preclinical rodent studies, where ‘sex’ is used exclusively. Many studies in humans (as discussed below) use the word ‘gender’ as an overarching term to denote both the social and biological underpinnings related to observed sex differences. In our clinical study

(Chapter 2) we chose to use the term ‘gender’ as well.

- 12 - A representative US study – the National Comorbidity Survey Replication Study

(NCS-R) – surveyed over 9000 people and quantified the lifetime risk of ‘any mood disorder’ to be 1.5x higher in women compared to men (Kessler et al., 2005). This result is in line with the most commonly observed gender differences in mood disorders – that women have twice the lifetime rates of depressive and anxiety disorders compared to men (Kessler et al., 1994; Gater et al., 1998). Moreover, women with BD have higher rates than men of BD-

II – which is marked by a higher preponderance of depressive symptoms (Baldassano et al., 2005). It is thought that women may have higher rates of BD-II because of higher rates of depression among women, and the decreased likelihood of women to experience purely manic episodes in BD (Kessler et al., 1994; Curtis 2005).

Gender differences also exist in the life course of mood disorders. There is no significant gender difference in age at onset for BD, or in the total number of mood episodes

(Merikangas et al., 2011). However, women with BD were more likely to experience depressive symptoms than men and are more likely to suffer depression at the onset of BD, and more men than women report mania at the start of BD-I (Hendrick et al., 2000a, 2000b;

Kawa et al., 2005; Parker et al., 2014). In contrast, women experience MDD more often than men, and tend to have a lower age at onset, a higher incidence rate, and a more chronic course than men (Essau et al., 2010; Azorin et al., 2014a, 2014b; Parker et al., 2014;

Schuch et al., 2014). Lower age-at-onset of MDD predicts a worse course for depression in women. Therefore, childhood depression is a more serious risk factor for girls than boys

(Essau et al., 2010).

Gender differences have been shown to exist in the symptomatology of mood disorders. Women with BD are more prone to mixed episodes – characterized by agitation, refractory anxiety, intractable insomnia, suicidality, obsessions and impulses, and hypersexuality (Akiskal et al., 1998). Women with BD are also more prone to rapid cycling

- 13 - – the experience of four or more episodes of depression or mania in one year – which is associated with increased rates of depression, suicidality, substance abuse, anxiety, and hypothyroidism (APA, 1994; Coryell et al., 2003; Calabrese et al., 1999).

There are also gender-based differences in psychiatric comorbidities with mood disorders; Women with BD are more at risk for post-traumatic stress disorder (PTSD), co- morbid anxiety disorders, weight change, appetite change, and bulimia (Baldassano et al.,

2005; Kawa et al., 2005; Saunders et al., 2012). Similarly, women with MDD were more likely to have comorbid panic and agoraphobia, and a lifetime diagnosis of an anxiety disorder compared to men (Schuch et al., 2014). Further, the risk for a migraine and the impact of a co-morbid migraine on poor mood outcome is more significant in women with

BD (Altshuler et al., 2010; Saunders et al., 2014). Moreover, poor sleep quality affects women with BD more than men with BD by increasing frequency and severity of mood episodes (Gruber et al., 2009, 2011; Saunders et al., 2015). Men with BD were more likely to report behavioral problems and being unable to hold a conversation during mania

(Hendrick et al., 2000a, 2000b). Gender differences in measures of impulsivity and mood instability during depression might contribute to the male preponderance in cases of completed suicide (Azorin et al., 2014). Also, more men than women with BD exhibited pathological gambling and conduct disorder as well as alcohol and cannabis abuse/dependence (Kawa et al., 2005).

1.6 Inflammation in Mood Disorders

The word ‘immune’ is derived from Latin immunis, meaning “free” or “untouched” (Menche,

2012); in line with its definition, the function of the immune system is primarily to protect the body from infection by pathogens in the environment, to which we are exposed to millions daily, through contact, ingestion, and inhalation (Alberts et al., 2002). There are two main

- 14 - branches of the immune system, which work closely together – innate and humoral immunity

(Menche, 2012). The innate immune system is the first line of defense against invading pathogens and includes cells such as phagocytes like macrophages, dendritic cells, neutrophils, as well as natural killer cells, mast cells, eosinophils, and basophils (Janeway et al., 2001). The cells of the innate immune system react to pathogen-associated molecular patterns, which can be cell surface markers or secreted markers from pathogens (Alberts et al., 2002). Innate immune cells can produce pro-inflammatory (e.g., IL-1β, IL-6, and TNF-

α) and anti-inflammatory (e.g., IL-1Ra, IL-4, IL-10, IL-11, and IL-13) cytokines, and chemokines, which are signaling messengers to propagate or inhibit the inflammatory cascade, respectively (Zhang et al., 2017a, 2017b).

For the past several decades increasing evidence has shown dysregulation of the immune/inflammatory system function in mood disorders. One of the first instances of this was demonstrated in a cohort of patients being treated with the cytokine interferon-α (IFN-

α) for hepatitis C infection (Chiu et al., 2017). Today, IFN-α induced depression is known to be a common and severe psychiatric disorder during IFN-α therapy (Chiu et al., 2017). It is understood that administration of IFN-α leads to depressed mood in as much as 50% of patients, and up to 30% develop depression in the first three months following treatment

(Udina et al., 2012). Moreover, almost all patients ultimately develop adverse effects from

IFN-α, including fatigue, malaise, muscle or joint pain, anorexia, apathy, and cognitive impairment (Davis et al., 1998a; Hoofnagle et al., 2006; Schaefer et al., 2006). This supports the idea that endogenous cytokines contribute to states of depression (Chiu et al., 2017).

Despite the serendipitous findings related to IFN-α and depressive symptoms, these were not the first recorded examples of a link between inflammation and mood. In fact, ancient scholars made notes related to this phenomenon, for instance Hippocrates (460-

370 BCE) observed that fever could calm epilepsy, and another Greek physician, Galen of

- 15 - Pergamon (130-210 AD), also described a case of melancholy ameliorated because of an attack of ‘quartan’ fever – a fever recurring every three to four days (Karamanou et al.,

2013). The first modern study to document a causal link between inflammation and mood was by the historically controversial figure – “teetering between brilliance and despicableness” – Julius Wagner-Jauregg (1857-1940) (Gartlehner et al., 2012). Wagner-

Jauregg was an Austrian neuropsychiatrist who used fever therapy to “treat” his patients suffering from neurosyphilis – a neurodegenerative disorder affecting the brain and spinal cord brought on by an untreated syphilitic infection and causing mood disturbances such as depression and agitation (Gartlehner et al., 2012; Karamanou et al., 2013). Wagner-Jauregg injected his ill patients with various agents including streptococci, tuberculin, and malaria

(Karamanou et al., 2013). In the years following his discovery, different alternative methods were employed to increase the body temperature of patients, including hot baths or a heating cabinet, electromagnetic fields, sulfur injections, and parasites (Karamanou et al.,

2013). These methods were used on neurosyphilis cases well into the mid-20th century until the introduction of penicillin made them obsolete (Karamanou et al., 2013). Also, around this time in the mid-20th century, the advent of monoamine oxidase inhibitor (MAOI) and tricyclic antidepressants sidelined the inflammation-based theories of mood disturbance for the burgeoning monoaminergic hypothesis – involving the neurotransmitters and neuromodulators derived from tyrosine, tryptophan, phenylalanine and others, e.g., serotonin, dopamine, norepinephrine, and epinephrine. (López-Muñoz and Alamo, 2009).

Increasing evidence has shown that the immune system is dysregulated in mood disorders (McNamara and Lotrich, 2012). It is known that markers of inflammation such as pro-inflammatory cytokines are elevated in MDD, BD depression and mania (Maes et al.,

1994, 1995, 1997a, 1997b; Dickerson et al., 2007; Drexhage et al., 2010; Rao et al., 2010) and inflammation may resolve in periods of euthymia (Brietzke et al., 2009). Further, studies

- 16 - looking at postmortem brain tissue (Gawryluk et al., 2011a, 2011b; Shelton et al., 2011) and

CSF (Lindqvist et al., 2009; Martinez et al., 2012) have found that markers of inflammation are increased in the brain.

1.6.1 Inflammatory Mechanisms to Depressive Symptoms

It was once thought that the brain was an immune-privileged organ, and while that is in part accurate, it is now known that the CNS is neither isolated nor passive in its interactions with the peripheral immune system (Carson et al., 2006). The brain has its own set of immune cells, including microglia – the resident CNS macrophage – and monitors the immune status of the body through various pathways (Carson et al., 2006). Immune cells are present in the meninges and choroid plexus, including macrophages and dendritic cells. These cells have toll-like receptors (TLRs) that bind to pathogen-associated molecules (Dantzer et al., 2008).

Macrophages and dendritic cells produce pro-inflammatory cytokines, which enter the brain through volume diffusion (Dantzer et al., 2008). Further, there exist leaky portions of the blood-brain barrier that correspond to some regions of the brain called circumventricular organs, as shown in Figure 1-2 (Dilger and Johnson, 2008). Circumventricular organs, characterized by small size, high permeability and fenestrated capillaries, allow for polypeptide hormones to leave the brain and permit substances to enter the brain that do not cross the blood-brain barrier (Ganong, 2000). Moreover, there are cytokine transporters in the blood-brain barrier, which allow access via the systemic circulation (Banks, 2009).

Lymphocytes are also capable of crossing the blood-brain barrier (Wilson et al., 2010;

Carson et al., 2006). Lastly, afferent nerve fibers can transduce immune-related signals from the gut and mouth, via the vagal and trigeminal nerves, respectively (Dantzer et al., 2008).

- 17 - Figure 1-2: Routes of communication between the immune system and brain.

Figure 1.2- There are various mechanisms by which the central nervous system (CNS) and peripheral immune system communicate. Pro-inflammatory cytokines such as IL-1β can gain access to the brain via Circumventricular organs, where the blood-brain-barrier (BBB) is “leaky”. There are also transporters for the cytokine IL-1β on the BBB. Two other modes of communication include vagal afferent fibers and monocyte infiltration into the CNS.

Adapted from: Dilger and Johnson, 2008.

Khandaker and colleagues (2014) aimed to test the causality of inflammation to increase future risk for depression. They performed a prospective, longitudinal study of 4500 individuals at the age of 9 and again at 18, and measured blood levels of IL-6 and CRP, and depression using two clinical mood scales (Khandaker et al., 2014). The authors found that higher levels of IL-6 in childhood were associated with an increased risk for the development of depression and psychosis in young adulthood (Khandaker et al., 2014). Similarly,

Valkanova et al. (2013) investigated whether inflammatory markers predicted an increased risk of subsequent depressive symptoms. The authors found a significant association

- 18 - between increased CRP and depressive symptoms and concluded that raised inflammatory markers had a small but significant association with the subsequent development of depressive symptoms (Valkanova et al., 2013).

Rodent models have allowed for the elucidation of direct mechanisms of action for inflammation in depressive mood states. Walker et al. (2013) assessed the role of NMDA receptor activation in LPS-induced depressive-like behavior in male mice. They found that a low dose of ketamine – an NMDA receptor antagonist – abrogated the development of

LPS-induced depressive-like behavior (Walker et al., 2013).

The ubiquitously expressed enzyme, indoleamine 2,3 dioxygenase (IDO), is known to be a critical molecular mediator for the development of the inflammation-based depressive phenotype (O’Connor et al., 2009b, 2009c). O’Connor et al. (2009) found the peripherally administered LPS in male mice activated IDO and culminated in a distinct depressive phenotype – in the forced swim test (FST) and tail suspension test (TST).

Administration of the anti-inflammatory compound minocycline, or IDO antagonist 1-MT, prevented the development of depressive-like behavior in this model (O’Connor et al.,

2009b, 2009c). Also, administration of L-kynurenine, a metabolite of tryptophan that is generated by IDO, also induced depressive-like behavior in male mice in this study

(O’Connor et al., 2009b, 2009c).

The catabolism of the essential amino acid tryptophan by IDO is the first, rate-limiting step in the production of kynurenine (KYN) – shown in Figure 1-3 (Mukherjee et al., 2018).

KYN is actively transported across the blood-brain barrier and is further catabolized into excitotoxic or neuroprotective molecules (Dantzer et al., 2008). KYN can be metabolized by perivascular macrophages, microglia, and astrocytes (Engin, 2015). Microglia metabolize

KYN into excitotoxic compounds quinolinic acid (QUIN) and 3-hydroxy kynurenine, which generates free radicals that cause oxidative stress, and lipid peroxidation (Lugo-Huitrón et

- 19 - al., 2013) Astrocytes produce kynurenic acid, an NMDA receptor antagonist (Engin, 2015).

Figure 1-3: The Metabolism of Tryptophan into Excitotoxic Molecules

TPHs Tryptophan Serotonin

TNF-α, INF-γ IDO TDO

Kynurenine

KMO KAT

3-hydroxykynurenine Kynureninc acid

KYNU

Quinolinic acid

Figure 1.3- The kynurenine pathway. INF-γ, interferon-gamma; TNF-α, tumor necrosis factor-alpha; TPH IDO, indolamine-2,3-dioxygenase; KAT, kynurenine aminotransferase;

KMO, kynurenine 3-monooxygenase; KYNU, kynureninase; TDO, tryptophan-2,3- dioxygenase; TPH, tryptophan hydroxylase. Adapted from: Mukherjee et al., 2018.

The proposed depressive-inducing mechanism of inflammation arises from the action of excitotoxic compounds produced by microglia, especially QUIN (Dantzer et al., 2008). QUIN is considered an endogenous neurotoxin and is expressed in the brain during an acute depressive episode (Felger and Lotrich, 2013b). QUIN is thought to contribute directly to excitotoxicity in the brain (Felger and Lotrich, 2013b). Excitotoxicity is the pathophysiological process by which neurons are damaged or killed by the over-activation (excessive

- 20 - stimulation) by neurotransmitters such as glutamate (Felger and Lotrich, 2013b). NMDA and

AMPA glutamate receptors are over-activated by glutamate and excitotoxins (like QUIN), which bind these receptors (Felger and Lotrich, 2013b). Excitotoxicity allows for high levels of calcium to enter the cell, which subsequently erroneously activates enzymes (Dantzer et al., 2008).

QUIN may contribute to excitotoxicity in a few ways; firstly, it is an NMDA receptor agonist (Felger and Lotrich, 2013b). QUIN also enhances the release of synaptosomal glutamate because of the inhibition of glutamate uptake into astrocytes that leads to overstimulation of receptors (Felger and Lotrich, 2013b). QUIN can also decrease the action of antioxidant enzymes, promote reactive oxygen species production, and generate lipid peroxidation (Felger and Lotrich, 2013b). QUIN may also inhibit mitochondrial complexes leading to an energy deficit, activating caspases and releasing cytochrome c (Felger and

Lotrich, 2013b). All of these factors induce cytoskeleton destabilization, DNA damage, and cell death (Felger and Lotrich, 2013b).

Oxidative damage results from the over-production of reactive oxygen species that overwhelms the antioxidant capacity (Felger and Lotrich, 2013b). Brain cells are more vulnerable than other cells to oxidative damage because the brain consumes about 20% of the body’s total oxygen, although it constitutes only 2% of total body weight (Felger and

Lotrich, 2013b). This data is in line with clinical evidence of increased oxidative stress in the post-mortem brain tissue of people with BD and MDD (Shelton et al., 2010).

Post-mortem studies show increased markers of oxidative stress and inflammation in the brains of people with BD (Kim et al., 2010a, 2010b; Rao et al., 2010). Such observations have contributed to the hypothesis of neuroprogression as part of the underlying pathophysiology of BD (Berk et al., 2011). Further, Che et al. (2010) analyzed oxidative damage to nucleic acids in the CA1, CA3 and dentate gyrus regions of postmortem

- 21 - hippocampal tissues, of patients with BD and MDD. They probed (with immunohistochemistry) with antibodies against 8-hydroxyguanosine in RNA and 8-hydroxy-

2’-deoxyguanosine in DNA. They found elevated oxidative damage to nucleic acids in all hippocampal areas assessed in both groups (Che et al., 2010). The authors concluded that the amount of oxidative damage might be associated with the severity of symptoms (Che et al., 2010). Increased lipid peroxidation and decreased activity of the antioxidant defense enzymes SOD and catalase were found in the plasma of patients with BD. Their study and others suggest that RNA oxidative damage may result from overproduction of ROS caused by mitochondrial dysfunction or abnormal function of monoamine oxidase (MAO) (Che et al., 2010).

The expression of the antioxidant enzyme glutathione s-transferase A4 and M3 subtypes were reduced in postmortem brain tissues from patients with BD (Gawryluk et al.

2011a, 2011b). Further, many studies have shown that mood-stabilizing drugs and antidepressants increase anti-oxidant production and decrease oxidative damage. In twin studies, the bipolar twin has more oxidative stress than the healthy twin, which was normalized after mood stabilizing treatment. Glutathione (GSH) is a primary free radical scavenger in the brain. Diminished GSH levels elevate cellular vulnerability toward oxidative stress. The authors looked at postmortem prefrontal cortex of patients with MDD and BD.

They also looked at protein levels of GCL, GSH reductase, and GSH peroxidase. They found diminished GSH (total, reduced and oxidized) in BD and MDD compared to healthy controls. They found reduced glutathione peroxidase in the BD group compared to the healthy control group (Gawryluk et al., 2011a, 2011b).

Patki et al. (2013) has shown that social defeat-stress paradigm causes anxiety-like behavior and memory impairment in rats. They also found reduced protein levels of GLO-1,

GSR-1, CAMKIV, CREB, and BDNF in the hippocampus. Zhao et al. (2017) found that

- 22 - acetaminophen attenuates LPS-induced cognitive impairment through antioxidant activity.

Allam et al. (2013) found that oxidative stress induced depressive-like behavior, and memory impairment. Grape powder increased levels of glyoxalase-1, glutathione reductase-

1 (GSR-1), calcium response element binding protein (CREB), brain-derived neurotrophic factor (BDNF) and calcium/calmodulin (CAMKIV), compared to animals given a compound that creates oxidative stress and anxiety-like behavior (Allam et al., 2013). The grape powder also attenuated anxiety-like behavior (Allam et al., 2013).

1.6.2 Inflammation and Gender/Sex Interaction

There are known gender disparities in immune activation. Male sex is an independent risk factor for severe infections in surgical patients (Offner et al., 1999), and women have better survival rates during sepsis (Schröder et al., 1998). In a rodent model of trauma-hemorrhage and sepsis, it was found that females in proestrus have significantly higher survival rates than males (Choudhry et al., 2005). This phenomenon may be related to the fact that the sex hormone 17-β estradiol (E2) inhibits the excess inflammatory response. E2 mediates broad anti-inflammatory effects in the body (Fish et al., 2008) and can inhibit the production of pro-inflammatory cytokines IL-1β, IL-6, and TNF-α from monocytes and macrophages in vitro (Kramer et al., 2004). Also, ovarian fluctuations of E2 can influence pro-inflammatory cytokine concentration (Bouman et al., 2005).

In line with this, there is evidence in the literature that the association between inflammation and mood varies by gender and sex. A study by Moieni et al. (2015b) aimed to determine whether women show greater peripheral levels pro-inflammatory cytokines, depressed mood, and social disconnection in response to an inflammatory challenge. The authors recruited 115 healthy participants (69 female) to complete this clinical trial in which participants were randomly assigned to receive either an infusion of endotoxin or placebo

- 23 - (Moieni et al., 2015b). The authors assessed IL-6, TNF, depressed mood and feelings of social disconnection hourly (Moieni et al., 2015b). Results showed that endotoxin caused an increase in depressed mood, feelings of disconnection and increased cytokines (Moieni et al., 2015b). Females had more significant depressed mood and feelings of social disconnection compared to males (Moieni et al., 2015b). Also, cytokines were correlated with feelings of social disconnection in females but not in males (Moieni et al., 2015b).

Kendler et al. (2014) sought to clarify the nature of sex differences in the etiological pathways of depressive episodes. They recruited 1,057 pairs of opposite-sex dizygotic twins. They then conducted two waves of interviews with these twin pairs wherein they performed retrospective and prospective assessments of risk factors and the occurrence of past-year MDD (Kendler et al., 2014). They found that some of the risk factors for MDD differed across sexes; personality and failures in interpersonal relationships played a stronger etiologic role in depression for women than for men (Kendler et al., 2014).

Externalizing psychopathology, prior depression, and acute stressors are more critical in the etiologic pathway to MDD for men (Kendler et al., 2014).

Eisenberger et al. (2009) examined the effect of pro-inflammatory cytokine activation on the neural correlates of socially painful experience and associated depressed mood. Men and women received either endotoxin or saline via intravenous infusion (Eisenberger et al.,

2009). They assessed IL-6 with hourly blood draws and depressed mood with hourly self- reports. Two hours after injection, participants went through a neuroimaging session in which they were socially excluded in an online game (Eisenberger et al., 2009). They found that people exposed to endotoxin showed increased IL-6 and depressed mood. They also found sex differences in the relationships between IL-6 increases and neural responses to exclusion among subjects exposed to endotoxin: among females exposed to endotoxin, increases in IL-6 were associated with increases in social pain-related neural activity that

- 24 - mediated the relationship between IL-6 increases and depressed mood (Eisenberger et al.,

2009).

Valentine et al. (2009) measured CRP and depressive symptoms in 127 elderly, community-dwelling men and women. They found that women reported 63% greater fatigue than men (Valentine et al., 2009). Women report more fatigue than men, which was independently associated with inflammation, depressive symptoms, physical activity and adiposity, whereas in men the only independent predictor was depression (Valentine et al.,

2009).

Maes et al. (2001) carried out a study to examine whether MDD induced sensitization with increased inflammatory responses to the stress of childbirth. Serum concentrations of cytokines were determined in women with (16) and without (50) a history of MDD. Blood was collected 3-6 days before childbirth and again 1-3 days after delivery. On each occasion, the women completed a depression mood rating scale. They found that blood cytokines are amplified after birth in women with a history of MDD, suggesting that MDD is accompanied by sensitization of the inflammatory system.

1.7 Animal Models

Animal models utilizing inflammatory instigators of depressive states have revealed sex- based differences as well. Pitychoutis et al. (2009) sought to identify whether “sickness behavior” in rodents is characterized by sexual dimorphism by focusing on both neurochemical and behavioral responses. The authors looked at the serotonergic and dopaminergic activity of various brain regions implicated in the pathophysiology of affective disorders – hypothalamus, hippocampus, prefrontal cortex, amygdala, and striatum – in response to lipopolysaccharide (LPS) challenge, in male and female rats (Pitychoutis et al.,

2009). They found that 2 hours post-LPS administration, the serotonergic function being

- 25 - enhanced in all brain regions in females (Pitychoutis et al., 2009). Females also had mild alterations of dopaminergic activity in the striatum and an enhanced stress response to LPS compared to males, via a measure of corticosterone (Pitychoutis et al., 2009). Males had no observed changes to dopaminergic activity after LPS. LPS also altered behaviors in a sexually dichotomous way in the FST, hot plate test and open field test (Pitychoutis et al.,

2009). The authors found that LPS-treated females coped better in the FST procedure, as evidenced by an increase in swimming duration (Pitychoutis et al., 2009). Males and females were similar on traditional measures of sickness, such as sucrose consumption and social exploration (Pitychoutis et al., 2009).

1.8 The Antioxidant Zinc

Zinc is an essential biometal; we know that over 300 enzymes require zinc for their activation or stability (Prasad, 2014). Clinical manifestations of dietary zinc deficiency include immune dysfunction, increased oxidative stress, and increased generation of inflammatory cytokines

(Prasad, 2014). Figure 1-4 depicts a few of the ways in which the micronutrient zinc acts as an antioxidant. The antioxidant effects of zinc include the induction of metallothionein – a zinc-binding protein – that is rich in cysteine residues and is an excellent scavenger of hydroxyl radicals and toxic metals (Marreiro et al., 2017; Prasad, 2014). Further, zinc is an inhibitor of NADPH oxidase, a plasma membrane-associated enzyme that catalyzes the production of superoxide from oxygen, using NADPH as an electron donor (Prasad, 2014).

The inhibition of NADPH oxidase results in the decreased generation of reactive oxygen species (Prasad, 2014). Zinc is also a cofactor for Cu/Zn superoxide dismutase (SOD1), an enzyme that catalyzes the breakdown of superoxides to hydrogen peroxide (Prasad, 2014).

- 26 - Figure 1-4: Zinc as an antioxidant.

Figure 1.4- The biometal zinc has a wide array of functions in the human body and is found in every cell, in every organ. Zinc has an important role in maintaining cellular oxidative homeostasis and plays a part as an antioxidant by associating with antioxidant proteins such as metallothionein (MT), superoxide dismutase-1 (SOD1). Zinc can also inhibit the action of certain transporters and enzymes that generate oxidative radicals, like the enzyme NADPH oxidase, which produces superoxides, and the NMDA receptor. Adapted from: Marreiro et al., 2017.

Zinc is also a constitutive co-factor of innate immune system activation. The acute phase immune response is triggered by different stimuli including trauma, infection, stress, neoplasia (tumor), and inflammation (Cray et al., 2009) – during which zinc is sequestered from the peripheral blood and taken into the liver (Liuzzi et al., 2005). Zinc sequestration likely occurs for a few reasons, including increasing zinc’s availability for the synthesis of

- 27 - acute phase proteins, the regulation of gluconeogenesis, and control of microbial growth and reactive oxygen species (e.g., nitric oxide and hydroxyl radicals) (Liuzzi et al., 2005). It is known that zinc transporters are responsible for the increase in zinc sequestration, in particular, the zinc importer Zip14; Zip14 is stimulated by the pro-inflammatory cytokine IL-

6 (Liuzzi et al., 2005). In this way, in response to infection or fever, hypozincemia occurs

(Liuzzi et al., 2005). Further, during the acute phase immune response, acute phase proteins are released from the liver, including C-reactive protein, named for its reactivity to pneumococcal C polysaccharide (Cray et al., 2009), and many other proteins essential for pathogen recognition and elimination, as well as coagulation, and propagation of the inflammatory response (Cray et al., 2009). Also, the acute phase response is marked by diminished peripheral albumin and transferrin protein levels, which bind zinc and iron, respectively (Cray et al., 2009). In this way, peripheral zinc concentration may act as a marker of inflammatory activation, which makes it a potentially useful biomarker in pathological mood states, which commonly have heightened inflammatory profiles.

In fact, research has consistently shown that peripheral zinc concentrations are significantly diminished during depressive states, and this diminishment is accompanied by a concomitant increase in pro-inflammatory markers (Maes et al., 1994a; Mcloughlin and

Hodge 1990). Stanley and Wakwe (2002) reported that zinc was reduced in participants with BD compared to healthy controls. Similarly, Nourmohammadi et al. (2007) reported hypozincemia in BD. Conversely, González-Estecha et al. (2011) found zinc to be increased in mania in a cohort of men and women with BD. Literature also indicates that peripheral zinc concentrations are negatively correlated with severity of depressive symptoms in men and women (Amani et al., 2010; Siwek et al., 2010). In this way, zinc could be a potential biomarker of mood symptom severity in BD and MDD.

- 28 - Zinc has also been shown to have antidepressant properties in human and rodent studies. Dietary zinc deficiency can precipitate the onset of a depressive phenotype, and zinc supplementation has been shown to reverse depressive-like behaviors (Szewczyk et al., 2002, 2009; Nowak et al., 2003a, 2003c; Cieslik et al., 2007; Tassabehji et al., 2008;

Mlyniec and Nowak, 2012). Zinc administration is known to increase mobility time in the forced swim test, an indication of increased resilience (Kroczka et al., 2000, 2001). Further, zinc augments imipramine (a tricyclic antidepressant) in the chronic unpredictable stress model in rats, and the action of various antidepressants (fluoxetine, paroxetine, imipramine, desipramine, and bupropion) in the tail suspension test (Cieslik et al., 2007; Cunha et al.,

2008). Zinc has also been shown to ameliorate psychotic symptoms in rodents (Joshi et al.,

2012). In the amphetamine-induced locomotor activity test, injected zinc produced a reduction in the total movement time, mean velocity and stereotypic movements, i.e., repetitive, non-functional movements (Joshi et al., 2012). In a double blind, randomized clinical trial, forty-four patients with MDD were randomly assigned to groups receiving antidepressant treatment plus (1) zinc supplementation or (2) placebo (Ranjbar et al., 2013).

Patients in the zinc group received daily supplementation adjunct to a selective serotonin reuptake inhibitor (SSRI) (Ranjbar et al., 2013). Their results indicated that zinc supplementation together with SSRI treatment improves depressive symptoms more efficiently than in patients given SSRI plus placebo (Ranjbar et al., 2013).

The antidepressant action of zinc has been linked to a neuroprotective mechanism.

One study showed that zinc chloride administration reversed a depressive-phenotype in the

‘malathion’ model of depression – malathion is an acetylcholinesterase inhibitor and potent neurotoxin (Brocardo et al., 2007). Malathion resulted in depressive-like behavior in the

FST, increased lipid peroxidation and reduced glutathione peroxidase activity in the cerebral cortex, reduced glutathione reductase activity in the hippocampus, and changed in the

- 29 - structure of chromatin in the dentate gyrus (Brocardo et al., 2007). All of these effects were attenuated by zinc chloride (Brocardo et al., 2007). Another study indicated that zinc’s antidepressant effects were related to the up-regulation of neuroprotective agents, including glutathione and BDNF (Franco et al., 2008). They found that chronic (30 days) zinc treatment was associated with an antidepressant effect in the FST, which was accompanied by a significant increase in total glutathione levels in the hippocampus and cerebral cortex

(Franco et al., 2008). They also found that chronic zinc treatment produced an increase in

ERK phosphorylation and BDNF expression in the cerebral cortex (Franco et al., 2008).

Despite accumulating evidence, the antidepressant and neuroprotective mechanisms of zinc are still being elucidated. However, research has shown the importance of the glutamatergic system in the antidepressant action of zinc. Some studies have also examined zinc in human post-mortem brain tissue. Two such reports, by Nowak et al. (2003) and Sowa-Kucma et al. (2013), hypothesized that an alteration in zinc homeostasis might exist in the brain tissue of suicide victims. These studies found that zinc had a decreased potency to inhibit NMDA receptors in the hippocampus, but not the prefrontal cortex in suicide victims (Nowak et al., 2003b; Sowa-Kucma et al., 2013). Szewczyk et al. (2010) showed that the antidepressant effect of zinc is reliant on zinc’s interaction with NMDA and

AMPA glutamate receptors. This study showed that the antidepressant activity of zinc in the

FST was abrogated by NMDA administration (Szewczyk et al., 2010). Further, zinc administration potentiated the release of glutamate and aspartate in the prefrontal cortex and hippocampus (Szewczyk et al., 2010). Other studies have examined the effect of zinc administration directly on the brain using microdialysis. Takeda et al. (2003) showed that the glutamate concentration in the CA1 region of the hippocampus was significantly decreased by zinc chloride. Conversely, GABA concentration was enhanced with the administration of zinc chloride in the perfusate (Takeda et al., 2003).

- 30 - It has become increasingly apparent in recent decades that zinc plays a dynamic role in the physiology of brain function (Sensi et al., 2011). Along with glutamate, zinc ions interact with many receptors and transporters in the brain, including those for serotonin and dopamine (Richfield et al., 1993; Schetz and Sibley, 1997; Schetz et al., 1999; Hubbard et al., 2000). Zinc ions are vesicularized at synaptic terminals exclusively in the forebrain – the cerebral cortex, basal ganglia, hippocampus, amygdala, olfactory bulb, thalamus, and hypothalamus – which is intriguing due to the anatomical correlation with mood disorders.

Zinc’s function as an antioxidant, a neuromodulator, and a putative cofactor during peripheral inflammation, indicates zinc is a relevant target of exploration in the field of mood disorders.

1.9 Summary and Specific Aims

Research in humans and rodents has implicated immune/inflammatory activation in the onset and progression of mood disorders – Bipolar Disorder (BD) and Major Depressive

Disorder (MDD). There are various lines of evidence that indicate immune/inflammatory dysregulation in mood disorders. For example, some pro-inflammatory cytokines are elevated in peripheral blood samples and cerebrospinal fluid in patients with MDD, BD- depression, and mania. Postmortem studies in human brain tissue have found that markers of inflammation are also increased. Prospective studies have revealed a potentially causative role of immune/inflammatory activation and depressive symptoms, so that higher levels of IL-6 at baseline are associated with an increased risk for the development of depression and psychosis a decade later.

Inflammation may induce oxidative stress in the brain via the production of kynurenine. Kynurenine is increased during immune/inflammatory activation and is actively transported across the blood-brain barrier. Kynurenine is further catabolized into excitotoxic

- 31 - molecules like quinolinic acid, which generates free radicals that cause oxidative stress.

Oxidative damage to brain tissue may result from the over-production of reactive oxygen species, which overwhelm the brain’s innate antioxidant capacity. Research has shown increased oxidative damage to nucleic acids in postmortem hippocampal tissue in people with BD and MDD. Further, some research has shown decreased antioxidant levels in BD and MDD in the postmortem brain. Studies have found diminished glutathione levels (total, reduced and oxidized) in BD and MDD compared to healthy controls. Also, peripheral zinc concentrations have been shown to be significantly diminished in depressive states, and this diminishment is accompanied by a concomitant increase in pro-inflammatory markers.

There are known gender disparities in immune activation; male sex is an independent risk factor for severe infections in surgical patients, and women have better survival rates during sepsis. This phenomenon may be related to the fact that the sex hormone 17-β estradiol (E2) inhibits the excess inflammatory response. E2 mediates broad anti-inflammatory effects in the body and can inhibit the production of pro-inflammatory cytokines IL-1β, IL-6, and TNF-α. In contrast, it is known that in general women have more robust cell-mediated and humoral immune responses to antigenic challenges compared with males. Further, there is evidence that females have increased feelings of depression after inflammatory challenge; in a recent study, women had higher self-reported feelings of depression and social isolation after endotoxin injection compared to men. Further, the pro- inflammatory cytokine IL-6 was found to be associated with the sensation of ‘social pain’ in women but not in men.

Despite the burgeoning body of literature related to inflammation and gender/sex in mood disorders, there are still many unanswered questions. For example, there are currently no known diagnostic, theranostic or prognostic biomarkers for BD. Therefore, biomarker identification and development are greatly needed in this field. We proposed that

- 32 - immune/inflammatory and antioxidant markers may be used prognostically to either predict an oncoming episode or to predict treatment response. Further, we posit that gender will be an important moderator between biomarkers and mood. Based on the evidence outlined above, there exists ample evidence to hypothesize that men and women differ in their inflammatory/immune responses in BD. Our focus on gender is also important regarding personalized treatments (theranostics), a field of study that has recently seen a surge of interest. Personalized treatment in BD is increasingly recognized as important due to the complex nature of this mental illness.

Another understudied area of research pertains to the sequelae from chronic peripheral inflammation in the brain. One of the most common observations in the study of

MDD and BD is that the hippocampus atrophies over time. Atrophy – particularly that of the hippocampus – has been hypothesized to at least partially underlie cognitive deficits in mood disorders. We wish to address gaps in knowledge related to the effect of peripheral inflammation on oxidative stress and apoptosis in the hippocampus, and the sex-based differences that may exist therein.

Aim 1. To examine the potential usefulness of inflammatory molecules and antioxidants as biomarkers in BD, we examined peripheral plasma samples from symptomatic human participants with BD. The biomarkers of interest were (1) neopterin – a marker of the acute phase immune response, and (2) zinc – an essential biometal and antioxidant in the human body. We hypothesized that neopterin would be increased in symptomatic participants BD and zinc decreased. We proposed that gender (men and women) moderates of the relationship between inflammation and mood in BD.

Aim 2. To examine the effect of the peripheral endotoxin exposure on depressive-like behaviors and hippocampal atrophy, we utilized male and female mice of the strain

C57BL/6. We hypothesized that a peripheral injection of lipopolysaccharide (LPS) – a

- 33 - component of the cell wall of gram-negative bacteria – would induce inflammation and oxidative stress, which would in turn precipitate hippocampal atrophy and reduced neural plasticity. Further, we postulated that males and females would respond differently to an injection of LPS acutely (24 hours after injection) and longitudinally (28 days after injection).

We hypothesized that there would be a sex-based dichotomy regarding the levels of oxidative stress and apoptosis in the brain after LPS injection as well as in behavioral responses to LPS.

- 34 - 1.10 References Akiskal HS, Hantouche EG, Bourgeois ML, Azorin JM, Sechter D, Allilaire JF, et al. (1998). Gender, temperament, and the clinical picture in dysphoric mixed mania: findings from a French national study (EPIMAN). J Affect Disord 50(2-3): 175-186. Alberts B, Academies N (2002). From the National Academies. Cell Biol Educ 1(4): 109- 110. Allam F, Dao AT, Chugh G, Bohat R, Jafri F, Patki G, et al. (2013). Grape powder supplementation prevents oxidative stress-induced anxiety-like behavior, memory impairment, and high blood pressure in rats. J Nutr 143(6): 835-842. Altshuler LL, Kupka RW, Hellemann G, Frye MA, Sugar CA, McElroy SL, et al. (2010). Gender and depressive symptoms in 711 patients with bipolar disorder evaluated prospectively in the Stanley Foundation bipolar treatment outcome network. Am J Psychiatry 167(6): 708-715. Amani R, Saeidi S, Nazari Z, Nematpour S (2010). Correlation between dietary zinc intakes and its serum levels with depression scales in young female students. Biol Trace Elem Res 137(2): 150-158. American Psychiatric Association (2013). Diagnostic and Statistical Manual of Mental Disorders, 5th edn.: Washington, DC. Angst J, Gamma A (2002a). A new bipolar spectrum concept: a brief review. Bipolar Disord 4 Suppl 1: 11-14. Angst J, Gamma A, Lewinsohn P (2002b). The evolving epidemiology of bipolar disorder. World Psychiatry 1(3): 146-14 Azorin JM, Belzeaux R, Fakra E, Kaladjian A, Hantouche E, Lancrenon S, et al. (2014b). Gender differences in a cohort of major depressive patients: further evidence for the male depression syndrome hypothesis. J Affect Disord 167: 85-92. Baldassano CF, Marangell LB, Gyulai L, Ghaemi SN, Joffe H, Kim DR, et al. (2005). Gender differences in bipolar disorder: retrospective data from the first 500 STEP- BD participants. Bipolar Disord 7(5): 465-470. Berk M, Kapczinski F, Andreazza AC, Dean OM, Giorlando F, Maes M, et al. (2011). Pathways underlying neuroprogression in bipolar disorder: focus on inflammation, oxidative stress and neurotrophic factors. Neurosci Biobehav Rev 35(3): 804-817. Bouman A, Heineman MJ, Faas MM (2005). Sex hormones and the immune response in humans. Hum Reprod Update 11(4): 411-423. Bowden CL (2005a). A different depression: clinical distinctions between bipolar and unipolar depression. J Affect Disord 84(2-3): 117-125. Brietzke E, Kauer-Sant'Anna M, Teixeira AL, Kapczinski F (2009). Abnormalities in serum chemokine levels in euthymic patients with bipolar disorder. Brain Behav Immun 23(8): 1079-1082. Brocardo PS, Assini F, Franco JL, Pandolfo P, Müller YM, Takahashi RN, et al. (2007). Zinc attenuates malathion-induced depressant-like behavior and confers neuroprotection in the rat brain. Toxicol Sci 97(1): 140-148. Calabrese JR (1999). Lamotrigine and the treatment of bipolar disorder. Introduction. Eur Neuropsychopharmacol 9 Suppl 4: S107-108. Carson MJ, Doose JM, Melchior B, Schmid CD, Ploix CC (2006). CNS immune privilege: hiding in plain sight. Immunol Rev 213: 48-65. CARTER HS (1954). Robert Burton and the anatomy of melancholy. Glasgow Med J 35(8): 202-212. Che Y, Wang JF, Shao L, Young T (2010). Oxidative damage to RNA but not DNA in the hippocampus of patients with major mental illness. J Psychiatry Neurosci 35(5):

- 35 - 296-302. Chiu WC, Su YP, Su KP, Chen PC (2017). Recurrence of depressive disorders after interferon-induced depression. Transl Psychiatry 7(2): e1026. Choudhry MA, Schwacha MG, Hubbard WJ, Kerby JD, Rue LW, Bland KI, et al. (2005). Gender differences in acute response to trauma-hemorrhage. Shock 24 Suppl 1: 101-106. Cieślik K, Klenk-Majewska B, Danilczuk Z, Wróbel A, Łupina T, Ossowska G (2007). Influence of zinc supplementation on imipramine effect in a chronic unpredictable stress (CUS) model in rats. Pharmacol Rep 59(1): 46-52. Coryell W, Solomon D, Turvey C, Keller M, Leon AC, Endicott J, et al. (2003). The long- term course of rapid-cycling bipolar disorder. Arch Gen Psychiatry 60(9): 914-920. Cosmacini G (1997a). [The heart of philosophers]. Cardiologia 42(6): 647-649. Cosmacini G (1997b). [The philosophy of blood and lymph circulation]. Cardiologia 42(12): 1289-1291. Cunha MP, Machado DG, Bettio LE, Capra JC, Rodrigues AL (2008). Interaction of zinc with antidepressants in the tail suspension test. Prog Neuropsychopharmacol Biol Psychiatry 32(8): 1913-1920. Curtis AL, Bethea T, Valentino RJ (2006). Sexually dimorphic responses of the brain norepinephrine system to stress and corticotropin-releasing factor. Neuropsychopharmacology 31(3): 544-554. Dantzer R, O'Connor JC, Freund GG, Johnson RW, Kelley KW (2008). From inflammation to sickness and depression: when the immune system subjugates the brain. Nat Rev Neurosci 9(1): 46-56. Davis GL, Beck JR, Farrell G, Poynard T (1998a). Prolonged treatment with interferon in patients with histologically mild chronic hepatitis C: a decision analysis. J Viral Hepat 5(5): 313-321. Delferrière O, Gobin R, Harrault F, Nyckees S, Tuske O (2014a). Improvement of extraction system geometry with suppression of possible Penning discharge ignition. Rev Sci Instrum 85(2): 02A939. Delferrière O, Gobin R, Harrault F, Nyckees S, Tuske O (2014b). Improvement of extraction system geometry with suppression of possible Penning discharge ignition. Rev Sci Instrum 85(2): 02A939. Dickerson F, Stallings C, Origoni A, Boronow J, Yolken R (2007). Elevated serum levels of C-reactive protein are associated with mania symptoms in outpatients with bipolar disorder. Prog Neuropsychopharmacol Biol Psychiatry 31(4): 952-955. Dilger RN, Johnson RW (2008). Aging, microglial cell priming, and the discordant central inflammatory response to signals from the peripheral immune system. J Leukoc Biol 84(4): 932-939. Dimopoulos N, Piperi C, Psarra V, Lea RW, Kalofoutis A (2008). Increased plasma levels of 8-iso-PGF2alpha and IL-6 in an elderly population with depression. Psychiatry Res 161(1): 59-66. Drexhage RC, van der Heul-Nieuwenhuijsen L, Padmos RC, van Beveren N, Cohen D, Versnel MA, et al. (2010). Inflammatory gene expression in monocytes of patients with schizophrenia: overlap and difference with bipolar disorder. A study in naturalistically treated patients. Int J Neuropsychopharmacol 13(10): 1369-1381. Eisenberger NI, Inagaki TK, Rameson LT, Mashal NM, Irwin MR (2009). An fMRI study of cytokine-induced depressed mood and social pain: the role of sex differences. Neuroimage 47(3): 881-890. Engin AB, Karahalil B, Karakaya AE, Engin A (2015). Helicobacter pylori and serum

- 36 - kynurenine-tryptophan ratio in patients with colorectal cancer. World J Gastroenterol 21(12): 3636-3643. Essau CA, Lewinsohn PM, Seeley JR, Sasagawa S (2010). Gender differences in the developmental course of depression. J Affect Disord 127(1-3): 185-190. Felger JC, Lotrich FE (2013b). Inflammatory cytokines in depression: neurobiological mechanisms and therapeutic implications. Neuroscience 246: 199-229. Neuropsychopharmacology 38(11): 2179-2187. Ferrari AJ, Charlson FJ, Norman RE, Patten SB, Freedman G, Murray CJ, et al. (2013). Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS Med 10(11): e1001547. Ferrari AJ, Stockings E, Khoo JP, Erskine HE, Degenhardt L, Vos T, et al. (2016). The prevalence and burden of bipolar disorder: findings from the Global Burden of Disease Study 2013. Bipolar Disord 18(5): 440-450. Fish EN (2008). The X-files in immunity: sex-based differences predispose immune responses. Nat Rev Immunol 8(9): 737-744. Franco JL, Posser T, Brocardo PS, Trevisan R, Uliano-Silva M, Gabilan NH, et al. (2008). Involvement of glutathione, ERK1/2 phosphorylation and BDNF expression in the antidepressant-like effect of zinc in rats. Behav Brain Res 188(2): 316-323. Ganong WF (2000). Circumventricular organs: definition and role in the regulation of endocrine and autonomic function. Clin Exp Pharmacol Physiol 27(5-6): 422-427. Gartlehner G, Stepper K (2012). Julius Wagner-Jauregg: pyrotherapy, simultanmethode, and 'racial hygiene'. J R Soc Med 105(8): 357-359. Gater R, Tansella M, Korten A, Tiemens BG, Mavreas VG, Olatawura MO (1998). Sex differences in the prevalence and detection of depressive and anxiety disorders in general health care settings: report from the World Health Organization Collaborative Study on Psychological Problems in General Health Care. Arch Gen Psychiatry 55(5): 405-413. Gawryluk J, Wang JF, Andreazza AC, Shao L, Yatham LN, Young LT (2011a). Prefrontal cortex glutathione S-transferase levels in patients with bipolar disorder, major depression and schizophrenia. Int J Neuropsychopharmacol 14(8): 1069-1074. Gawryluk JW, Wang JF, Andreazza AC, Shao L, Young LT (2011b). Decreased levels of glutathione, the major brain antioxidant, in post-mortem prefrontal cortex from patients with psychiatric disorders. Int J Neuropsychopharmacol 14(1): 123-130. Gonzalez-Estecha M, Trasobares EM, Tajima K, Cano S, Fernandez C, Lopez JL, et al. (2011). Trace elements in bipolar disorder. Journal of Trace Elements in Medicine and Biology 25: S78-S83. Gruber J, Harvey AG, Wang PW, Brooks JO, Thase ME, Sachs GS, et al. (2009). Sleep functioning in relation to mood, function, and quality of life at entry to the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD). J Affect Disord 114(1-3): 41-49. Gruber J, Miklowitz DJ, Harvey AG, Frank E, Kupfer D, Thase ME, et al. (2011). Sleep matters: sleep functioning and course of illness in bipolar disorder. J Affect Disord 134(1-3): 416-420. Hendrick V, Altshuler LL, Gitlin MJ, Delrahim S, Hammen C (2000a). Gender and bipolar illness. J Clin Psychiatry 61(5): 393-396; quiz 397. Hendrick V, Altshuler LL, Gitlin MJ, Delrahim S, Hammen C (2000b). Gender and bipolar illness. J Clin Psychiatry 61(5): 393-396; quiz 397. Hippocrates (2002a). On the articulations. The genuine works of Hippocrates. Clin Orthop Relat Res(400): 19-25.

- 37 - Hippocrates (2002b). On the articulations. The genuine works of Hippocrates. Clin Orthop Relat Res(400): 19-25. Hirschfeld RM, Vornik LA (2004). Recognition and diagnosis of bipolar disorder. J Clin Psychiatry 65 Suppl 15: 5-9. Hoofnagle JH, Seeff LB (2006). Peginterferon and ribavirin for chronic hepatitis C. N Engl J Med 355(23): 2444-2451. Horwitz Allan V., JCW, and Lorenzo Lorenzo-Luaces (2016). History of Depression. In: Strunk RJDaDR (ed). The Oxford Handbook of Mood Disorders. Oxford University Press. Hubbard PC, Lummis SC (2000). Zn(2+) enhancement of the recombinant 5-HT(3) receptor is modulated by divalent cations. Eur J Pharmacol 394(2-3): 189-197. Joshi M, Akhtar M, Najmi AK, Khuroo AH, Goswami D (2012). Effect of zinc in animal models of anxiety, depression and psychosis. Hum Exp Toxicol 31(12): 1237-1243. Judd LL, Akiskal HS (2003). Depressive episodes and symptoms dominate the longitudinal course of bipolar disorder. Curr Psychiatry Rep 5(6): 417-418. Judd LL, Schettler PJ, Akiskal HS, Maser J, Coryell W, Solomon D, et al. (2003). Long- term symptomatic status of bipolar I vs. bipolar II disorders. Int J Neuropsychopharmacol 6(2): 127-137. Karamanou M, Liappas I, Antoniou C, Androutsos G, Lykouras E (2013). Julius Wagner- Jauregg (1857-1940): Introducing fever therapy in the treatment of neurosyphilis. Psychiatriki 24(3): 208-212. Kawa I, Carter JD, Joyce PR, Doughty CJ, Frampton CM, Wells JE, et al. (2005). Gender differences in bipolar disorder: age of onset, course, comorbidity, and symptom presentation. Bipolar Disord 7(2): 119-125. Kemp DE, Gao K, Chan PK, Ganocy SJ, Findling RL, Calabrese JR (2010). Medical comorbidity in bipolar disorder: relationship between illnesses of the endocrine/metabolic system and treatment outcome. Bipolar Disord 12(4): 404- 413. Kendler KS, Gardner CO (2014). Sex differences in the pathways to major depression: a study of opposite-sex twin pairs. Am J Psychiatry 171(4): 426-435. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE (2005a). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 62(6): 593-602. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE (2005b). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 62(6): 593-602. Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE (2005c). Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 62(6): 617-627. Kessler RC, McGonagle KA, Nelson CB, Hughes M, Swartz M, Blazer DG (1994). Sex and depression in the National Comorbidity Survey. II: Cohort effects. J Affect Disord 30(1): 15-26. Khandaker GM, Dantzer R, Jones PB (2017). Immunopsychiatry: important facts. Psychol Med: 1-9. Khandaker GM, Pearson RM, Zammit S, Lewis G, Jones PB (2014). Association of serum interleukin 6 and C-reactive protein in childhood with depression and psychosis in young adult life: a population-based longitudinal study. JAMA Psychiatry 71(10): 1121-1128. Kilbourne AM, Cornelius JR, Han X, Pincus HA, Shad M, Salloum I, et al. (2004). Burden

- 38 - of general medical conditions among individuals with bipolar disorder. Bipolar Disord 6(5): 368-373. Kim HW, Rapoport SI, Rao JS (2010a). Altered expression of apoptotic factors and synaptic markers in postmortem brain from bipolar disorder patients. Neurobiol Dis 37(3): 596-603. Kim HW, Rapoport SI, Rao JS (2010b). Altered expression of apoptotic factors and synaptic markers in postmortem brain from bipolar disorder patients. Neurobiol Dis 37(3): 596-603. Kramer PR, Kramer SF, Guan G (2004). 17 beta-estradiol regulates cytokine release through modulation of CD16 expression in monocytes and monocyte-derived macrophages. Arthritis Rheum 50(6): 1967-197. Krishnan KR (2005). Psychiatric and medical comorbidities of bipolar disorder. Psychosom Med 67(1): 1-8. Kroczka B, Branski P, Palucha A, Pilc A, Nowak G (2001). Antidepressant-like properties of zinc in rodent forced swim test. Brain Res Bull 55(2): 297-300. Kroczka B, Zieba A, Dudek D, Pilc A, Nowak G (2000). Zinc exhibits an antidepressant- like effect in the forced swimming test in mice. Pol J Pharmacol 52(5): 403-406. Leboyer M, Soreca I, Scott J, Frye M, Henry C, Tamouza R, et al. (2012). Can bipolar disorder be viewed as a multi-system inflammatory disease? J Affect Disord 141(1): 1-10. Lindqvist D, Janelidze S, Hagell P, Erhardt S, Samuelsson M, Minthon L, et al. (2009). Interleukin-6 is elevated in the cerebrospinal fluid of suicide attempters and related to symptom severity. Biol Psychiatry 66(3): 287-292. Liuzzi JP, Lichten LA, Rivera S, Blanchard RK, Aydemir TB, Knutson MD, et al. (2005). Interleukin-6 regulates the zinc transporter Zip14 in liver and contributes to the hypozincemia of the acute-phase response. Proc Natl Acad Sci U S A 102(19): 6843-6848. Lugo-Huitrón R, Ugalde Muñiz P, Pineda B, Pedraza-Chaverrí J, Ríos C, Pérez-de la Cruz V (2013). Quinolinic acid: an endogenous neurotoxin with multiple targets. Oxid Med Cell Longev 2013: 104024. López-Muñoz F, Alamo C (2009). Monoaminergic neurotransmission: the history of the discovery of antidepressants from 1950s until today. Curr Pharm Des 15(14): 1563-1586. Maes M, Bosmans E, De Jongh R, Kenis G, Vandoolaeghe E, Neels H (1997a). Increased serum IL-6 and IL-1 receptor antagonist concentrations in major depression and treatment resistant depression. Cytokine 9(11): 853-858. Maes M, D'Haese PC, Scharpé S, D'Hondt P, Cosyns P, De Broe ME (1994). Hypozincemia in depression. J Affect Disord 31(2): 135-140. Maes M, Meltzer HY, Bosmans E, Bergmans R, Vandoolaeghe E, Ranjan R, et al. (1995). Increased plasma concentrations of interleukin-6, soluble interleukin-6, soluble interleukin-2 and transferrin receptor in major depression. J Affect Disord 34(4): 301-309. Maes M, Vandoolaeghe E, Neels H, Demedts P, Wauters A, Meltzer HY, et al. (1997b). Lower serum zinc in major depression is a sensitive marker of treatment resistance and of the immune/inflammatory response in that illness. Biol Psychiatry 42(5): 349-358. Maes M, Ombelet W, De Jongh R, Kenis G, Bosmans E (2001). The inflammatory response following delivery is amplified in women who previously suffered from major depression, suggesting that major depression is accompanied by a

- 39 - sensitization of the inflammatory response system. J Affect Disord 63(1-3): 85-92. Martinez JM, Garakani A, Yehuda R, Gorman JM (2012). Proinflammatory and "resiliency" proteins in the CSF of patients with major depression. Depress Anxiety 29(1): 32- 38. Marreiro DD, Cruz KJ, Morais JB, Beserra JB, Severo JS, de Oliveira AR (2017). Zinc and Oxidative Stress: Current Mechanisms. Antioxidants (Basel) 6(2). McLoughlin IJ, Hodge JS (1990). Zinc in depressive disorder. Acta Psychiatr Scand 82(6): 451-453. McNamara RK, Lotrich FE (2012). Elevated immune-inflammatory signaling in mood disorders: a new therapeutic target? Expert Rev Neurother 12(9): 1143-1161. Menche N. (ed.) Biologie Anatomie Physiologie. Munich: Urban & Fischer/ Elsevier; 2012. Merikangas KR, Akiskal HS, Angst J, Greenberg PE, Hirschfeld RM, Petukhova M, et al. (2007). Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Arch Gen Psychiatry 64(5): 543-552. Merikangas KR, Jin R, He JP, Kessler RC, Lee S, Sampson NA, et al. (2011). Prevalence and correlates of bipolar spectrum disorder in the world mental health survey initiative. Arch Gen Psychiatry 68(3): 241-251. Miklowitz DJ, Johnson SL (2006). The psychopathology and treatment of bipolar disorder. Annu Rev Clin Psychol 2: 199-235. Moieni M, Irwin MR, Jevtic I, Olmstead R, Breen EC, Eisenberger NI (2015b). Sex differences in depressive and socioemotional responses to an inflammatory challenge: implications for sex differences in depression. Neuropsychopharmacology 40(7): 1709-1716. Mukherjee D, Krishnamurthy VB, Millett CE, Reider A, Can A, Groer M, et al. (2017). Total sleep time and kynurenine metabolism associated with mood symptom severity in bipolar disorder. Bipolar Disord. Młyniec K, Davies CL, Budziszewska B, Opoka W, Reczyński W, Sowa-Kućma M, et al. (2012a). Time course of zinc deprivation-induced alterations of mice behavior in the forced swim test. Pharmacol Rep 64(3): 567-575. Młyniec K, Nowak G (2012b). Zinc deficiency induces behavioral alterations in the tail suspension test in mice. Effect of antidepressants. Pharmacol Rep 64(2): 249-255. Nourmohammadi I, Ghaderi M, Hydar S, Noormohammadi E (2007). Hypozincemia in bipolar I disorder (BID) patients. Current Topics in Nutraceutical Research 5(2-3): 135-138. Nowak G, Siwek M, Dudek D, Zieba A, Pilc A (2003a). Effect of zinc supplementation on antidepressant therapy in unipolar depression: a preliminary placebo-controlled study. Pol J Pharmacol 55(6): 1143-1147. Nowak G, Szewczyk B, Sadlik K, Piekoszewski W, Trela F, Florek E, et al. (2003b). Reduced potency of zinc to interact with NMDA receptors in hippocampal tissue of suicide victims. Pol J Pharmacol 55(3): 455-459. Nowak G, Szewczyk B, Wieronska JM, Branski P, Palucha A, Pilc A, et al. (2003c). Antidepressant-like effects of acute and chronic treatment with zinc in forced swim test and olfactory bulbectomy model in rats. Brain Res Bull 61(2): 159-164. O'Connor JC, Lawson MA, André C, Moreau M, Lestage J, Castanon N, et al. (2009b). Lipopolysaccharide-induced depressive-like behavior is mediated by indoleamine 2,3-dioxygenase activation in mice. Mol Psychiatry 14(5): 511-522. O'Connor JC, Lawson MA, André C, Moreau M, Lestage J, Castanon N, et al. (2009c). Lipopolysaccharide-induced depressive-like behavior is mediated by indoleamine 2,3-dioxygenase activation in mice. Mol Psychiatry 14(5): 511-522.

- 40 - Offner PJ, Moore EE, Biffl WL (1999). Male gender is a risk factor for major infections after surgery. Arch Surg 134(9): 935-938; discussion 938-940. Oyebode, F. (2010) ‘Dearest, I feel certain I am going mad again’: The suicide note of Virgina Woolf. Advances in Psychiatric Treatment, 16(4), 280- 280.doi:10.1192/apt.16.4.280 Parker G, Blanch B, Crawford J (2011). Does gender influence response to differing psychotherapies by those with unipolar depression? J Affect Disord 130(1-2): 17- 20. Patki G, Solanki N, Atrooz F, Allam F, Salim S (2013). Depression, anxiety-like behavior and memory impairment are associated with increased oxidative stress and inflammation in a rat model of social stress. Brain Res 1539: 73-86. Phillips JA (2014). A changing epidemiology of suicide? The influence of birth cohorts on suicide rates in the United States. Soc Sci Med 114: 151-160. Pitychoutis PM, Nakamura K, Tsonis PA, Papadopoulou-Daifoti Z (2009). Neurochemical and behavioral alterations in an inflammatory model of depression: sex differences exposed. Neuroscience 159(4): 1216-1232. Prasad AS (2014a). Zinc is an Antioxidant and Anti-Inflammatory Agent: Its Role in Human Health. Front Nutr 1: 14. Prasad AS (2014b). Zinc: an antioxidant and anti-inflammatory agent: role of zinc in degenerative disorders of aging. J Trace Elem Med Biol 28(4): 364-371. Raison CL, Rutherford RE, Woolwine BJ, Shuo C, Schettler P, Drake DF, et al. (2013b). A randomized controlled trial of the tumor necrosis factor antagonist infliximab for treatment-resistant depression: the role of baseline inflammatory biomarkers. JAMA Psychiatry 70(1): 31-41. Rao JS, Harry GJ, Rapoport SI, Kim HW (2010). Increased excitotoxicity and neuroinflammatory markers in postmortem frontal cortex from bipolar disorder patients. Mol Psychiatry 15(4): 384-392. Richfield EK (1993). Zinc modulation of drug binding, cocaine affinity states, and dopamine uptake on the dopamine uptake complex. Mol Pharmacol 43(1): 100- 108. Saunders EF, Fitzgerald KD, Zhang P, McInnis MG (2012). Clinical features of bipolar disorder comorbid with anxiety disorders differ between men and women. Depress Anxiety 29(8): 739-746. Saunders EF, Nazir R, Kamali M, Ryan KA, Evans S, Langenecker S, et al. (2014). Gender differences, clinical correlates, and longitudinal outcome of bipolar disorder with comorbid migraine. J Clin Psychiatry 75(5): 512-519. Saunders EF, Reider A, Singh G, Gelenberg AJ, Rapoport SI (2015). Low unesterified:esterified eicosapentaenoic acid (EPA) plasma concentration ratio is associated with bipolar disorder episodes, and omega-3 plasma concentrations are altered by treatment. Bipolar Disord 17(7): 729-742. Schaefer C, Hidalgo TR, Cashion L, Petry H, Brooks A, Szymanski P, et al. (2006). Gene- based delivery of IFN-beta is efficacious in a murine model of experimental allergic encephalomyelitis. J Interferon Cytokine Res 26(7): 449-454. Schetz JA, Chu A, Sibley DR (1999). Zinc modulates antagonist interactions with D2-like dopamine receptors through distinct molecular mechanisms. J Pharmacol Exp Ther 289(2): 956-964. Schetz JA, Sibley DR (1997). Zinc allosterically modulates antagonist binding to cloned D1 and D2 dopamine receptors. J Neurochem 68(5): 1990-1997. Schmaal L, Veltman DJ, van Erp TG, Sämann PG, Frodl T, Jahanshad N, et al. (2016).

- 41 - Subcortical brain alterations in major depressive disorder: findings from the ENIGMA Major Depressive Disorder working group. Mol Psychiatry 21(6): 806- 812. Schröder J, Kahlke V, Staubach KH, Zabel P, Stüber F (1998). Gender differences in human sepsis. Arch Surg 133(11): 1200-1205. Schuch JJ, Roest AM, Nolen WA, Penninx BW, de Jonge P (2014). Gender differences in major depressive disorder: results from the Netherlands study of depression and anxiety. J Affect Disord 156: 156-163. Sensi SL, Paoletti P, Koh JY, Aizenman E, Bush AI, Hershfinkel M (2011). The neurophysiology and pathology of brain zinc. J Neurosci 31(45): 16076-16085. Shelton RC, Claiborne J, Sidoryk-Wegrzynowicz M, Reddy R, Aschner M, Lewis DA, et al. (2011). Altered expression of genes involved in inflammation and apoptosis in frontal cortex in major depression. Mol Psychiatry 16(7): 751-762. SIGERIST HE (1951). On the hundredth anniversary of Julius Pagel's birth, 29 May 1851. Bull Hist Med 25(3): 203-206. Siwek M, Dudek D, Schlegel-Zawadzka M, Morawska A, Piekoszewski W, Opoka W, et al. (2010). Serum zinc level in depressed patients during zinc supplementation of imipramine treatment. J Affect Disord 126(3): 447-452. Sowa-Kucma M, Szewczyk B, Sadlik K, Piekoszewski W, Trela F, Opoka W, et al. (2013). Zinc, magnesium and NMDA receptor alterations in the hippocampus of suicide victims. Journal of Affective Disorders 151(3): 924-931. Stanley PC, Wakwe VC (2002). Toxic trace metals in the mentally ill patients. Niger Postgrad Med J 9(4): 199-204. Sterpellone L (2002). [The remarkable odyssey of the potato]. Kos(199): 58-62. Szewczyk B, Poleszak E, Sowa-Kućma M, Wróbel A, Słotwiński S, Listos J, et al. (2010a). The involvement of NMDA and AMPA receptors in the mechanism of antidepressant-like action of zinc in the forced swim test. Amino Acids 39(1): 205- 217. Szewczyk B, Poleszak E, Sowa-Kućma M, Wróbel A, Słotwiński S, Listos J, et al. (2010b). The involvement of NMDA and AMPA receptors in the mechanism of antidepressant-like action of zinc in the forced swim test. Amino Acids 39(1): 205- 217. Takeda A, Minami A, Seki Y, Oku N (2003). Inhibitory function of zinc against excitation of hippocampal glutamatergic neurons. Epilepsy Res 57(2-3): 169-174. Tasca C, Rapetti M, Carta MG, Fadda B (2012a). Women and hysteria in the history of mental health. Clin Pract Epidemiol Ment Health 8: 110-119. Tasca C, Rapetti M, Carta MG, Fadda B (2012b). Women and hysteria in the history of mental health. Clin Pract Epidemiol Ment Health 8: 110-119. Tassabehji NM, Corniola RS, Alshingiti A, Levenson CW (2008). Zinc deficiency induces depression-like symptoms in adult rats. Physiol Behav 95(3): 365-369. Tsirigotis K, Gruszczynski W, Tsirigotis M (2011). Gender differentiation in methods of suicide attempts. Med Sci Monit 17(8): PH65-70. Udina M, Castellví P, Moreno-España J, Navinés R, Valdés M, Forns X, et al. (2012). Interferon-induced depression in chronic hepatitis C: a systematic review and meta-analysis. J Clin Psychiatry 73(8): 1128-1138. Valentine RJ, McAuley E, Vieira VJ, Baynard T, Hu L, Evans EM, et al. (2009). Sex differences in the relationship between obesity, C-reactive protein, physical activity, depression, sleep quality and fatigue in older adults. Brain Behav Immun 23(5): 643-648.

- 42 - Valkanova V, Ebmeier KP, Allan CL (2013). CRP, IL-6 and depression: a systematic review and meta-analysis of longitudinal studies. J Affect Disord 150(3): 736-744. Walker AK, Budac DP, Bisulco S, Lee AW, Smith RA, Beenders B, et al. (2013). NMDA receptor blockade by ketamine abrogates lipopolysaccharide-induced depressive- like behavior in C57BL/6J mice. Neuropsychopharmacology 38(9): 1609-1616. Weissman MM, Bland R, Joyce PR, Newman S, Wells JE, Wittchen HU (1993). Sex differences in rates of depression: cross-national perspectives. J Affect Disord 29(2-3): 77-84. Weissman MM, Bland RC, Canino GJ, Greenwald S, Hwu HG, Lee CK, et al. (1994). The cross national epidemiology of obsessive compulsive disorder. The Cross National Collaborative Group. J Clin Psychiatry 55 Suppl:5-10. Zhang Y, Igwe OJ (2017a). Lipopolysaccharide (LPS)-mediated priming of toll-like receptor 4 enhances oxidant-induced prostaglandin E2 biosynthesis in primary murine macrophages. Int Immunopharmacol 54: 226-237. Zhang Y, Sun X, Nai Y, Ba M, Fu Q, Wu F (2017b). Hippocampal infusion of lipopolysaccharide induces immune responses and results in seizures in rats. Neuroreport 28(4): 200-207. Zhao WX, Zhang JH, Cao JB, Wang W, Wang DX, Zhang XY, et al. (2017). Acetaminophen attenuates lipopolysaccharide-induced cognitive impairment through antioxidant activity. J Neuroinflammation 14(1): 17. Zimmermann P, Brückl T, Nocon A, Pfister H, Lieb R, Wittchen HU, et al. (2009). Heterogeneity of DSM-IV major depressive disorder as a consequence of subthreshold bipolarity. Arch Gen Psychiatry 66(12): 1341-1352.

- 43 -

Chapter 2

Peripheral zinc and neopterin concentrations are associated with mood severity in bipolar disorder in a gender-specific manner

2.1 Abstract

Bipolar disorder (BD) is a recurrent, episodic mood disorder for which there is no current diagnostic, prognostic or theranostic biomarker. Two peripheral markers of the acute phase immune response, zinc and neopterin, are consistently associated with severity of depression in literature. Given gender differences in clinical presentation of BD and in inflammatory processes, we aimed to explore the interaction between gender and immune biomarkers to predict mood severity in BD. Participants with DSM IV BD I and II were recruited through the Pennsylvania Psychiatric Institute during an acute mood episode.

Healthy controls (HC) were recruited through advertisements. Participants fasted for at least

6 hours when blood was drawn for biomarkers. We found that zinc concentrations were significantly lower in the BD group at baseline (p<.05), and there was also a significant interaction between gender and zinc (p<.05), associated with depression severity. Also, we found a significant interaction between gender and neopterin, associated with mania severity (p<.05). We found that mania severity was associated with neopterin in men, while depression severity was positively associated with zinc in women. Our report bears replication in larger samples and highlights the potential for differences in the underlying pathophysiology between men and women with BD.

- 44 - 2.2 Introduction

Bipolar disorder (BD) is a recurrent, episodic mood disorder with an estimated lifetime prevalence of 3.9% in the U.S. (Kessler et al., 2005). Mood episodes may be experienced in BD as major depressive, manic, hypomanic and mixed episodes (American Psychiatric

Association, 2013). BD type I (BD-I) is defined by the lifetime experience of at least one manic episode, and BD type II (BD-II) is defined by the lifetime experience of at least one hypomanic episode and at least one depressive episode (American Psychiatric Association,

2013). Though many pharmacotherapies exist for BD, individuals living with BD still suffer with mood symptoms approximately 50% of the time (Judd et al., 2002). While BD is equally prevalent in men and women, the clinical presentation and co-morbidities can differ. Women with BD have higher rates than men of BD-II, co-morbid anxiety disorders, post-traumatic stress disorder (PTSD) and bulimia (Baldassano et al., 2005, Saunders et al., 2012). The risk for migraine and the impact of co-morbid migraine on poor mood outcome is greater in women with BD (Altshuler et al., 2010; Saunders et al., 2014). Moreover, poor sleep quality affects women with BD more than men with BD by increasing frequency and severity of mood episodes (Gruber et al., 2009, 2011; Saunders et al., 2015). This suggests that gender is an important moderator of symptomatology in BD and may point to differing underlying pathophysiology in men and women with BD.

Zinc is an essential micronutrient and plays multiple roles in the brain, as a signaling element, as a co-factor in enzymatic reactions and as a modulator of the dopaminergic system (Frederickson et al., 2005). In human studies of major depressive disorder (MDD), peripheral zinc concentrations were shown to be significantly lower in groups of men and women with depression compared to healthy controls, and zinc concentration has been inversely correlated to severity of depression in men and women (McLoughlin and Hodge,

1990; Maes et al., 1994; Amani et al., 2010; Siwek et al., 2010). Also, studies in male rodents

- 45 - have shown that dietary zinc deficiency is causally related to depressive-like phenotypes, and zinc supplementation can reverse depressive-like behaviors (Szewczyk et al., 2002,

2009; Nowak et al., 2003; Cieślik et al., 2007; Mlyniec and Nowak, 2012b). In this way, literature has supported peripheral zinc as having an important role as a biomarker for both men and women in MDD. However, to our knowledge, no studies have directly probed gender differences in the relationship between peripheral zinc and BD.

Zinc concentration in the periphery is reduced in the presence of pro-inflammatory cytokines (Liuzzi et al., 2006). While some inflammatory markers have been found to be associated with mood states in BD and in MDD, the role of inflammation as an inciting, moderating or mediating factor in the development or persistence of mood episodes is unclear. Neopterin is a circulating signaling marker of cellular inflammation produced by activated macrophages and has been shown to have an association with depression severity (Maes et al., 1994). Also, neopterin negatively correlates with peripheral zinc in men and women with MDD (Maes et al., 1994). Moreover, while there are clear gender differences in autoimmune disease, and gender-based differences in inflammatory activation due to sex hormones (Fish et al., 2008), the role of gender differences in inflammation in mood disorders has not been explored fully.

To address the gaps in knowledge regarding circulating biomarkers of zinc and inflammatory status, we measured the peripheral zinc and neopterin concentrations in symptomatic men and women with BD, in depressed or mixed states, and compared them to healthy control participants. We also examined the relationship between mood severity, zinc, and neopterin as a function of gender in participants with BD.

- 46 - 2.3 Methods

2.3.1 Participants

Participants with DSM IV BD I and II (N=27, men=14, women=13; age range=19-55y) were recruited through the Pennsylvania Psychiatric Institute (PPI) in Harrisburg, PA. Healthy control participants (HC, N=31, men=13, women=18; age range=20-58y), with no personal or family history of mood disorders, were recruited through advertisements posted in the

Penn State College of Medicine and Penn State Milton S. Hershey Medical Center.

Exclusion criteria included inability to consent, pregnancy, intoxication with alcohol or substances of abuse, major endocrinological or rheumatological illness, and use of non- steroidal anti-inflammatory drugs (NSAIDs). The study was approved by the institutional review board (IRB 39364EP) at Penn State Hershey College of Medicine and complied with the Declaration of Helsinki. No research was conducted until consent was obtained from each participant. If a participant was found to be unable to demonstrate adequate insight into illness, they were considered unable to consent.

2.3.2 Rating scales

The Mini Neuropsychiatric Interview, a DSM-IV TR-based structured interview assessment, was performed at baseline to confirm diagnosis of BD (Sheehan et al., 1998). Interviews were conducted by physicians and trained research assistants. Demographic information was collected from the individual, and clinical information including current medication use, smoking status was collected via self-report. Depressive symptoms were rated with

Hamilton Depression Rating Scale 21 plus atypical items (HDRS-21 + AT: Hamilton, 1960).

Manic symptoms were assessed with Young Mania Rating Scale (YMRS: Young et al.,

1978). A combination of clinically significant manic and depressed symptoms defined a mixed-manic phenotype (MM). HDRS-21 + AT scores between 0 and 6 indicated no

- 47 - depression, scores between 7 and 17 indicated mild depression, scores between 18 and 24 indicated moderate depression, and scores over 24 indicated severe depression

(Zimmerman et al., 2013). YMRS scores below 7 were considered not manic. Scores between 7 and 12 were mild, and above 12 were severe manic phenotypes (Young et al.,

1978). Height and weight were measured, and body mass index (BMI) was calculated using the formula: weight (kg) divided by [height (m)] x [height (m)]. Biological samples were collected as described below.

During the study, participants received naturalistic treatment, and dietary intake was not controlled. After discharge from the hospital or partial hospital program, participants were followed each week by phone and assessed for clinical improvement. A return visit was scheduled with repeat measures of mood and a blood draw when the subject was asymptomatic, or after 3 months had elapsed (regardless of if they had resolved symptoms).

Due to irregular contact with some participants, the maximum number of days for follow-up was 187 (the median was 22 days and average was 52 days). Due to unstable housing situations, approximately 50% of the subjects were lost to follow-up (14 out of 27). At the return visit, height and weight were measured, medication use was recorded, and mood was assessed using the HDRS-21 + AT and YMRS.

2.3.3 Sample Collection and Biomarker Analysis

Blood samples were drawn for serum in vacutainers in the morning between 7:15 and 10:30

AM for the BD group (average time was 8:24 AM), and between 8:00 AM and 2:00 PM for the HC group (average time was 9:19 AM). All participants had been fasting for at least 6 hours at the time of blood draw. After centrifugation for 10 min at 654.03 g, the serum was extracted and stored in plastic tubes at -80ᵒ C until used. For zinc analysis, serum samples were digested in nitric acid (1N) for at least 24 hours and Zn2+ concentration was assessed

- 48 - using flame atomic absorption spectrometry (AAS) as previously described (Dempsey, et al., 2012). Neopterin was measured by enzyme-linked immunosorbent assay (BRAHMS,

Hennigsdorf, Germany) according to the manufacturer’s instructions, with a detection limit of 2 nmol/L.

2.3.4 Statistical Analysis

For comparison of zinc concentrations between Healthy Controls (HC) vs Bipolar Subjects

(BD), and women vs. men groups, we used a two-way analysis of covariance (ANCOVA) with sex and diagnosis as predictor variables. Body Mass Index (BMI) was significantly different between the BD and HC groups (t= 3.457, p=0.001) and was a covariate in the

ANCOVA analysis. Multiple linear regression analyses were performed to analyze the predictive power of three main effects, zinc, neopterin and gender, on mood severity. For comparison of two time points, baseline and follow-up, zinc concentrations and mood scales, a paired t-test was used. Categorical variables between BD and HC groups were compared using a Χ2 test. All statistical analyses were performed using SPSS version 22,

IBM SPSS Statistics (IBM Corporation, Armonk, NY, USA) software. A significant effect was documented at p<.05.

2.4 Results

2.4.1 Demographic and Clinical Data

Table 2.1 describes the demographic, clinical and biological data from the sample. Thirty

BD subjects were recruited, and 27 had samples available for this analysis (13 women and

14 men). The BD and HC groups did not differ in age, gender, race, or ethnicity. A lesser proportion of the BD group was married and employed. BMI was higher in the BD group than the HC group, and a greater proportion of the BD group smoked cigarettes. The BD

- 49 - group had 78% BD-I, and the treatments being taken at the time of the study are described in Table 1. On average, the BD group was experiencing severe depressive symptoms and moderate manic symptoms at the time of the study. At baseline, there were 8 depressed BD participants, 18 mixed-state BD participants and 1 manic BD participant.

Table 2-1: Demographic and Clinical Data

- 50 - Table 2.1. Demographic and clinical data of the participants with bipolar disorder (BD) and healthy controls (HC) at baseline. Data represented as Mean (± SD) or N (%). P values represent comparisons between ‘BD all’ and ‘HC all’ groups. *Neopterin N values

(BD=27 men=14, women=13; HC=31 men=13, women=18).

2.4.2 Peripheral Biomarkers in BD and HC

The unadjusted mean zinc concentration (±SD) for BD was 94.2 (±15.6) µg/dL, and in HC it was 106.3 (±17.9) µg/dL. We found mean zinc to be significantly lower in the BD participants than the HC participants (p=0.037) (Table 2.1, Figure 2.1). The range of zinc concentrations was greater in the BD participants than the HC subjects (Figure 2.1). High outliers were defined as any value above the 75th percentile score plus 1.5*interquartile range (IQR). Low outliers were defined as any value below the 25th percentile minus 1.5*IQR. Three outliers were present in the BD participants and were removed because they were not representative of physiological concentrations of zinc. The bounds for acceptable serum zinc concentrations were made based on recommended clinical cutoffs (Hotz et al., 2003).

The total number of BD participants after removal of outliers was 24 (11 women and 13 men). One subject was zinc-deficient. There was no statistically significant difference in mean zinc by gender (Figure 2.2).

The unadjusted mean neopterin concentrations (±SD) were 6.2 (±2.5) nM in BD and

5.4 (±1.4) nM in HC. We found no significant difference in neopterin between HC and BD groups (not shown). We found no significant correlation between neopterin and zinc in our population (F (3, 20)=.655, p=.589).

- 51 - Figure 2-1: Serum Zinc in Symptomatic Bipolar Disorder

Figure 2-1. Box and whisker plot for peripheral zinc in bipolar disorder (BD) and healthy control (HC) groups at baseline. The range of zinc concentrations was larger for the BD population (Range=129.25), compared to HCs (Range=63.5). Women with BD had two outliers, and men had one. Middle bar in each box plot is the median value. Asterisks

(*) represent outliers.

- 52 - Figure 2-2: Lower Serum Zinc in BD Compared to HCs

Figure 2-2. Two-way ANCOVA of peripheral zinc concentrations by diagnosis and gender. A main effect showed that participants with bipolar disorder (BD) had significantly lower zinc concentrations on average (p=.037), as compared to healthy controls (HC). BD

N=24 (men N=13 and women N=11) and HCs N=31 (men N=13 and women N=18). Data represents marginal means ±SE. Marginal means are means adjusted for the covariate BMI.

- 53 - 2.4.3 Differential Association Between Zinc, Neopterin and Mood Symptoms by

Gender

We then assessed for an association between peripheral zinc and neopterin concentrations, and mood symptoms by gender. A multiple linear regression was calculated to determine association between depression severity score and zinc by gender (F(3, 20)= 2.336, p=.1).

The interaction of gender × zinc was significantly associated with depression severity scores

(Figure 2.3, t=2.41, p<.05). As illustrated in Figure 2.3, zinc and depression severity scores were positively associated in women and negatively associated in men.

We then calculated a multiple linear regression to determine the association between mania severity score and neopterin by gender (F(3,20)=4.583, p=.013). There was a significant interaction between gender and neopterin associated with mania severity (t=-

2.72, p<.05). As seen in Figure 2.4, mania severity had a positive association with neopterin concentration in men, but not in women. A multiple linear regression was calculated to determine association between depression severity and neopterin by gender; no significant regression equation was found (F(3, 20)=1.118, p=.365). Also, a multiple linear regression was calculated to determine association between mania severity and zinc, and no significant equation was found (F(3,20)=.372, p=.774).

- 54 - Figure 2-3: Zinc Correlates with Depression Severity in Women

Figure 2-3. Peripheral zinc is associated with depression severity in women. There is a significant interaction between zinc and gender in prediction of depression severity

(HDRS-21 +AT). Women have a positive correlation between zinc and depression severity

(Pearson’s R=0.61), and men have a negative correlation (Pearson’s R=-0.33).

- 55 - Figure 2-4: Neopterin is Associated with Mania Severity in Men

Figure 2-4. Neopterin predicts mania severity in men. There is a significant interaction between gender and neopterin in prediction of mania severity (YMRS). The correlation was stronger for men (Pearson’s R=.74) than women (Pearson’s R=.38).

- 56 - 2.4.4 Follow-up Zinc and Mood State Analysis Twelve BD participants completed the follow-up visit (4 women and 8 men). As seen in

Figure 2.5, peripheral zinc increased between baseline and follow-up (t=-2.64, p=0.02). At the follow up time point, zinc concentration in the BD group increased to a level exceeding that of zinc in the HC group at baseline.

Figure 2-5: Zinc is Significantly Increased at the Follow-up Time Point

Figure 2-5. Baseline and follow up values for participants with bipolar disorder.

Zinc concentration significantly increased at follow up. N=12 (men N=4 and women N=8).

Data represented as mean ± SE. ***p<.001, **p<.01, *p<.05.

- 57 - 2.5 Discussion

Here we have shown that zinc concentrations were lower in a cohort of symptomatic participants with BD than HC, and that zinc increased at follow-up when the BD participants were asymptomatic. Moreover, we have shown for the first time a gender- specific relationship between zinc and depression severity, and neopterin and mania severity.

BD is diagnosed as a syndrome based on clinical symptoms. Evidence supports that there are multiple possible etiological causes for BD (Belmaker, 2004). The search for biomarkers in this field serves several purposes – to investigate etiological hypotheses or to develop as diagnostic, prognostic, or theranostic markers. While low-level inflammation has been associated with MDD and BD, inflammation is a consequence of the immune response that has evolved to protect the organism and may have evolved sex-specific differences in this system due to differing reproductive needs (Raison and Miller, 2013), and energy demands. Inflammatory markers may not distinguish between diagnoses but may be used prognostically to either predict an oncoming episode or to predict treatment response. In this study, we investigated associations between neopterin as an inflammatory marker and zinc as a constitutive co-factor in inflammatory response, and mood state in BD.

We then asked if the relationship between neopterin, zinc, and mood state differed between men and women because mood phenotype in BD and inflammatory responses differ by gender.

Zinc has the potential to influence mood through its action as a neuromodulator in the brain. Zinc ion interacts with many receptors and transporters in the brain, including those for glutamate, serotonin and dopamine (Richfield et al., 1993; Wu et al., 1997; Schetz et al., 1997, 1999; Hubbard et al., 2000). Also, during periods of oxidative stress, labile zinc can accumulate and act as a potent neurotoxin and contribute to neuronal and glial death

(Sensi et al., 2011). Post-mortem studies show increased markers of oxidative stress and

- 58 - inflammation in the brains of people with BD (Kim et al., 2010; Rao et al., 2010). Such observations have contributed to the hypothesis of neuroprogression as part of the underlying pathophysiology of BD, though a connection to zinc has not been made (Berk et al., 2011). There exists a paucity of literature examining how peripheral inflammation and zinc concentrations correspond to labile zinc in the brain, and in turn, how this associates with mood. Pre-clinical models have begun to probe into these questions and more work in this area is greatly needed.

In the majority of studies examining zinc as marker of depressed mood state in men and women with MDD, lower peripheral zinc concentration has been reliably reproduced.

However, the literature is fairly sparse with respect to zinc in symptomatic men and women with BD. Stanley and Wakwe (2002) reported that, in a group of in-and-out patient BD subjects being treated with antipsychotic medication, zinc was reduced in participants with

BD compared to healthy controls. However, the authors did not include gender as a component of their analyses. Conversely, González-Estecha et al. (2011) found zinc to be increased in mania in a cohort of men and women with BD. We examined a group of men and women who were hospitalized for BD and many were experiencing both manic and depressive symptoms at the time of the study. We therefore examined mood symptoms quantitatively rather than categorically. We found significantly lower zinc concentrations in

BD participants at baseline regardless of gender, and a significant increase in zinc at follow- up that was correlated with a decrease in mood symptoms. The changes in zinc that we observed fit within the context of the current literature.

It has yet to be fully elucidated why peripheral zinc decreases based on mood state in symptomatic BD. Although reduced zinc has been implicated in mood disorders, the underlying biological etiology and implication for the illness remains vague. While decades of literature show that dietary zinc deficiency hinders neurological development, and can

- 59 - induce depression (Prasad, 1969), peripheral zinc is not necessarily altered in periods of dietary zinc deficiency, and does not reflect cellular zinc status, due to tight homeostatic control mechanisms (Maret and Sandstead, 2006). Therefore, we cannot conclude that our

BD cohort was zinc deficient based on this measure alone. However, there is evidence in literature to support other sources contributing to zinc diminishment in BD.

Another underlying factor to consider is inflammation. It is known that markers of inflammation such as pro-inflammatory cytokines are elevated in unipolar depression, bipolar depression and mania (Maes et al., 1994, 1995, 1997a, 1997b; Dickerson et al.,

2007; Drexhage et al., 2010a; Rao et al., 2010) and inflammation may resolve in periods of euthymia (Brietzke et al., 2009). Pro-inflammatory cytokines, such as IL-1β and IL-6, enable the removal of zinc from the periphery as part of the acute phase response (Liuzzi et al.,

2005). Also, peripheral zinc levels are negatively correlated with increased concentration of neopterin, IL-6, and CD4/CD8 ratios in MDD (Maes et al., 1994, 1997a, 1997b, 1999).

Further, studies looking at post mortem brain tissue (Gawryluk et al., 2011a, 2011b; Shelton et al., 2011) and CSF (Lindqvist et al., 2009; Martinez et al., 2012) have found that markers of inflammation are increased in the brain in MDD.

Although there is ample evidence in the literature to support an inverse relationship between inflammation and peripheral zinc, we did not observe a significant increase in neopterin – a marker of cellular immune system activation – in our men and women with

BD. Our result seems to be consistent with prior literature. Hoekstra et al. (2006) measured neopterin in symptomatic and euthymic BD patients and found it to be decreased relative to healthy controls and MDD participants, irrespective of symptomatic status. Also,

Reininghaus et al. (2014) observed that peripheral neopterin was significantly lower in euthymic BD participants compared to healthy controls. A possible explanation for these observations centers on the role of the cytokine IFN-γ, which is required for neopterin

- 60 - release from monocytes/macrophages (Murr et al., 2002). IFN-γ release has been reported to be lower in BD (Barbosa et al., 2014). In this way, perhaps neopterin is not being stimulated for release in our BD group. In the current study we have shown that, while there were no differences in mean neopterin concentration between groups, neopterin was differentially associated with mania severity between genders. Although neopterin is positively associated with mania severity in men, the causal relationship remains an unknown, due to the possibility of increased cytokine production caused by enhanced psychomotor activities in manic states.

There are known gender disparities in immune activation, which may partly underlie the gender difference we observed in the association between neopterin and mania severity.

In a rodent model of trauma hemorrhage and sepsis, it was found that females in proestrus have significantly higher survival rates than males (Choudhry et al., 2005). Male sex is an independent risk factor for severe infections in surgical patients (Offner et al., 1999), and women have better survival rates during sepsis (Schrӧder et al., 1998). This phenomenon may be related to the fact that the sex hormone 17-β estradiol (E2) inhibits excess inflammatory response. E2 mediates broad anti-inflammatory effects in the body (Fish et al., 2008) and can inhibit the production of pro-inflammatory cytokines IL-1β, IL-6, and TNF-

α from monocytes and macrophages in vitro (Kramer et al., 2004). Also, ovarian fluctuations of E2 can influence pro-inflammatory cytokine concentration (Bouman et al., 2005).

Additionally, there is some evidence to indicate that peripheral zinc fluctuates during the female menstrual cycle. One study found that zinc is highest during the menses and follicular phases of the menstrual cycle (Deuster et al., 1987). It should be noted that of the 11 women in the BD group, 4 women reported having regular menstrual periods, 6 women reported irregular periods, and 1 woman did not report on the frequency of her period.

- 61 - In this way, it is possible that the lack of E2 in males may influence neopterin and mania severity, while the presence of E2 may influence zinc and depression severity in women. The effect of the ovulatory cycle on mood variability is not well characterized in women with BD (Sit et al., 2011). Some women with BD-I may experience perimenstrual mood worsening (Blehar et al., 1998), and oral contraceptives have been shown to relieve mood variability in women with BD who experience mood worsening around their menses

(Rasgon et al., 2003). However, studies have failed to find a significant association between mood symptoms and menstrual phase in women with BD (Shivakumar et al., 2008; Sit et al., 2011). Future work should explore differences in biology and phenotype of BD between men and women.

Due to the small sample size, we were not able to control for medications in the BD participants, which is clearly a distinct difference from the HC population. Also, we found that more participants in the BD group use tobacco than in the HC group (p<.0001).

However, statistical analysis revealed no significant differences in zinc concentrations between BD smokers and BD non-smokers (p=0.20). Another limitation of this study was the high dropout rate of the participants in the follow-up study. Some dropouts may have been due to transient housing situations and contact issues. Due to the number of dropouts, a response bias may be present (i.e., only the subjects with resolved mood symptoms were motivated to continue their participation). As is common in BD, our BD participants were obese – the average BMI was 30.1 (±5.8), and that obesity is associated with low-level, chronic inflammation (Gregor and Hotamisligil, 2011). Also, obesity is linked to reduced peripheral zinc (Tussing-Humphries and Nguyen, 2007). BMI was accounted for in our statistical modeling, and we found zinc to be increased at follow-up, while the average BMI remained unchanged. This indicates that obesity-related inflammation is unlikely to be the entire cause of the reduction in zinc. Another limitation of this study was the use of peripheral

- 62 - biomarkers, which do not necessarily reflect inflammation or zinc in the CNS. However,

Leboyer et al., (2012) discussed BD within a whole-body framework, arguing that BD can be conceptualized as a multi-systemic inflammatory disease. The authors’ argument lends credence to the value of peripheral biomarker analyses.

Here we have shown that men and women with BD differ in terms of how the concentration of peripheral biomarkers associates with depression and mania severity.

Mania severity in men is associated with neopterin, whereas depression severity in women is associated with zinc. These results may indicate that identifying mood symptoms in men and women with BD requires the use of different biomarker measures. Future work should extend this study toward identifying biomarkers for mood state in BD that differ by gender, and elucidate the mechanism underlying these gender differences. Further, we recommend that any biomarker research in BD using men and women should include gender as a component in the analysis.

- 63 - 2.6 References Altshuler, L.L., Kupka, R.W., Hellemann, G., Frye, M.A., Sugar, C.A., McElroy, S.L., Nolen, W.A., Grunze, H., Leverich, G.S., Keck, P.E., Zermeno, M., Post, R.M., Suppes, T., 2010. Gender and depressive symptoms in 711 patients with bipolar disorder evaluated prospectively in the Stanley Foundation bipolar treatment outcome network. Am J Psychiatry 167(6), 708-715. Amani, R., Saeidi, S., Nazari, Z., Nematpour, S., 2010. Correlation between dietary zinc intakes and its serum levels with depression scales in young female students. Biol Trace Elem Res 137(2), 150-158. American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders, 5th ed, Washington, DC. Baldassano, C.F., Marangell, L.B., Gyulai, L., Ghaemi, S.N., Joffe, H., Kim, D.R., Sagduyu, K., Truman, C.J., Wisniewski, S.R., Sachs, G.S., Cohen, L.S., 2005. Gender differences in bipolar disorder: retrospective data from the first 500 STEP- BD participants. Bipolar Disord 7(5), 465- 470. Barbosa, I.G., Machado-Vieira, R., Soares, J.C., Teixeira, A.L., 2014. The immunology of bipolar disorder. Neuroimmunomodulation 21(2-3), 117-122. Belmaker, R.H., 2004. Bipolar disorder. N Engl J Med 351(5), 476-486. Berk, M., Kapczinski, F., Andreazza, A.C., Dean, O.M., Giorlando, F., Maes, M., Yücel, M., Gama, C.S., Dodd, S., Dean, B., Magalhães, P.V., Amminger, P., McGorry, P., Malhi, G.S., 2011. Pathways underlying neuroprogression in bipolar disorder: focus on inflammation, oxidative stress and neurotrophic factors. Neurosci Biobehav Rev 35(3), 804-817. Blehar, M.C., DePaulo, J.R., Gershon, E.S., Reich, T., Simpson, S.G., Nurnberger, J.I., 1998. Women with bipolar disorder: findings from the NIMH Genetics Initiative sample. Psychopharmacol Bull 34(3), 239-243. Bouman, A., Heineman, M.J., Faas, M.M., 2005. Sex hormones and the immune response in humans. Hum Reprod Update 11(4), 411-423. Brietzke, E., Kauer-Sant'Anna, M., Teixeira, A.L., Kapczinski, F., 2009. Abnormalities in serum chemokine levels in euthymic patients with bipolar disorder. Brain Behav Immun 23(8), 1079- 1082. Cervantes, P., Gelber, S., Kin, F.N., Nair, V.N., Schwartz, G., 2001. Circadian secretion of cortisol in bipolar disorder. J Psychiatry Neurosci 26(5), 411-416. Choudhry, M.A., Schwacha, M.G., Hubbard, W.J., Kerby, J.D., Rue, L.W., Bland, K.I., Chaudry, I.H., 2005. Gender differences in acute response to trauma-hemorrhage. Shock 24 Suppl 1, 101-106. Cieślik, K., Klenk-Majewska, B., Danilczuk, Z., Wróbel, A., Łupina, T., Ossowska, G., 2007. Influence of zinc supplementation on imipramine effect in a chronic unpredictable stress (CUS) model in rats. Pharmacol Rep 59(1), 46-52. Dempsey, C., McCormick, N.H., Croxford, T.P., Seo, Y.A., Grider, A., Kelleher, S.L., 2012. Marginal maternal zinc deficiency in lactating mice reduces secretory capacity and alters milk composition. J Nutr 142(4), 655-660. Deuster, P.A., Dolev, E., Bernier, L.L., Trostmann, U.H., 1987. Magnesium and zinc status during the menstrual cycle. Am J Obstet Gynecol 157(4 Pt 1), 964-968. Dickerson, F., Stallings, C., Origoni, A., Boronow, J., Yolken, R., 2007. Elevated serum levels of C-reactive protein are associated with mania symptoms in outpatients with bipolar disorder. Prog Neuropsychopharmacol Biol Psychiatry 31(4), 952-955. Drexhage, R.C., Knijff, E.M., Padmos, R.C., Heul-Nieuwenhuijzen, L., Beumer, W., Versnel, M.A., Drexhage, H.A., 2010. The mononuclear phagocyte system and its

- 64 - cytokine inflammatory networks in schizophrenia and bipolar disorder. Expert Rev Neurother 10(1), 59-76. Fish, E.N., 2008. The X-files in immunity: sex-based differences predispose immune responses. Nat Rev Immunol 8(9), 737-744. Frederickson, C.J., Koh, J.Y., Bush, A.I., 2005. The neurobiology of zinc in health and disease. Nat Rev Neurosci 6(6), 449-462. Gawryluk, J.W., Wang, J.F., Andreazza, A.C., Shao, L., Yatham, L.N., Young, L.T., 2011a. Prefrontal cortex glutathione S-transferase levels in patients with bipolar disorder, major depression and schizophrenia. Int J Neuropsychopharmacol 14(8), 1069- 1074. Gawryluk, J.W., Wang, J.F., Andreazza, A.C., Shao, L., Young, L.T., 2011b. Decreased levels of glutathione, the major brain antioxidant, in post-mortem prefrontal cortex from patients with psychiatric disorders. Int J Neuropsychopharmacol 14(1), 123- 130. González-Estecha, M., Trasobares, E.M., Tajima, K., Cano, S., Fernández, C., López, J.L., Unzeta, B., Arroyo, M., Fuentenebro, F., 2011. Trace elements in bipolar disorder. J Trace Elem Med Biol 25 Suppl 1, S78-83. Gregor, M.F., Hotamisligil, G.S., 2011. Inflammatory mechanisms in obesity. Annu Rev Immunol 29, 415-445. Gruber, J., Harvey, A.G., Wang, P.W., Brooks, J.O., Thase, M.E., Sachs, G.S., Ketter, T.A., 2009. Sleep functioning in relation to mood, function, and quality of life at entry to the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD). J Affect Disord 114(1-3), 41-49. Gruber, J., Miklowitz, D.J., Harvey, A.G., Frank, E., Kupfer, D., Thase, M.E., Sachs, G.S., Ketter, T.A., 2011. Sleep matters: sleep functioning and course of illness in bipolar disorder. J Affect Disord 134(1-3), 416-420. HAMILTON, M., 1960. A rating scale for depression. J Neurol Neurosurg Psychiatry 23, 56-62. Hoekstra, R., Fekkes, D., Pepplinkhuizen, L., Loonen, A.J., Tuinier, S., Verhoeven, W.M., 2006. Nitric oxide and neopterin in bipolar affective disorder. Neuropsychobiology 54(1), 75-81. Hubbard, P.C., Lummis, S.C., 2000. Zn(2+) enhancement of the recombinant 5-HT(3) receptor is modulated by divalent cations. Eur J Pharmacol 394(2-3), 189-197. Hotz, C., Peerson, J.M., Brown, K.H., 2003. Suggested lower cutoffs of serum zinc concentrations for assessing zinc status: reanalysis of the second National Health and Nutrition Examination Survey data (1976-1980). Am J Clin Nutr 78(4), 756- 764. Jamison, K.R., 2000. Suicide and bipolar disorder. J Clin Psychiatry 61 Suppl 9, 47-51. Judd, L.L., Akiskal, H.S., Schettler, P.J., Endicott, J., Maser, J., Solomon, D.A., Leon, A.C., Rice, J.A., Keller, M.B., 2002. The long-term natural history of the weekly symptomatic status of bipolar I disorder. Arch Gen Psychiatry 59(6), 530-537. Kessler, R.C., Berglund, P., Demler, O., Jin, R., Merikangas, K.R., Walters, E.E., 2005. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 62(6), 593-602. Kim, H.W., Rapoport, S.I., Rao, J.S., 2010. Altered expression of apoptotic factors and synaptic markers in postmortem brain from bipolar disorder patients. Neurobiol Dis 37(3), 596-603. Kramer, P.R., Kramer, S.F., Guan, G., 2004. 17 beta-estradiol regulates cytokine release through modulation of CD16 expression in monocytes and monocyte-derived

- 65 - macrophages. Arthritis Rheum 50(6), 1967-1975. Leboyer, M., Soreca, I., Scott, J., Frye, M., Henry, C., Tamouza, R., Kupfer, D.J., 2012. Can bipolar disorder be viewed as a multi-system inflammatory disease? J Affect Disord 141(1), 1- 10. Lindqvist, D., Janelidze, S., Hagell, P., Erhardt, S., Samuelsson, M., Minthon, L., Hansson, O., Björkqvist, M., Träskman-Bendz, L., Brundin, L., 2009. Interleukin-6 is elevated in the cerebrospinal fluid of suicide attempters and related to symptom severity. Biol Psychiatry 66(3), 287-292. Liuzzi, J.P., Lichten, L.A., Rivera, S., Blanchard, R.K., Aydemir, T.B., Knutson, M.D., Ganz, T., Cousins, R.J., 2005. Interleukin-6 regulates the zinc transporter Zip14 in liver and contributes to the hypozincemia of the acute-phase response. Proc Natl Acad Sci U S A 102(19), 6843-6848. Maes, M., Bosmans, E., De Jongh, R., Kenis, G., Vandoolaeghe, E., Neels, H., 1997a. Increased serum IL-6 and IL-1 receptor antagonist concentrations in major depression and treatment resistant depression. Cytokine 9(11), 853-858. Maes, M., Bosmans, E., De Jongh, R., Kenis, G., Vandoolaeghe, E., Neels, H., 1997b. Increased serum IL-6 and IL-1 receptor antagonist concentrations in major depression and treatment resistant depression. Cytokine 9(11), 853-858. Maes, M., D'Haese, P.C., Scharpé, S., D'Hondt, P., Cosyns, P., De Broe, M.E., 1994. Hypozincemia in depression. J Affect Disord 31(2), 135-140. Maes, M., De Vos, N., Demedts, P., Wauters, A., Neels, H., 1999. Lower serum zinc in major depression in relation to changes in serum acute phase proteins. J Affect Disord 56(2-3), 189- 194. Maes, M., Meltzer, H.Y., Bosmans, E., Bergmans, R., Vandoolaeghe, E., Ranjan, R., Desnyder, R., 1995. Increased plasma concentrations of interleukin-6, soluble interleukin-6, soluble interleukin-2 and transferrin receptor in major depression. J Affect Disord 34(4), 301-309. Maes, M., Vandoolaeghe, E., Neels, H., Demedts, P., Wauters, A., Meltzer, H.Y., Altamura, C., Desnyder, R., 1997c. Lower serum zinc in major depression is a sensitive marker of treatment resistance and of the immune/inflammatory response in that illness. Biol Psychiatry 42(5), 349- 358. Maret, W., Sandstead, H.H., 2006. Zinc requirements and the risks and benefits of zinc supplementation. J Trace Elem Med Biol 20(1), 3-18. Martinez, J.M., Garakani, A., Yehuda, R., Gorman, J.M., 2012. Proinflammatory and "resiliency" proteins in the CSF of patients with major depression. Depress Anxiety 29(1), 32-38. McLoughlin, I.J., Hodge, J.S., 1990. Zinc in depressive disorder. Acta Psychiatr Scand 82(6), 451-453. Młyniec, K., Nowak, G., 2012. Zinc deficiency induces behavioral alterations in the tail suspension test in mice. Effect of antidepressants. Pharmacol Rep 64(2), 249-255. Murr, C., Widner, B., Wirleitner, B., Fuchs, D., 2002. Neopterin as a marker for immune system activation. Curr Drug Metab 3(2), 175-187. Nowak, G., Szewczyk, B., Wieronska, J.M., Branski, P., Palucha, A., Pilc, A., Sadlik, K., Piekoszewski, W., 2003. Antidepressant-like effects of acute and chronic treatment with zinc in forced swim test and olfactory bulbectomy model in rats. Brain Res Bull 61(2), 159-164. Offner, P.J., Moore, E.E., Biffl, W.L., 1999. Male gender is a risk factor for major infections after surgery. Arch Surg 134(9), 935-938; discussion 938-940. Paoletti, P., Vergnano, A.M., Barbour, B., Casado, M., 2009a. Zinc at glutamatergic

- 66 - synapses. Neuroscience 158(1), 126-136. Paoletti, P., Vergnano, A.M., Barbour, B., Casado, M., 2009b. Zinc at glutamatergic synapses. Neuroscience 158(1), 126-136. Prasad, A.S., 1969. A century of research on the metabolic role of zinc. Am J Clin Nutr 22(9), 1215-1221. Raison, C.L., Miller, A.H., 2013. Role of inflammation in depression: implications for phenomenology, pathophysiology and treatment. Mod Trends Pharmacopsychiatri 28, 33-48. Rao, J.S., Harry, G.J., Rapoport, S.I., Kim, H.W., 2010. Increased excitotoxicity and neuroinflammatory markers in postmortem frontal cortex from bipolar disorder patients. Mol Psychiatry 15(4), 384-392. Rasgon, N., Bauer, M., Glenn, T., Elman, S., Whybrow, P.C., 2003. Menstrual cycle related mood changes in women with bipolar disorder. Bipolar Disord 5(1), 48-52. Reininghaus, E.Z., McIntyre, R.S., Reininghaus, B., Geisler, S., Bengesser, S.A., Lackner, N., Hecht, K., Birner, A., Kattnig, F., Unterweger, R., Kapfhammer, H.P., Zelzer, S., Fuchs, D., Mangge, H., 2014. Tryptophan breakdown is increased in euthymic overweight individuals with bipolar disorder: a preliminary report. Bipolar Disord 16(4), 432-440. Richfield, E.K., 1993. Zinc modulation of drug binding, cocaine affinity states, and dopamine uptake on the dopamine uptake complex. Mol Pharmacol 43(1), 100- 108. Saunders, E.F., Fernandez-Mendoza, J., Kamali, M., Assari, S., McInnis, M.G., 2015. The effect of poor sleep quality on mood outcome differs between men and women: A longitudinal study of bipolar disorder. J Affect Disord 180, 90-96. Saunders, E.F., Fitzgerald, K.D., Zhang, P., McInnis, M.G., 2012. Clinical features of bipolar disorder comorbid with anxiety disorders differ between men and women. Depress Anxiety 29(8), 739-746. Saunders, E.F., Nazir, R., Kamali, M., Ryan, K.A., Evans, S., Langenecker, S., Gelenberg, A.J., McInnis, M.G., 2014. Gender differences, clinical correlates, and longitudinal outcome of bipolar disorder with comorbid migraine. J Clin Psychiatry 75(5), 512- 519. Schetz, J.A., Chu, A., Sibley, D.R., 1999. Zinc modulates antagonist interactions with D2- like dopamine receptors through distinct molecular mechanisms. J Pharmacol Exp Ther 289(2), 956-964. Schetz, J.A., Sibley, D.R., 1997. Zinc allosterically modulates antagonist binding to cloned D1 and D2 dopamine receptors. J Neurochem 68(5), 1990-1997. Schröder, J., Kahlke, V., Staubach, K.H., Zabel, P., Stüber, F., 1998. Gender differences in human sepsis. Arch Surg 133(11), 1200-1205. Sensi, S.L., Paoletti, P., Koh, J.Y., Aizenman, E., Bush, A.I., Hershfinkel, M., 2011. The neurophysiology and pathology of brain zinc. J Neurosci 31(45), 16076-16085. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, Baker, R., Dunbar, G.C., 1998. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD- 10. J Clin Psychiatry 59 Suppl 20, 22- 33;quiz 34-57. Shelton, R.C., Claiborne, J., Sidoryk-Wegrzynowicz, M., Reddy, R., Aschner, M., Lewis, D.A., Mirnics, K., 2011. Altered expression of genes involved in inflammation and apoptosis in frontal cortex in major depression. Mol Psychiatry 16(7), 751-762.

- 67 - Shivakumar, G., Bernstein, I.H., Suppes, T., Keck, P.E., McElroy, S.L., Altshuler, L.L., Frye, M.A., Nolen, W.A., Kupka, R.W., Grunze, H., Leverich, G.S., Mintz, J., Post, R.M., Network, S.F.B., 2008. Are bipolar mood symptoms affected by the phase of the menstrual cycle? J Womens Health (Larchmt) 17(3), 473-478. Sit, D., Seltman, H., Wisner, K.L., 2011. Menstrual effects on mood symptoms in treated women with bipolar disorder. Bipolar Disord 13(3), 310-317. Siwek, M., Dudek, D., Schlegel-Zawadzka, M., Morawska, A., Piekoszewski, W., Opoka, W., Zieba, A., Pilc, A., Popik, P., Nowak, G., 2010. Serum zinc level in depressed patients during zinc supplementation of imipramine treatment. J Affect Disord 126(3), 447-452. Stanley, P.C., Wakwe, V.C., 2002. Toxic trace metals in the mentally ill patients. Niger Postgrad Med J 9(4), 199-204. Szewczyk, B., Brański, P., Wierońska, J.M., Pałucha, A., Pilc, A., Nowak, G., 2002. Interaction of zinc with antidepressants in the forced swimming test in mice. Pol J Pharmacol 54(6), 681- 685. Szewczyk, B., Poleszak, E., Wlaź, P., Wróbel, A., Blicharska, E., Cichy, A., Dybała, M., Siwek, A., Pomierny-Chamioło, L., Piotrowska, A., Brański, P., Pilc, A., Nowak, G., 2009. The involvement of serotonergic system in the antidepressant effect of zinc in the forced swim test. Prog Neuropsychopharmacol Biol Psychiatry 33(2), 323- 329. Tussing-Humphries, L., Nguyen, V.T.Q., 2007. Obesity and Micronutrient Deficiencies, in: Fantuzzi, G., Braunschweig, C. (Eds.), Adipose Tissue and Adipokines in Obesity in Health and Disease. New York, New York, pp. 143-145. Wu, Q., Coffey, L.L., Reith, M.E., 1997. Cations affect [3H]mazindol and [3H]WIN 35,428 binding to the human dopamine transporter in a similar fashion. J Neurochem 69(3), 1106-1118. Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. A rating scale for mania: reliability, validity and sensitivity. Br J Psychiatry 133, 429-435. Zimmerman, M., Martinez, J.H., Young, D., Chelminski, I., Dalrymple, K., 2013. Severity classification on the Hamilton Depression Rating Scale. J Affect Disord 150(2), 384-388.

- 68 - Chapter 3

The sex-specific effects of LPS on depressive-like behavior and oxidative stress in the hippocampus of the mouse

3.1 Abstract Many human, postmortem and animal studies have implicated inflammation as a mediator of oxidative stress and a possible contributor to hippocampal atrophy in the pathophysiology of major depression. We aimed to examine the effect of peripheral, systemic inflammation on oxidative stress and apoptosis in the hippocampi of male and female mice. We hypothesized that (1) lipopolysaccharide (LPS) would induce depressive-like behavior and hippocampal oxidative stress after 1 day in males, and (2) that one systemic LPS exposure would result in hippocampal apoptosis at 28 days. Further, we predicted that female sex would confer protection from the deleterious effects of LPS molecularly, as well as behaviorally. Behavioral experiments including the forced swim and open-field tests were performed on male and female mice 24 hours post-injection (IP) of either LPS or saline, and again at 28 days. Inflammatory, oxidative stress and neurotrophic factors were measured in the hippocampi of male and female mice at both time-points. Males exhibited depressive- like behavior 24 hours after LPS administration compared to the males given saline. Female mice did not exhibit depressive-like behavior after LPS compared to females in the saline group. Males given LPS also had increased markers of oxidative stress including protein levels of superoxide dismutase (SOD) I in the hippocampus by 24 hours. Correlational analysis revealed a significant, positive association between transcript levels of glutathione reductase, an enzyme important for resisting oxidative stress, and immobility time in males

(p<0.05). At 28 days, we found decreased levels of the neurotrophic BDNF protein in males given LPS compared to males given saline (p<0.05). Moreover, males given LPS showed increased apoptosis by higher cleaved caspase-3 protein levels and TUNEL positive cells

- 69 - compared to males in the saline group (p<0.05). The current study revealed that male mice exhibited depressive-like behavior and signs of oxidative stress in the hippocampus 1 and

28 days after peripheral administration of LPS. Future work will aim to understand more fully the role of oxidative stress in depressive-states and existing sex differences.

- 70 - 3.2 Introduction

In recent decades, immune activation has been shown to be a significant contributor to the onset and progression of mood symptoms in a variety of psychiatric disorders. Major depressive disorder (MDD) is associated with activation of cell-mediated immunity (CMI), marked by an increase in acute phase reactants such as C-reactive protein (CRP), as well as a decrease in peripheral antioxidants, like zinc (Maes et al., 1994a). Studies examining clinical populations with MDD have revealed an association between peripheral pro- inflammatory cytokines, TNF-α and IL6, and depressive symptoms (Liu et al., 2012). Also, cytokines have been shown to be elevated in the cerebrospinal fluid (CSF), and levels correlate with the severity of depressive symptoms (Levine et al., 1999a; Raison et al., 2010;

Young et al., 2014). Post-mortem brain analyses have shown increased levels of cytokines, apoptotic factors and decreased metallothionein in people with MDD (Shelton et al., 2010b).

Moreover, a burgeoning body of literature has revealed that inflammation is not only associated with depressive symptoms, but also precipitates the onset of such symptoms in healthy individuals: a meta-analytic review of longitudinal studies found that inflammatory markers have a small but significant association with development of future depressive symptoms (Valkanova et al., 2013).

When considering biological factors that influence the expression of the depressive phenotype, we must consider that women of reproductive age experience MDD at twice the rate of men (Baxter et al., 2014; Whiteford, et al., 2010). Moreover, there are known gender- based differences in the immune response (Bouman et al., 2005; Fish et al., 2008). Women have better outcomes relative to males in states of enhanced inflammation, like sepsis and trauma-hemorrhage (Choudhry et al., 2005; Schröder et al., 1998). Enhanced outcomes in women might be due to the female sex hormone 17-β estradiol (E2), which is broadly anti- inflammatory and can regulate cytokine release (Kramer et al., 2004). Delineating the role

- 71 - of gender in modulating immune system activation in MDD is imperative to enhance understanding and treatment of this common and heterogeneous mood disorder. However, despite the above findings, there have been only a limited number of studies examining mechanisms underlying gender or sex-based differences in inflammation-based depression.

Preclinical rodent models have allowed for a mechanistic understanding of sex- based differences in depression. The lipopolysaccharide (LPS) mouse model of depression is a relevant model to probe not only the behavioral characteristics emergent from inflammation, and the mechanistic underpinnings of depressive states, but also sex difference that may exist therein. It is known that the LPS-to-depression pathway is mediated by the production of excitotoxic kynurenine metabolites in the brain, such as quinolinic acid (QUIN) (Dantzer et al., 2008, 2011; O’Connor et al., 2009a). Excitotoxic molecules cause oxidative and nitrosative stress (O&NS) in the brain. This is in line with clinical evidence of increased oxidative stress in the post-mortem brain tissue of people with

MDD (Shelton et al., 2010). The hippocampus has been shown to be particularly sensitive to oxidative stress, and, also to atrophy in MDD (Schmaal et al., 2016). It is still unknown whether systemic, peripheral inflammation can precipitate atrophy of the hippocampus via production of O&NS, and whether males and females differ in their responses to inflammation in this regard.

To address the gaps in knowledge related to potential sex differences in inflammation-based depression, we measured behavior and markers of oxidative stress and apoptosis in male and female C57BL/6 mice acutely and longitudinally. We hypothesized that (1) LPS would induce depressive-like behavior and hippocampal oxidative stress after

1 day in males, and (2) that one systemic LPS exposure would result in hippocampal apoptosis at 28 days. Further, we predicted that female sex would confer protection from

- 72 - the deleterious effects of LPS molecularly as well as behaviorally.

3.3 Materials and Methods

3.3.1 Animals

Male (n=36) and female (n=36) C57BL/6 mice age 9-14 weeks, weighing 20-30g were used in this study. Females and males were housed in separate rooms throughout the study. All mice were housed with a 12h-light, 12-h dark cycle (lights on at 0700 h) and food and water was given ad libitum. All mice were acclimated to the housing environment for 5 days, and subsequently habituated to experimenter handling for at least 5 days prior to injection. Mice were injected in the peritoneal cavity (IP) with either 0.83mg/kg lipopolysaccharide (LPS,

Escherichia coli O127:B8, Cat No. L3137, Sigma), or saline vehicle (0.9% NaCl), 24 hours prior to the start of behavioral testing. The experimenter was blind to treatment at the time of injection, and during subsequent behavioral testing. Behavioral tests were performed between 1900-2300 hours, in the dark under red light. All experiments were conducted in accordance with the NIH guidelines and with an approved protocol from Penn State

Institutional Animal Care and Use Committee (IACUC).

3.3.2 Experimental design

The study consisted of two experimental settings as follows:

Experiment 1 (acute): Figure 2A shows the timeline for Experiment 1. Male (N=24) and female

(N=24) mice were group housed (N=4/cage). Female estrous cycles were synchronized via the Whitten effect (Whitten, 1956). Prior to injection, vaginal smears were taken, and cytology was used to verify the estrous cycle stage of females (Byers et al., 2012). Males received a mock smear prior to injection. Following this, mice received an IP injection of either LPS

(males, N=12, females, N=12) or saline (males, N=12, females, N=12). On day 1, mice were tested for depressive and anxiety-like behaviors, as well as in mobility measures. Mice were

- 73 - euthanized with CO2 plus cardiac puncture immediately following behavioral testing.

Pilot Experiment 2 (longitudinal): Figure 5A shows the timeline for Pilot Experiment 2. Male

(N=12) and female (N=12) mice were singly housed. It is known that males and females respond differently to housing conditions; males tend to be more anxious when group housed, and females tend to be more anxious when singly housed (Palanza and Parmigiani, 2017).

We singly housed the mice in the pilot study to draw out these behavioral differences.

Females were randomly cycling throughout the 28-day pilot experiment. Females could randomly cycle because female mice must be group housed to engage the Whitten effect, and these females were singly housed. Mice were injected IP with either LPS (males, N=6, females, N=6) or saline (males, N=6 females, N=6) on day 0. Behavioral testing was performed on day 1, 14, and 28 after injection. Mice were euthanized immediately following behavioral testing in the same fashion as in Experiment 1.

3.3.3 Behavioral assessments

Forced swim test (FST). The FST was performed in both Experiment 1 and 2 under similar conditions: In Experiment 1 the FST was performed 24-28 hours after injection. Standard two- liter PYREX beakers (Corning Life Sciences, NY) were used for the FST and the test was performed in the dark, under red light. In Pilot Experiment 2 the FST was performed 24 hours,

14 days and 28 days after the injection, and plastic buckets were used. Each FST container was filled to a height of 12-14 cm with water, which was between 25-30 °C. The FST was six minutes in duration.

Open-field test (OFT). The OFT was used to assess mobility and anxiety in Experiment 1 immediately prior to the FST (Gould et al., 2009). Each test was performed in the dark, under red light, 24-28 hours post injection. At the start of the test session, an individual mouse was placed in the center of the open-field arena (50 x 50 cm) and allowed to freely explore for 5 minutes. Time spent in the center square, total distance traveled, line crossings, and rearings

- 74 - were recorded with ANYMAZE software (Stoelting, Co., USA). The sum of rearings and line crosses made up an overall ‘locomotor activity’ (LMA) value. The number of fecal boli and the presence of urine were also noted for each animal. The open-field arena was cleaned with

70% ethanol between each mouse.

3.3.4 Plasma corticosterone and zinc measurements

After cardiac puncture plasma was extracted and stored in plastic tubes at −80 °C until used.

Plasma corticosterone was measured with an ELISA from ENZO LifeSciences (Cat # 89141-

146) following manufacturer’s instructions. For zinc analysis samples were digested in nitric acid (1N) for at least 24 h and Zn2+ concentration was assessed using flame atomic absorption spectrometry (AAS) as previously described (Dempsey et al., 2012).

3.3.5 Brain dissections and preparation of homogenates

Mice were euthanized with CO2 plus cardiac puncture immediately following the final behavioral test. Brains were quickly removed after euthanasia and rinsed in ice cold PBS, then frozen in isopentane over dry ice or hemisected and immersed in RNAlater (Sigma, Cat

No. R0901). All tissue was subsequently stored at -80 °C until used. Blood was collected by cardiac puncture and plasma was separated immediately and stored at -80 °C. Later, the right and left hippocampi were grossly dissected (Paxinos and Watson, 1980a) and homogenized for protein and RNA, respectively.

3.3.6 Western blot analysis

20μg of hippocampal tissue homogenate proteins diluted with RIPA buffer (Cat No. R0278,

Sigma) with 1X HALT protease and phosphatase inhibitor cocktail (ThermoFisher, Cat No.

1861280) were subjected to SDS-polyacrylamide gel electrophoresis (PAGE) and western blotting using the NuPAGE system (Invitrogen). Primary antibody dilutions were as follows:

From Novus Biologicals (Littleton, CO) – Cu/Zn SOD (1:1000, Cat No. NBP224915ss),

- 75 - cleaved caspase-3 (1:1000, Cat No. NB10056113T). Others – BDNF (1:1000, ThermoFisher,

IL, Cat No. 08B00017W), synapsin-1 (1:1000, Millipore, Billerica, MA, Cat No. AB1543) and

β- actin (1:10,000, Sigma, Cat No. 2228). Following primary antibody hybridization, membranes were incubated at room temperature for 2 hours in one of the following secondary antibodies: anti-rat horseradish peroxidase (HRP)-conjugated secondary antibody (1:10,000), anti-rabbit HRP-conjugated secondary antibody (1:10,000) or anti-mouse HRP-conjugated secondary antibody (1:10,000). Images of immunoblots were captured by the Fluorochem M imaging system (FM 0241, Protein Simple, San Jose, CA). Quantification of band intensity was performed with Alpha View 3.3.1 software (Cell Biosciences, Inc., San Jose, CA). All protein band intensities were normalized to β-actin internal control, and data is presented as a relative fold change with saline control assigned a value of 1.

3.3.7 RNA isolation and real-time quantitative PCR (qPCR)

RNA was extracted from frozen hippocampi using the TRIzol chloroform method (Invitrogen,

Cat No.10296010). The quantity and quality of RNA was determined with a Nanodrop spectrophotometer and Northern blotting. Equal amounts of RNA from each sample were converted into cDNA using the IMPROM-II kit (Promega, Cat No. A3800). qPCR was performed using Taqman reagents, including the Universal PCR Master Mix (Thermo Fisher,

Cat No. 4324018 or 4304437) and the following primers: βactin (Mm02619580_g1), Mt1

(Mm00496660_g1), Mt2 (Mm00809556_s1), Mt3 (Mm00496661_g1). Syn1

(Mm00449772_m1), Bdnf (Mm04230607_s1), Gsr (Mm00439154_m1), Aif1

(Mm00479862_g1), Gfap (Mm01253033_m1), Sod1 (Mm01344233_g1), Sod2

(Mm01313000_m1). Gene expression was analyzed on the QuantStudio 12K Flex PCR system (Thermo Fisher). All gene expression was normalized to the β-actin house-keeping gene. Relative mRNA expression level was calculated using the standard comparative –ΔΔCt method.

- 76 - 3.3.8 Immunofluorescence

8-OHdG staining. Brains were dissected and subsequently flash frozen with isopentane over dry ice then stored at -80 °C until slicing. Coronal sections of the hippocampus were taken using a cryostat at a thickness of 20μm. Hippocampal sections were thawed at RT for at least

30 minutes prior to fixation in 3.7% paraformaldehyde. Slices were then rinsed 3 x 10 minutes in 1X PBS. Then, slices were permeabilized for two hours in a 1% Triton X-100 solution (1mL

Triton X-100 in 100mL PBS). After permeabilization, slices were repeatedly rinsed in 1X PBS, and blocked in a 5% normal goat serum (NGS) solution with .1% Triton X-100. Slides were subsequently incubated overnight in their respective primary antibodies, as follows: mouse monoclonal 8-OHdG antibody (1:1000, Abcam, Cat No: 62623). On the second day of the staining procedure, slides were rinsed 3 x 30 minutes in a 1% NGS solution. Then, slides were incubated in a goat anti-mouse secondary antibody conjugated to red-fluorescent Alexa

Fluor 568 (1:10,000), at 37 °C for one hour. Finally, all slides were incubated in DAPI nuclear stain for ten minutes (5μM, Molecular Probes, Cat No. D3571).

TSQ (N-(6-Methoxy-8-Quinolyl)-Para-Toluenesulfonamide) staining. Frozen, unfixed, 20μm coronal sections were air dried for at least 30 minutes, and subsequently immersed for 60 seconds in a solution of 5μM TSQ and then rinsed with 0.9% saline for 60 seconds. Slides were imaged immediately after staining. Quantification of TSQ signal intensity was performed using Adobe Photoshop. Mean fluorescent intensity (AU) was measured in a 100mm square within the regions of interest (mossy-fiber terminal area CA3 and DG/hilus). A background correction was made on each image by subtracting the mean fluorescence from a 100mm square of white matter tracts adjacent to the hippocampus.

Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). TUNEL labeling was carried out using the TACS TdT fluorescein labeling kit (R&D Systems, Cat No. 4812-30-K) following manufacturer’s instructions.

- 77 - 3.3.9 Microscopy

Fluorescent signal was viewed under a wide field fluorescent microscope DeltaVision RT

(Applied Precision) system with Olympus IX70 inverted microscope, CoolSNAP camera

(Roper Scientific, Duluth, GA), and SoftWorx (Applied Precision, Seattle, WA) imaging software.

3.3.10 Quantification of labeled cells

8-OHdG positive cells were counted from 20× images of the CA3 and DG/hilus hippocampal regions. For each animal, the right and left hemispheres of (typically 2 to 3) slices were counted. The number of 8-OHdG positive cells were counted using the cell counter function of Image J.

3.3.11 Statistical analysis

Data are expressed as mean ±SEM unless otherwise specified. Statistical significance was determined using non-paired Student’s t-test (Graphpad PRISM Software, Inc. San Diego,

CA). A value of p<0.05 was considered significant.

3.4 Results

3.4.1 General parameters

In Experiment 1 we found that both male and female mice lost a significant amount of weight

24 h after injection with LPS (p<0.0001, Figure 1A). LPS had no significant effect on plasma corticosterone (CORT) concentrations in males (p=0.24) or females (p=0.47) at 24 h post injection (Figure 1B). Figure 1C shows the pooled plasma zinc concentrations in males and females. Male zinc concentrations were numerically less after LPS and female zinc concentration appeared to increase after LPS, although statistics could not be performed on the pooled samples (Figure 1C).

- 78 - 3.4.2 Experiment 1: Anxiety and depressive-like behavior tests

The FST was used to assess depressive-like behavior. We found a significant effect of LPS on males in the FST (p<0.01, Figure 2B). Females showed no difference between LPS and saline treated groups (p=0.44, Figure 2B). The open field test (OFT) was used to assess general levels of mobility and anxiety in males and females (Figure 2C-E). Locomotor activity

(LMA) – a measure of general activity levels – significantly decreased after LPS in females

(p=0.03, Figure 2C). LMA was not changed in in males after LPS (p=0.11, Figure 2C). Further, total distance traveled was decreased after LPS in females (p=0.04, Figure 2D). Total distance traveled was not changed in males given LPS (p=0.08, Figure 2D). Neither male (p=0.43) nor female (p=0.12) mice exhibited anxiety-like behavior in the OFT, as measured by the number of entries into the center square (Figure 2E).

3.4.3 Measures of inflammation and oxidative stress in the hippocampus

We found altered gene expression in the hippocampus in both males and females after LPS administration (Figure 3A and 3B, respectively). Males given LPS showed a decrease in iba1 mRNA, a marker of microglial activation, relative to saline (p<.01, Figure 3A). Similarly, females given LPS decreased Iba1 (p<0.05, Figure 3B), as well as a marker of astrocytes, glial-fibrillary acidic protein (gfap) (p=.02, Figure 3B). In contrast, females given LPS showed a trend toward an increase in mRNA levels of metallothionein II (mt2), a small molecular weight protein with an important role in sequestering oxidative species intracellularly (p=.09,

Figure 3B). Further, we found that LPS caused an increase in protein levels of the antioxidant superoxide dismutase I (sod1) in the male hippocampus (p=0.03, Figure 3C). Females also had a trending increase in protein level of sod1 after LPS (p=0.07, Figure 3D). 8-OHdG is a maker of oxidative lesions on DNA, and we failed to find any significant increase in 8-OHdG staining in the hippocampus in males and females (not shown).

- 79 - 3.4.4 Markers of plasticity in the hippocampus

We found that both males and females had increased protein markers of synaptic plasticity in the hippocampus after LPS administration. LPS caused an increase in synapsin-1 (syn1) in the male hippocampus (‘p=0.03, Figure 3C). Females also had an increase in synapsin-1 protein after LPS (p=0.02, Figure 3D).

3.4.5 Measures of labile zinc in the hippocampus

Accumulation of labile zinc might be indicative of a state of increased oxidative stress in the hippocampus, and we therefore examined labile zinc using the fluorescent stain TSQ. We found that females reduced TSQ staining intensity in the CA3 region of the mossy fiber pathway after LPS (Figure 4B, p<.01). There were no observable changes in TSQ staining in the male hippocampus (Figure 4B, p=0.3).

3.4.6 Correlational examination of behavioral and transcript measures

Table 1 lists the results from a series of correlational analyses between behavioral parameters and molecular changes levels in males and females given LPS. First, immobility time (IT) in males was positively correlated with glutathione reductase mRNA levels, an enzyme that catalyzes the reduction of glutathione, an important antioxidant in the brain (p<0.05). In females, IT was negatively correlated with metallothionein III (mt3), the isoform of metallothionein found exclusively in neurons (p<0.01).

Regarding mobility measures, male LMA was trending toward a positive correlation with metallothionein I mRNA levels (mt1) (p= 0.07), sodF1 (p<0.05), and an astrocytic marker, gfap (p<0.05). Further, we found that the number of center square entries made by males was positively correlated with mt2 (p=0.03), mt-III (p=0.03), sod1 (p<.01), gfap (p=0.02) and Iba1

(p=0.02). In contrast, total distance traveled (Tdt) was negatively correlated with mt1 (p<0.01) and trend with sod1 (p=0.05) and mt1 (p=0.07). In females, the number of center square entries was negative trend with mt1 (p= 0.06).

- 80 - Figure 3-1. General Parameters

Figure 3.1. General parameters. Data is represented as mean ±SEM, n=12 mice/group for (a) weight, (b) 6-8 for CORT and (c) pooled plasma zinc samples.

- 81 - Figure 3-2: The Effect of LPS on Male and Female Behavior

a Behavioral tests Injection & harvest

Induce anestrous in females Whitten effect time day [-24] day [-3] day [0] hours [24-28] bc

Forced swim test Locomotor activity saline LPS 400 saline LPS 350 * * 300 300

250 200

200 activity locomotor 100 total time immobile (s) immobile time total

150 0 male female male female

de Total distance traveled (Tdt) Entries into center square 25 saline LPS * 20 saline LPS

20 15

15 10 10 5 5 center square entries (#) total distance traveled (m) traveled distance total

0 0 male female male female

Figure 3.2. Examination of depressive and anxiety-like behaviors in males and females. (a)

Timeline of behavioral testing. (b) The forced swim test determined total immobility time for males and females. (c) Locomotor activity. (d) Total distance traveled. (e) Center square entries. ‘*’ = significantly different from control, p<0.05. Data is represented as mean ±SEM, n=12 mice/group.

- 82 - Figure 3-3: The Effect of LPS on Oxidative Stress in the Hippocampus

Figure 3.3-Examination of transcript and protein in the male and female hippocampus. (a) qPCR in males and (b) females, all set relative to saline (as one). (c) Males showed significantly increased SODI after LPS. (*) significantly different from saline p<0.05. (#)

- 83 - p<0.09. Data is represented as mean ±SEM.

Figure 3-4: Females Reduce Labile Zinc in the Hippocampus

a b

150 Zinc staining in the hippocampus saline LPS

100 male *

50 female TSQ% salinerelative to

0

3 3 A ilus A lus C h C hi le le e e ma ma femal femal

Figure 3.4. Quantification of labile zinc in the hippocampus. TSQ quantification goes down

in females given LPS (c) but not in males (b). (*) = significantly different from control, p<.05.

Data is represented as means ±SD, N=4 mice/group. Bar is 100μm.

- 84 - Table 3-1: Examinations of correlations (Pearson's R) between behavioral and molecular changes in mice given LPS (*P<0.05)

MT1 MT 2 MT 3 SOD 1 SOD 2 GSR GFAP IBA1 SYN1 BDNF

IT -0.03 -0.05 -0.04 0.03 0.09 0.73* 0.22 0.31 0.39 -0.09 M LMA 0.66 0.57 0.59 0.73* 0.57 0.19 0.77* 0.61 0.11 -0.39 TDT 0.4 0.33 0.36 0.47 0.49 0.1 0.34 0.34 0.04 -0.31 CSQ 0.67 0.74* 0.75* 0.90* 0.77* 0.6 0.6 0.78* 0.56 -0.08

IT -0.3 0.37 -0.99* -0.4 -0.49 -0.53 -0.90 0.03 -0.88 0.02

LMA -0.8 0.54 -0.43 -0.69 -0.56 -0.67 -0.42 -0.61 0.04 0.49 F TDT -0.99* 0.89 -0.17 -0.94 -0.85 -0.89 -0.44 -0.12 0.23 -0.07 CSQ -0.94 0.77 0.02 -0.84 -0.7 -0.76 -0.22 -0.26 0.44 0.06

Table 3.1. Correlational analysis of transcript levels and behavioral parameters in LPS treated animals. (*) = significantly different from control, p<0.05. MT (metallothionein); SOD

(superoxide dismutase); GSR (glutathione reductase); GFAP (glial fibrillary acidic protein);

IBA1 (ionized calcium-binding adaptor molecule 1); SYN1 (synapsin-1); BDNF (brain derived neurotrophic factor).

- 85 - 3.4.7 Pilot Experiment 2: Longitudinal effects of LPS on male and female depressive- like behavior

In the longitudinal study, males and females performed the FST on 1 day, 14 days, and 28 days after injection (Figure 5B). We found a significant interaction between ‘time’ and ‘sex’ (F

(6, 36) = 2.865, p=0.02). Post hoc testing revealed a significant difference in immobility times between saline treated males (296.2 s) and females (330.6 s) on day 14 (p=0.04). There was also a trend toward an increase in immobility time (p=0.08) on day 1 in males treated with

LPS (275.8 s) versus those given saline (247 s).

3.4.8 Longitudinal effects of LPS on protein markers of plasticity and apoptosis

Males given LPS also showed increased protein levels of cleaved caspase-3 28 d post injection (p<0.05, Figure 5C). In concordance with this, BDNF protein levels were also significantly decreased in males given LPS (p=0.03, Figure 5C). Females showed no change in cleaved caspase-3 protein levels (p=0.22, Figure 5C).

3.4.9 Longitudinal effects of LPS on markers of apoptosis in the hippocampus

We examined the effect of LPS on markers of apoptosis in the hippocampus longitudinally. At

28 d after a single injection of LPS, males showed significantly increased TUNEL (+) cells in the hippocampus compared to males in the saline group (DG/hilus: p=0.03, Figure 5D and

CA3: p<0.01, Figure 5E). Females showed no difference between the saline and LPS conditions in either TUNEL staining (DG/hilus: p=0.88, Figure 5D, and CA3: p=0.5, Figure

5E).

- 86 - Figure 3-5: Longitudinal Examination of LPS Effects

- 87 -

Figure 3.5. Longitudinal examination of protein levels in the hippocampus. (a)

Timeline of the longitudinal study. (b) Forced swim test behavior at three timepoints in males and females. (c) Males increased cleaved caspase-3 and decreased BDNF protein levels after LPS. (d) males had increased TUNEL (+) cells after LPS. (*) = significantly different from control, p<.05. Data is represented as means ±SEM, N=3 mice/group.

- 88 - Figure 3-6: Male and female C57BL/6 mice respond differently to a single injection of lipopolysaccharide (LPS).

Figure 3.6. We propose that sex is a moderator of the effect of lipopolysaccharide (LPS) on sickness and depressive-like behavior, and the subsequent development of oxidative stress and apoptotic markers in the hippocampus. Our data indicate that the male hippocampus is less resilient that the female hippocampus in response to LPS challenge. In future, we hope to elucidate the level of hippocampal atrophy that may occur in this model in males, and the effect this has on memory-related tasks.

- 89 - 3.5 Discussion

Here we have shown that administration of LPS causes depressive-like behavior in male mice associated with increased hippocampal oxidative stress at 24 hours, and apoptosis at 28 days.

However, we found no evidence of depressive-like behavior or hippocampal apoptosis in females after injection with LPS. This is one of the first studies to show that LPS is associated with hippocampal oxidative stress and apoptosis long-term in male mice, and that female sex may confer some resilience to the adverse effects of LPS behaviorally as well as in the brain.

In our model (Figure 6), female mice decreased their level of general mobility in the open field test but did not develop depressive-like behaviors in the forced swim test (FST), thereby indicating a sickness-like phenotype. Conversely, male mice given LPS increased their immobility in the FST, indicating a depressive-like phenotype. We therefore propose that sex is a moderator of the effect of LPS on behavior (Figure 6). We also found that superoxide dismutase protein level was increased in the male hippocampus 24 hours after LPS, indicating increased oxidative stress. By 28 days, male mice had increased levels of hippocampal apoptosis and decreased brain-derived neurotrophic factor, indicating that the male hippocampus is less resilient that the female hippocampus in response to LPS challenge. In future, we hope to elucidate the level of hippocampal atrophy that may occur in this model in males, and the effect this has on memory tasks.

Major depressive disorder (MDD) is a common mood disorder, and a leading cause of disability worldwide (Whiteford et al., 2010). It is known that women of childbearing age experience MDD at approximately twice the rate of men (Baxter et al., 2014; Whiteford et al.,

2010), and are two and a half times more likely to take antidepressants than men (Pratt et al.,

2011). The reason for the preponderance of depression in women of childbearing age is largely a mystery, although there are likely a variety of contributing factors, both biological (Solomon et al., 2009) and social (Kessler et al., 2000). There is an emerging body of literature that shows

- 90 - women and men differ not only in the rate of MDD, but also in the presentation of mood symptoms. Women experience more somatic depression, marked by fatigue, sleep disturbance, pain and anxiety (Silverstein et al., 2013), whereas men may present with symptoms such as aggression, impulsivity and substance abuse (Martin et al., 2013).

It has come to light in recent decades that immune activation influences the behavioral, neuroendocrine and neurochemical features of MDD. Interestingly, females and males have known differences in immune activation (Fish et al., 2008), and so the inciting, moderating or mediating influence of inflammation in MDD may also differ between the sexes, and could potentially underlie the gender-based differences in MDD symptomatology and prevalence.

LPS is a highly applicable and robust model of inflammation-based depression in rodents.

Systemic inflammation by LPS triggers activation of the ubiquitous enzyme indoleamine dioxygenase (IDO), which induces tryptophan metabolism through the kynurenine pathway

(Dantzer et al., 2008). Excitotoxic kynurenine metabolites such as quinolinic acid (QUIN) induce depressive behavior through agonizing the NMDA receptor in the brain (Walker et al.,

2013b). This leads to changes in food intake, weight, and mood in male mice 24 hours after

LPS administration (Dantzer et al., 2008), like clinical features of depression (Dantzer et al.,

2008) and consistent with our results in males at 24 h. Our work corresponds with previous studies that showed male CD-1 mice given the same dose of LPS increased immobility in both the FST and tail suspension test (TST) at 24 hours (O’Connor et al., 2009b). O’Connor et al.

(2009b) also found that locomotor activity (LMA) was reduced at 6 hours in males. Similarly, we found LMA – a measure of general activity levels – to be reduced in females given LPS at

24 hours, which is indicative of sickness-like behaviors and not depressive-like behavior

(O’Connor et al., 2009b).

However, we found that females were resistant to LPS generating depressive-like behavior. Only a few studies have examined female behavior after systemic LPS

- 91 - administration. Painsipp et al. found that group-housed, randomly cycling C57BL/6 female mice decreased immobility in the FST 24 hours after LPS administration, indicating perhaps an increase in motivation. In the present study, we found that female behavior did not change in the FST 24 h after LPS. This discrepancy might be explained by the fact that the Painsipp

(2009) study used randomly cycling females, whereas we used females in estrus. Further,

Pitychoutis et al. (2009b) found that both males and females reduce vertical activity (rearings) after LPS, in line with our findings related to general mobility.

Many clinical and preclinical studies report the crucial role of oxidative and nitrosative stress and hippocampal atrophy on the pathophysiology of MDD (Maes et al., 2011a; Moylan et al., 2014). Oxidative stress is defined as a state where reactive oxygen and reactive nitrogen species (ROS and RNS), such as nitric oxide, superoxide, peroxide and peroxynitrite, are present in excess amounts (Moylan et al., 2014). Increased O&NS, which arises from increased production of ROS and RNS and/or decreased antioxidant function, may damage cellular components and ultimately lead to cell death (Moylan et al., 2014). A recent meta- analysis found that 8-OHdG, a marker of DNA-damage-related oxidative stress, and F2- isoprostanes, a marker of lipid peroxidation, were significantly increased in MDD (Black et al.

2015). Other studies have found increased lipid and arachidonic acid peroxidation in MDD

(Dimopoulos et al., 2008; Yager et al., 2010), as well as oxidative damage to RNA in the post- mortem hippocampus (Che et al., 2010). Furthermore, many studies have revealed a reduction in antioxidant levels in MDD, including reduced serum zinc, superoxide dismutase

(SOD) and glutathione peroxidase (Maes, et al., 2011a, Moylan et al. 2014). In our study we found that females altered hippocampal zinc after LPS, including increasing a zinc-binding protein and antioxidant, metallothionein. Interestingly, a recent study by Shelton et al. (2011), found reduced metallothionein transcript levels in the postmortem frontal cortex of persons who had lived with MDD. Considering that our female mice were resilient to depressive-like

- 92 - behavior in the FST, perhaps their compensatory alteration of metallothionein is related to a protective mechanism against oxidative stress and depressive-like behavior.

There is evidence that O&NS underlie atrophy of the hippocampus (Che et al., 2010), which is a consistently replicated finding in MDD literature (Schmaal et al., 2016). In line with this, pre-clinical models of depression and social stress indicate that the hippocampus is particularly sensitive to oxidative stress compared to other brain areas (Allam et al., 2013;

Patki et al., 2013). We found this to be the case in the present study, where markers of hippocampal oxidative stress (sod1 protein levels) were significantly increased in response to

LPS in males. Also, in line with our findings, male mice given microinjections of LPS in the hippocampus increase apoptotic cell death at 24 hours (Zhao et al., 2017).

Interestingly, we found a positive association between immobility time and glutathione reductase (gsr) transcript levels in males given LPS. The enzyme glutathione reductase has an important role in maintaining cellular oxidant balance, via the reduction of glutathione disulfide (GSSG) to glutathione (GSH) – an important cellular antioxidant (Yan et al., 2013).

Yan and colleagues have shown that glutathione reductase is an essential enzyme for host defense against the bacteria Escherichia coli in a mouse model of sepsis (Yan et al., 2013).

Yan et al. (2012) found that glutathione reductase-null mice were more susceptible to E. coli challenge, indicated by dramatically increased bacterial burden, cytokine storm, and elevated mortality. In our study we used an E. coli strain of LPS to induce depressive-like behavior; therefore, in males there may exist a link between LPS-induced immobility and a compensatory increase in glutathione reductase levels.

We extended our initial behavioral findings by determining that depressive-like behavior resolved by 28 days after LPS administration. In line with this, prior work has indicated that singly housed C57BL/6 females given LPS do not significantly differ from females given saline at 28 d (Painsipp et al., 2011). Further, we found a significant interaction

- 93 - between sex and treatment over time, and post-testing revealed a sex difference at 14 days, i.e., females had significantly higher immobility than males in the saline group. This suggests that females may respond to repeat exposures earlier than males.

Despite the apparent resolution of depressive-like behavior, deleterious effects on the hippocampus were present in males at 28 days. We found that males had higher TUNEL staining and cleaved caspase-3 protein levels in the hippocampus and significantly reduced

BDNF protein levels four weeks after LPS administration. These results are in line with literature; it is commonly found that LPS is associated with a reduction of BDNF (Guan and

Fang, 2006; Walker et al., 2013b; Wu et al., 2011). Further, a single LPS injection has been shown to cause a delayed and progressive loss of neurons in male mice at 7 and 10 months post administration (Qin et al., 2007). Zhang et al. (2017) showed that LPS causes the number of neurons in the hippocampus to decrease.

The dimorphism we have found in terms of response to LPS is consistent with differences in clinical presentation between the sexes/genders. It is known that women with

MDD tend to present with somatic symptoms, such as fatigue, headaches, sleep disturbances, appetite changes, and breathing problems, which are reminiscent of sickness behaviors that we found in our female mice (Silverstein et al., 2013). On the other hand, men with MDD tend to report loss of interest in once pleasurable activities, rather than sadness. Further, women with MDD are more likely to present with internalizing symptoms such as anxiety disorder, agoraphobia, and panic (Parker et al., 2014), whereas men have more externalizing symptoms such as aggression/violence, impulsivity and drug abuse

(Martin et al., 2013), which are consistent with the higher rates of suicide completion in men than women with MDD (APA/DSM-5, 2013).

It has been proposed that men and women may experience MDD at equal rates, but the externalizing symptomatology in males is perhaps overlooked in the diagnostic criteria

- 94 - for MDD (Martin et al., 2013). Our findings also indicate that the sex dimorphism should be considered when testing pharmacological agents in pre-clinical studies in males vs. females.

For example, it is known that women respond more slowly and are less likely to achieve an optimal response to treatment for depression (Wizemann et al., 2012).

Here we have shown that male and female mice differ in their short and long-term responses to LPS-induced inflammation. Females appear to have increased sickness-like behavior, indicated by decreased mobility in the OFT, but no hippocampal oxidative stress.

However, males have increased depressive behavior and have greater hippocampal oxidative stress. Overall, the current work reveals that males and females differ in their response to

LPS, and future work should aim to extend this study to identify the mechanism underlying the observed sex differences. Finally, the hippocampus seems to represent a common brain area that potentially mediates depression-like behaviors and is especially sensitive to oxidative stress. Future studies should directly examine causality of hippocampal oxidative stress in depressive-like behaviors in rodent models of depression, and the sex differences that may exist therein.

- 95 - 3.6 References

Allam, F., Dao, A.T., Chugh, G., Bohat, R., Jafri, F., Patki, G., et al. 2013. Grape powder supplementation prevents oxidative stress-induced anxiety-like behavior, memory impairment, and high blood pressure in rats. J Nutr. 143, 835-42. Baxter, A.J., Scott, K.M., Ferrari, A.J., Norman, R.E., Vos, T., Whiteford, H.A., 2014. Challenging the myth of an "epidemic" of common mental disorders: trends in the global prevalence of anxiety and depression between 1990 and 2010. Depress Anxiety. 31, 506-16. Bouman, A., Heineman, M.J., Faas, M.M., 2005. Sex hormones and the immune response in humans. Hum Reprod Update. 11, 411-23. Byers, S.L., Wiles, M.V., Dunn, S.L., Taft, R.A., 2012. Mouse estrous cycle identification tool and images. PLoS One. 7, e35538. Che, Y., Wang, J.F., Shao, L., Young, T., 2010. Oxidative damage to RNA but not DNA in the hippocampus of patients with major mental illness. J Psychiatry Neurosci. 35, 296-302. Choudhry, M.A., Schwacha, M.G., Hubbard, W.J., Kerby, J.D., Rue, L.W., Bland, K.I., et al. 2005. Gender differences in acute response to trauma-hemorrhage. Shock. 24 Suppl 1, 101-6. Curtis AL, Bethea T, Valentino RJ (2006). Sexually dimorphic responses of the brain norepinephrine system to stress and corticotropin-releasing factor. Neuropsychopharmacology 31(3): 544-554 Dalla C, Pitychoutis PM, Kokras N, Papadopoulou-Daifoti Z (2011). Sex differences in response to stress and expression of depressive-like behaviours in the rat. Curr Top Behav Neurosci 8: 97-118. Dantzer, R., O'Connor, J.C., Freund, G.G., Johnson, R.W., Kelley, K.W., 2008. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat Rev Neurosci. 9, 46-56. Dantzer, R., O'Connor, J.C., Lawson, M.A., Kelley, K.W., 2011. Inflammation-associated depression: from serotonin to kynurenine. Psychoneuroendocrinology. 36, 426-36. Dempsey, C., McCormick, N. H., Croxford, T. P., Seo, Y. A., Grider, A., Kelleher, S. L., 2012. Marginal maternal zinc deficiency in lactating mice reduces secretory capacity and alters milk composition. J Nutr 142, 655-660. Dimopoulos, N., Piperi, C., Psarra, V., Lea, R.W., Kalofoutis, A., 2008. Increased plasma levels of 8-iso-PGF2alpha and IL-6 in an elderly population with depression. Psychiatry Res. 161, 59-66. Drossopoulou G, Antoniou K, Kitraki E, Papathanasiou G, Papalexi E, Dalla C, et al. (2004). Sex differences in behavioral, neurochemical and neuroendocrine effects induced by the forced swim test in rats. Neuroscience 126(4): 849-857. Fish, E.N., 2008. The X-files in immunity: sex-based differences predispose immune responses. Nat Rev Immunol. 8, 737-44. Galea LA, McEwen BS, Tanapat P, Deak T, Spencer RL, Dhabhar FS (1997). Sex differences in dendritic atrophy of CA3 pyramidal neurons in response to chronic restraint stress. Neuroscience 81(3): 689-697. Gaillard RC, Spinedi E (1998). Sex- and stress-steroids interactions and the immune system: evidence for a neuroendocrine-immunological sexual dimorphism. Domest Anim Endocrinol 15(5): 345-352. Gandhi M, Aweeka F, Greenblatt RM, Blaschke TF (2004). Sex differences in pharmacokinetics and pharmacodynamics. Annu Rev Pharmacol Toxicol 44: 499-

- 96 - 523. Todd D. Gould, D. T. D., Colleen E. Kovacsics, 2009. The Open Field Test. In: Gould, T. D., (Ed), Mood and Anxiety Related Phenotypes in Mice. Humana Press. Guan, Z., Fang, J., 2006. Peripheral immune activation by lipopolysaccharide decreases neurotrophins in the cortex and hippocampus in rats. Brain Behav Immun. 20, 64- 71. Handa RJ, Burgess LH, Kerr JE, O'Keefe JA (1994). Gonadal steroid hormone receptors and sex differences in the hypothalamo-pituitary-adrenal axis. Horm Behav 28(4): 464-476. Kessler, R.C., 2000. The epidemiology of pure and comorbid generalized anxiety disorder: a review and evaluation of recent research. Acta Psychiatr Scand Suppl. 7-13. KITAY JI (1961). Sex differences in adrenal cortical secretion in the rat. Endocrinology 68: 818-824. Klein SL (2000). The effects of hormones on sex differences in infection: from genes to behavior. Neurosci Biobehav Rev 24(6): 627-638. Kramer, P.R., Kramer, S.F., Guan, G., 2004. 17 beta-estradiol regulates cytokine release through modulation of CD16 expression in monocytes and monocyte-derived macrophages. Arthritis Rheum. 50, 1967-75. Levine, J., Barak, Y., Chengappa, K.N., Rapoport, A., Rebey, M., Barak, V., 1999a. Cerebrospinal cytokine levels in patients with acute depression. Neuropsychobiology. 40, 171-6. Liu, Y., Ho, R.C., Mak, A., 2012. Interleukin (IL)-6, tumour necrosis factor alpha (TNF-α) and soluble interleukin-2 receptors (sIL-2R) are elevated in patients with major depressive disorder: a meta-analysis and meta-regression. J Affect Disord. 139, 230-9. Maes, M., D'Haese, P.C., Scharpé, S., D'Hondt, P., Cosyns, P., De Broe, M.E., 1994a. Hypozincemia in depression. J Affect Disord. 31, 135-40. Maes, M., Galecki, P., Chang, Y.S., Berk, M., 2011a. A review on the oxidative and nitrosative stress (O&NS) pathways in major depression and their possible contribution to the (neuro)degenerative processes in that illness. Prog Neuropsychopharmacol Biol Psychiatry. 35, 676-92. Martin, L.A., Neighbors, H.W., Griffith, D.M., 2013. The experience of symptoms of depression in men vs women: analysis of the National Comorbidity Survey Replication. JAMA Psychiatry. 70, 1100-6. Moylan, S., Berk, M., Dean, O.M., Samuni, Y., Williams, L.J., O'Neil, A., et al. 2014. Oxidative & nitrosative stress in depression: why so much stress? Neurosci Biobehav Rev. 45, 46-62. O'Connor, J.C., André, C., Wang, Y., Lawson, M.A., Szegedi, S.S., et al. 2009a. Interferon-gamma and tumor necrosis factor-alpha mediate the upregulation of indoleamine 2,3-dioxygenase and the induction of depressive-like behavior in mice in response to bacillus Calmette-Guerin. J Neurosci. 29, 4200-9. O'Connor, J.C., Lawson, M.A., André, C., Moreau, M., Lestage, J., et al. 2009b. Lipopolysaccharide-induced depressive-like behavior is mediated by indoleamine 2,3-dioxygenase activation in mice. Mol Psychiatry. 14, 511-22. Painsipp, E., Köfer, M.J., Sinner, F., Holzer, P., 2011. Prolonged depression-like behavior caused by immune challenge: influence of mouse strain and social environment. PLoS One. 6, e20719. Palanza P, Parmigiani S (2017). How does sex matter? Behavior, stress and animal models of neurobehavioral disorders. Neurosci Biobehav Rev 76(Pt A): 134-143.

- 97 - Paoletti, P., Vergnano, A.M., Barbour, B., Casado, M., 2009. Zinc at glutamatergic synapses. Neuroscience. 158, 126-36. Parker G, Blanch B, Crawford J (2011). Does gender influence response to differing psychotherapies by those with unipolar depression? J Affect Disord 130(1-2): 17- 20. Patki, G., Solanki, N., Atrooz, F., Allam, F., Salim, S., 2013. Depression, anxiety-like behavior and memory impairment are associated with increased oxidative stress and inflammation in a rat model of social stress. Brain Res. 1539, 73-86. Paxinos, G., Watson, C. R., Emson, P. C., 1980a. AChE-stained horizontal sections of the rat brain in stereotaxic coordinates. J Neurosci Methods 3, 129-149. Pinos H, Collado P, Rodríguez-Zafra M, Rodríguez C, Segovia S, Guillamón A (2001). The development of sex differences in the locus coeruleus of the rat. Brain Res Bull 56(1): 73-78. Pitychoutis, P.M., Nakamura, K., Tsonis, P.A., Papadopoulou-Daifoti, Z., 2009b. Neurochemical and behavioral alterations in an inflammatory model of depression: sex differences exposed. Neuroscience. 159, 1216-32. Pratt, L.A., Brody, D.J., Gu, Q., 2011. Antidepressant use in persons aged 12 and over: United States, 2005-2008. NCHS Data Brief. 1-8. Qin, L., Wu, X., Block, M.L., Liu, Y., Breese, G.R., et al. 2007. Systemic LPS causes chronic neuroinflammation and progressive neurodegeneration. Glia. 55, 453-62. Raison, C.L., Dantzer, R., Kelley, K.W., Lawson, M.A., Woolwine, B.J., et al. 2010b. CSF concentrations of brain tryptophan and kynurenines during immune stimulation with IFN-alpha: relationship to CNS immune responses and depression. Mol Psychiatry. 15, 393-403. Roberts CW, Walker W, Alexander J (2001). Sex-associated hormones and immunity to protozoan parasites. Clin Microbiol Rev 14(3): 476-488. Schmaal, L., Hibar, D.P., Sämann, P.G., Hall, G.B., Baune, B.T., et al. 2017. Cortical abnormalities in adults and adolescents with major depression based on brain scans from 20 cohorts worldwide in the ENIGMA Major Depressive Disorder Working Group. Mol Psychiatry. 22, 900-909. Schröder, J., Kahlke, V., Staubach, K.H., Zabel, P., Stüber, F., 1998. Gender differences in human sepsis. Arch Surg. 133, 1200-5. Seale JV, Wood SA, Atkinson HC, Bate E, Lightman SL, Ingram CD, et al. (2004a). Gonadectomy reverses the sexually diergic patterns of circadian and stress- induced hypothalamic-pituitary-adrenal axis activity in male and female rats. J Neuroendocrinol 16(6): 516-524. Shelton, R.C., Claiborne, J., Sidoryk-Wegrzynowicz, M., Reddy, R., Aschner, M., et al. 2011b. Altered expression of genes involved in inflammation and apoptosis in frontal cortex in major depression. Mol Psychiatry. 16, 751-62. Silverstein B, Edwards T, Gamma A, Ajdacic-Gross V, Rossler W, et al. (2013). The role played by depression associated with somatic symptomatology in accounting for the gender difference in the prevalence of depression. Soc Psychiatry Psychiatr Epidemiol 48(2): 257-263. Solomon, M.B., Herman, J.P., 2009. Sex differences in psychopathology: of gonads, adrenals and mental illness. Physiol Behav. 97, 250-8. Valkanova, V., Ebmeier, K.P., Allan, C.L., 2013. CRP, IL-6 and depression: a systematic review and meta-analysis of longitudinal studies. J Affect Disord. 150, 736-44. Walker, A.K., Budac, D.P., Bisulco, S., Lee, A.W., Smith, R.A., et al. 2013b. NMDA receptor blockade by ketamine abrogates lipopolysaccharide-induced depressive-

- 98 - like behavior in C57BL/6J mice. Neuropsychopharmacology. 38, 1609-16. Whiteford, H.A., Degenhardt, L., Rehm, J., Baxter, A.J., Ferrari, A.J., et al. 2013. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet. 382, 1575-86. WHITTEN, W. K., 1956. Modification of the oestrous cycle of the mouse by external stimuli associated with the male. J Endocrinol 13, 399-404. Wizemann TM (ed) (2012). SEX-SPECIFIC REPORTING OF SCIENTIFIC RESEARCH: A WORKSHOP SUMMARY. Institute of Medicine of the National Academies. 500 Fifth Street, NW • Washington, DC 20001. THE NATIONAL ACADEMIES PRESS: 500 Fifth Street, NW • Washington, DC 20001. Wu, S.Y., Wang, T.F., Yu, L., Jen, C.J., Chuang, J.I., et al. 2011. Running exercise protects the substantia nigra dopaminergic neurons against inflammation-induced degeneration via the activation of BDNF signaling pathway. Brain Behav Immun. 25, 135-46. Yager, S., Forlenza, M.J., Miller, G.E., 2010. Depression and oxidative damage to lipids. Psychoneuroendocrinology. 35, 1356-62. Yan J, Meng X, Wancket LM, Lintner K, Nelin LD, Chen B, et al. (2012). Glutathione reductase facilitates host defense by sustaining phagocytic oxidative burst and promoting the development of neutrophil extracellular traps. J Immunol 188(5): 2316-2327. Yan J, Ralston MM, Meng X, Bongiovanni KD, Jones AL, Benndorf R, et al. (2013). Glutathione reductase is essential for host defense against bacterial infection. Free Radic Biol Med 61: 320-332. Young, J.J., Bruno, D., Pomara, N., 2014. A review of the relationship between proinflammatory cytokines and major depressive disorder. J Affect Disord. 169, 15- 20 Zhang, Y., Sun, X., Nai, Y., Ba, M., Fu, Q., et al. 2017. Hippocampal infusion of lipopolysaccharide induces immune responses and results in seizures in rats. Neuroreport. 28, 200-207. Zhao, W.X., Zhang, J.H., Cao, J.B., Wang, W., Wang, D.X., et al. 2017. Acetaminophen attenuates lipopolysaccharide-induced cognitive impairment through antioxidant activity. J Neuroinflammation. 14, 17.

- 99 -

Chapter 4

Sex and Gender-Based Differences in Mood Disorders – Summary and Implications

4.1 Summary of Main Findings of the Dissertation

The current work aimed to fill gaps in knowledge related to inflammation and sex/gender in manic and depressive mood states. We presented two studies, one clinical and one preclinical. Our clinical research (Chapter 2) revealed gender-based differences in the relationship between inflammatory biomarkers and mood severity in Bipolar Disorder (BD).

Our preclinical study (Chapter 3) uncovered sex-based differences in behavioral and molecular measures in response to an inflammatory stimulus. Each study contributed new knowledge in areas that have been understudied.

In Chapter 2 we measured two peripheral biomarkers, neopterin, and zinc, in a cohort of individuals with symptomatic BD and compared them to a group of healthy controls (HC).

We then examined the relationship between mood severity, zinc, and neopterin as a function of gender in participants with BD. Finally, we reported on follow up data related to serum zinc concentrations and mood severity. We found that the BD group had significantly reduced serum zinc concentrations on average compared to the HCs. However, we did not see a significant difference in neopterin concentration between the two groups.

We then performed a series of multiple linear regression analyses to assess the interaction between gender and peripheral biomarkers to predict mood severity. We found that serum zinc concentrations predict severity of depression in subjects with BD, in a gender-dependent manner. Women had a positive, moderately strong association

- 100 - (Pearson's R=0.61) between zinc and depression, i.e., as serum zinc increased, so did the severity of depressive symptoms in women. Men had a weak, negative association

(Pearson's R=-0.33) between zinc and depression; as zinc concentration diminished, so did the severity of depression in men. We also found that neopterin concentrations predicted mania severity in men (Pearson's R=.74), but not in women (Pearson's R=-.38). Finally, we found that zinc concentrations were significantly increased at the follow-up time point, when the participants were asymptomatic.

In our clinical study we have shown for the first time a gender-specific relationship between zinc and depression severity, and neopterin and mania severity. Overall our study highlights the potential differences in the underlying pathophysiology of BD in men and women.

Our preclinical study built upon our clinical study by pursuing the causal role of inflammation in depressive states. In male and female mice given an endotoxin (LPS) injection, we measured peripheral zinc and corticosterone concentrations, depressive and anxiety-like behaviors, and molecular markers of oxidative stress, plasticity, and apoptosis in the brain. Further, we conducted a pilot study examining the effect of LPS on males and females 28 days after a single LPS injection. The rationale for looking longitudinally, as opposed to only acutely (24 hours), was that chronicity is a significant factor in mood disorder pathophysiology and is a contributor to the burden of these illnesses.

In Chapter 3 we have shown that administration of LPS causes depressive-like behavior in male mice, which associated with increased hippocampal oxidative stress at 24 hours and apoptosis at 28 days. We found no evidence of depressive-like behavior or hippocampal apoptosis in females after injection with LPS.

- 101 - This is one of the first studies to show that LPS associated with hippocampal oxidative stress and apoptosis long-term in male mice, and that female sex may confer some resilience to the adverse effects of LPS behaviorally as well as in the brain.

4.2 Why Are Males and Females Different?

The clinical and preclinical studies presented in this dissertation revealed significant sex and gender-based differences related to mood, peripheral biomarkers, and molecular markers in the brains of mice. The underlying causes of the observed differences are likely multitudinous and involve a complex interplay between sexually dichotomous physiological responses to stress and inflammation, sex hormones, environment, and genetics. A few of the main sex and gender-based differences described in this dissertation are discussed below.

In Chapter 3 we found significant differences between males and females in the forced swim test (FST) – a putative measure of depressive-like behavior. Our analysis revealed that females given saline had higher immobility time in the FST compared to males given saline. This result was confirmed by prior literature (Kokras et al., 2015). However, the existing body of literature on sex differences in the FST has many inconsistencies. A recent review examined over 60 studies that compared male and female behavior in the

FST and reported that 1/3rd of studies showed females had higher immobility than males

(corroborating our study), 1/3rd reported that males had higher immobility than females, and the final third failed to show any significant difference between the sexes.

Many confounding factors can influence FST behavioral performance in males and females and may explain discrepancies in literature. One factor that has been proposed is the length of time each rodent is handled before testing. It is thought that if females were handled more often, or for longer periods of time, due to monitoring of their estrous cycle,

- 102 - this would alter their performance on the FST. In fact, long-term handling has been shown to affect immobility in male rodents (Cannizzaro et al., 2002). In contrast to this, rodents are known to habituate – become accustomed to – repeated handling by the same person

(Dobrakovova et al., 1993; Kokras et al., 2015). In our study, males and females were habituated for at least five days before testing, and males were given a 'mock smear' to control for differences in handling due to estrous cycle measures in females.

Other methodological differences may partly underlie the disparate results in the literature. Various factors can potentially influence results in the FST, including the temperature and depth of water in the swimming tank, and whether the tests are scored manually or in an automated fashion. Kokras et al. (2015) found that female rats had higher immobility when the water depth was in a higher range (30-50 cm), as compared to the studies in which males had higher immobility (18-30 cm). For FST studies using mice, behavioral protocols recommend a water height between 10-15 cm (Castagné et al., 2011;

Can et al., 2012). In our behavioral tests, the height of the water in the swimming tank was always between 12-14 cm – well within these limits. Further, the water temperature in the

FST was recommended to be 'tepid' - between 24-30 °C, and in our studies the water was always between 25-30 °C.

There are also established sex differences in stress coping strategies (Dalla et al.,

2011). There are known sexual dichotomies in the stress system; females have a higher volume and number of neurons than males in the locus coeruleus (LC) – which produces norepinephrine – and is required for activation of the stress response, along with the hypothalamic-pituitary-adrenal (HPA) axis (Pinos et al., 2001). Neuronal activity in the LC is more responsive to stressors in females than in males; this effect is due to the increased sensitivity to corticotropin-releasing factor (CRF) in the female LC (Curtis et al., 2006).

- 103 - Further, female rodents are known to have higher levels of plasma corticosterone at rest and following acute and repeated stress exposure, compared to males (Galea et al.,

1997; Seale et al., 2004a, 2004b; Handa et al., 1994; Kitay 1961). Our data are consistent with this observation - female mice had higher plasma corticosterone compared to males

24 hours after injection with either saline or LPS.

There are also sex-based differences in monoamine utilization in the FST, which may also underlie the behavioral differences we observed. It is known that the FST decreases serotonergic activity in the hippocampus and the hypothalamus of female rats

(Dalla et al., 2011). In contrast, serotonergic activity is increased in the hypothalamus of males by the FST (Dalla et al., 2011). Dopaminergic activity is increased in the hippocampus and prefrontal cortex of male rats exposed to FST, while there is no effect in females (Dalla et al., 2008a). Similarly, hypothalamic serotonin mRNA levels are decreased in female rats after the FST, and hippocampal serotonin mRNA levels are increased in males after the

FST (Drossopoulou et al., 2004).

Another sex-based difference observed in Chapter 3 was the susceptibility of males to the deleterious effects of LPS, and the apparent protective effect of the female sex to development of depressive-like behavior after LPS. We found that 24 hours after LPS administration, males had significantly increased immobility time in the forced swim test, indicating increased depressive-like behavior. This finding was in line with literature; previous studies showed male CD-1 mice given the same dose of LPS increased immobility in both the FST and tail suspension test (TST) at 24 hours (O'Connor et al., 2009b).

In contrast, we found that LPS did not change female behavior in the FST. Only a few studies have examined female behavior after systemic LPS administration. Painsipp et al. (2011) found that female mice decreased immobility in the FST 24 hours after LPS administration. Similarly, Pitychoutis et al. (2009) found a beneficial effect of LPS

- 104 - administration on the FST in female rats; LPS treatment increased swimming duration in females but not males (Pitychoutis et al., 2009). In this way, it appears as though female rodents implement an advantageous coping strategy in response to LPS in the FST. LPS administration (100 mg/kg) induced a beneficial female-specific enhancement of coping ability (Pitychoutis et al., 2009). However, it has been shown that higher doses of LPS (e.g.,

2 mg/kg) reliably induced an increase in floating time in rats of both sexes, at 24 h post- injection (Tonelli et al., 2008).

One reason for the sex-based difference in behavior after LPS may be the sexually dichotomous utilization of monoamines after LPS administration. It is known that administration of LPS or cytokines induces cerebral changes in neurotransmitter metabolism (Ader et al., 2007). Pitychoutis et al. (2009) found that LPS treatment increased serotonergic activity in various regions of the limbic system, including the hypothalamus, hippocampus, and striatum, as well as prefrontal cortical areas in females. In contrast, LPS decreased serotonergic activity in many of these areas of male rats (Pitychoutis et al., 2009).

Further, LPS treatment altered the utilization of dopamine in the hippocampus and prefrontocortical brain regions in females (Pitychoutis et al., 2009).

Sex-based differences in neurochemical and behavioral responses after LPS may originate from the differences in the activity of the immune system between males and females (Fish 2008; Gaillard and Spinedi, 1998). It is known that the proportion and severity of infections are higher in males than females for viral, bacterial, fungal and parasitic diseases, except sexually transmitted infections (Klein 2000; Roberts et al., 2001). In general, females have more robust cell-mediated and humoral immune responses to antigenic challenges compared with males (Fish, 2008). In contrast, it is known that the female sex hormone 17-β estradiol (E2) is broadly anti-inflammatory and confers protection against sepsis. For example, males are known to be more susceptible to bacterial sepsis

- 105 - than females; in vitro experiments have shown that LPS-challenged macrophages derived from male mice produced higher levels of inflammatory cytokines than similarly treated female-derived macrophages (Marriott et al., 2006a, 2006b). This data is in line with clinical observations, which have shown that male sex is an independent risk factor for severe infections in surgical patients (Offner et al., 1999), and women have better survival rates during sepsis (Schröder et al., 1998). Further, female rodents in proestrus have significantly higher survival rates than males in a model of trauma-hemorrhage and sepsis (Choudhry et al., 2005).

Estrogen receptors (ERα and ERβ) are found in various immune cell types: T cells,

B cells, dendritic cells, macrophages, neutrophils and natural killer (NK) cells (Fish, 2008).

Within immune cells, ER complexes can affect transcription of various genes by binding to

ER sensitive segments on DNA (Fish 2008). The effects of estrogen on the innate immune responses are primarily repressive; for example, estrogens decrease the cytotoxicity of NK cells (Hao et al., 2007), are negative regulators of CD4 (+) T-cell-derived tumor-necrosis factor (TNF), inhibit CD16 expression in monocytes and macrophages - leading to the reduction in pro-inflammatory cytokines IL-1, IL-6 and TNF (Fish, 2008). Estrogen also up- regulates the production of nitric-oxide synthase (NOS) and nitric oxide in neutrophils (Fish,

2008). Further, E2 exerts a biphasic effect on T helper-cell polarization: low levels promote

T helper 1-cell differentiation and higher levels promote T helper 2-cell polarization, with consequent impact on the production of cytokines that are associated with each of the T helper-cell populations (Fish, 2008). Suppression of T helper 1-cell responses by estrogens results in decreased disease activity in patients with rheumatoid arthritis (Fish, 2008).

Various autoimmune diseases affect more women than men, such as systemic lupus erythematosus (SLE), multiple sclerosis, rheumatoid arthritis and scleroderma

- 106 - (Lockshin 2006; Whitacre 2001). Estrogen decreases the negative selection of naive B cells, enhances the survival of autoreactive B cells (Fish, 2008). These B-cell effects might contribute to the increased incidence of many autoimmune diseases in women (Fish, 2008).

Another factor that is important to mention is the role of sex chromosomes. The X chromosome encodes approximately 1,100 genes, some of which may partly underlie sex- based differences in immune responses (Fish, 2008). For example, a mutation in a γ- chain subunit protein - a significant component of cytokine receptors for interleukins 2, IL-4, IL-7,

IL-9, IL-15 and IL-21 - results in X linked severe combined immunodeficiency (XSCID)

(Schmalstieg and Goldman, 2002). Males are more vulnerable to X-linked diseases relative to females – due to the fact they only have a single X chromosome, whereas females have two. In this way, females experience the benefit of X-inactivation, which results in cellular mosaicism (Migeon et al., 2006).

Evolutionary arguments may also help explain sex-based differences in behavior.

For example, Darwin's theory of sexual selection predicted that behavioral strategies in coping with environmental challenges would differ in males and females when a discrepancy in parental investment exists (Palanza and Parmigiani, 2017). For example, in mammalian species, the mother typically invests more in the offspring than the father. Based on this theory, each sex would possess a behavioral repertoire designed to deal with the adaptive challenges that it has faced over evolutionary history (Palanza and Parmigiani, 2017).

Behavioral systems for stress coping, or in response to pathogenic insults, may also be a target for sexual selection, and therefore be differentially expressed in males and females

(Palanza and Parmigiani, 2017). Darwin's theory might help explain the common observation that females of reproductive age (15-44y) experience MDD at twice the rate of males (Fiske et al., 2009; Seedat et al., 2009a, 2009b).

- 107 - The pathogen-host defense theory of depression (PATHOS-D) posits that ancestral females developed depression during their reproductive years because depressive symptoms promoted behaviors that decreased the risk of pathogen exposure and provided increased protection from pathogens in ancestral environments (Raison and Miller, 2016).

The PATHOS-D theory also states that the human inflammatory response promoted our ancestral survival by encouraging individuals – particularly females – to endure the burdens of living in a group (Raison and Miller, 2017).

A study by Moieni and colleagues (2015b) showed that females responded to an injection of endotoxin (LPS) with feelings of social isolation and depression, which correlated with the level of peripheral cytokines. Ostracism posed a severe threat to survival for ancestral humans (Raison and Miller, 2017). An ostracism-detection mechanism might have affected females because their children would be less likely to survive without them (Sear and Mace, 2008).

- 108 - 4.3 How Good is the Model?

Rodent models of human mental illnesses are assessed for their quality and validity by meeting a few criteria: the model has symptomatic overlap with the disorder (also known as

'face validity'); the model has similar causative factors to the disorder ('etiological validity'); the mood symptoms are ameliorated by available medications for the disorder ('predictive validity'); and the model has similar postmortem changes found in disorder ('pathological validity') (Krishnan and Nestler, 2011). The Porsolt forced swim test (Porsolt et al., 1977a,

1977b) is the most commonly used test to measure antidepressant efficacy in rodents

(Krishnan and Nestler, 2011). During the test, a mouse is placed in a tank of water, and following a period of swimming and climbing the animal assumes an immobile posture

(Krishnan and Nestler, 2011). The time the animal spends immobile is used to infer

"depressive-like" behavior and is a measure of "behavioral despair" that the animal experiences, as though they have given up hope on escaping the tank (Cryan et al., 2005a;

Lucki et al., 2001).

Regarding meeting the markers of validity for a behavioral test, the FST has both good and bad qualities as a putative test for "depressive-like" behaviors in rodents. The FST is thought to have a high level of predictive validity because it is sensitive to compounds that are effective in humans as antidepressants and insensitive to compounds that are not effective (McKinney and Bunney, 1969; Willner, 1984). However, the FST in-and-of-itself is not considered to have high face validity with human depression. There is little similarity between the clinical symptoms of depression in humans and the behaviors measured in the test (Commons et al., 2017). Others have argued that the FST is incapable of measuring

"depressive-like behaviors" at all; depression is a subjective human emotional state, and subjective emotion is something we will simply never know about regarding rodents

(Commons et al., 2017). It has been argued that the FST measures more "stress coping"

- 109 - strategies than depressive-like behaviors (Commons et al., 2017). Further, regarding etiological validity, an assessment of the FST has limited values, since the etiology of depression is incompletely understood (Krishnan and Nestler, 2011).

In our study, we did not use the FST as a model of depression, but as a mechanism by which to measure the depressogenic effects of lipopolysaccharide (LPS) – a paradigm that is more in line with the FST's strength as a validation model. Moreover, we verified the results of FST with a separate behavioral test that measured overall activity - a methodology that has been used previously in various studies (Gould, 2009; O'Connor et al., 2009).

Using an inflammatory stimulus such as LPS to model depression in rodents has applicability to real-world cases (Krishnan and Nestler, 2011). For example, depression is prevalent in medical illnesses, such as multiple sclerosis (MS); up to 50% of patients experience depression (Krishnan and Nestler, 2011). Therefore, the LPS model may share similar causative underpinnings with medical-related depression – making LPS an etiologically valid model. Further, the symptomatology induced by LPS in rodents is like sickness and depressive behaviors in humans during depressive episodes – such as anhedonia, fatigue, body weight changes, and reduced locomotor, exploratory, and social behaviors. Therefore, the LPS model may also have reasonable face validity with some forms of depression in humans. Further, chronic treatment with SSRI antidepressants reverses many of the behavioral effects of LPS, as well as LPS-induced changes in body temperature, adrenocortical activation, hypothalamic serotonin release, and the expression of splenic TNF-α mRNA (Dantzer, 2009). In this way, the LPS model of depression may also has reasonable predictive validity for human depression.

- 110 - 4.4 Can We Treat Mood Disorders with Anti-Inflammatory Drugs?

As we have shown in Chapter 3, inflammation can cause depressive symptoms in male mice. Many groups have confirmed similar findings in male mice. Based on these and other pieces of evidence, many have posited that anti-inflammatory agents might treat depressive episodes.

A proof-of-concept study by Raison et al. (2014) examined the effect of the TNFα inhibitor infliximab on individuals with treatment-resistant depression. Their study included sixty male and female outpatients (25-60 years old) with MDD (they also included BD- depressed) (Raison et al., 2014). Although this study did not report on sex-based differences in response to infliximab, they did indicate that the patient sample was majority female (66%)

(Raison et al., 2014). The authors found no overall difference in depressive scores (HAM-

D) between treatment groups across the 12-week study (Raison et al., 2014). However, further analysis revealed that HAM-D scores improved in a subset of participants with higher baseline C-reactive protein (CRP) concentrations (Raison et al., 2014). Therefore, the authors concluded that TNF-α antagonism does not have generalized efficacy but may help patients with high baseline inflammatory biomarkers (Raison et al., 2014).

A more recent study by Husain et al. (2017a, 2017b) examined the efficacy of minocycline – a commonly used antibiotic – as an adjunct treatment in forty-one patients with treatment-resistant MDD (TRD). This study did not attempt to describe sex-based differences in response to minocycline; however, they reported the cohort to be 45% male in the placebo group, and 55% male in the treatment group (Husain et al., 2017a, 2017b).

Overall the authors found a significant improvement in HAM-D scores in the treatment group by 12 weeks (Husain et al., 2017a, 2017b). They concluded that minocycline may be helpful in TRD, but this study needs replication in a larger sample (Husain et al., 2017).

- 111 - A similar study by Soczynska et al. (2017) aimed to determine if adjunctive minocycline mitigated depressive symptoms and improved cognitive function in individuals with BD I/II. Using a small sample size of 27 participants, they found that adjunctive minocycline reduced depressive symptom severity on the Montgomery-Åsberg Depression

Rating Scale (MADRS) and HAM-D (Soczynska et al., 2017). They did not report on sex- based differences, although the cohort was 51.9% female (Soczynska et al., 2017).

A meta-analysis by Köhler et al. (2014) aimed to systematically review the antidepressant effects of anti-inflammatory interventions, and the possible adverse effects of these interventions (e.g., gastrointestinal or cardiovascular events, or infections). The authors identified ten trials that evaluated the use of NSAIDs – with 4258 participants total, and four trials that investigated cytokine inhibitors, with 2004 participants total (Köhler et al.,

2014). Their analysis suggested that anti-inflammatory treatment, in particular the selective cyclooxygenase two inhibitor celecoxib, decreases depressive symptoms without increased risks of adverse effects (Köhler et al., 2014). The authors concluded that although there is heterogeneity in literature, their study supports a proof-of-concept regarding the use of anti- inflammatory treatments for depression (Köhler et al., 2014). Further, there may be a benefit to analyzing subgroups of people who may benefit more from these interventions (Köhler et al., 2014).

There is promising evidence of anti-inflammatories as an adjunctive treatment for depressive episodes in human studies. However, there is conflicting evidence in rodent models. Vogt et al. (2016) examined the effect of minocycline on behavioral correlates of mood disorders in the forced swim test and open field test. They used adult male C57BL/6 mice – the most widely used strain in pharmacological and genetic animal models (Vogt et al., 2016). They found that minocycline did not improve depressive behavior in the FST in contrast to imipramine, which reduced depression in the mice (Vogt et al., 2016). The

- 112 - authors also found that minocycline did not reduce anxiety in the male mice, in contrast to diazepam, which increased exploratory behavior in the males (Vogt et al., 2016). In line with this, a study by Warner-Schmidt et al. (2011) found that ibuprofen (an NSAID) antagonized the effects of selective serotonin reuptake inhibitors (SSRIs) in adult male mice. The furthered this finding by extracting human data from the "sequenced treatment alternatives to relieve depression" (STAR*D) dataset. The authors examined the effect of NSAID treatment on remission within the first level of the STAR*D treatment sequence, in which the study participants were treated with citalopram (Warner-Schmidt et al., 2011).

Interestingly, they found that of those subjects who took an NSAID at least once during the study, 45% were in remission and 55% were treatment resistant (Warner-Schmidt et al.,

2011). Of those subjects who did not take an NSAID, 55% were in remission, and 45% being treatment resistant (Warner-Schmidt et al., 2011). The authors found that a higher percentage of participants were treatment resistant to citalopram if they had at some point taken an NSAID (Warner-Schmidt et al., 2011).

There are a few critical factors that might explain the discrepancy in some of the literature regarding the effectiveness of anti-inflammatory agents to treat depressive symptoms. First, the three agents discussed above - infliximab, minocycline, and NSAIDs - have different mechanisms of action. An explanation for this discrepancy may lie in the target of ibuprofen versus celecoxib; most common NSAIDs (such as aspirin and ibuprofen) inhibit the activity of cyclooxygenase-1 (COX-1) and cyclooxygenase-2 (COX-2). Another limitation of the literature is the lack of female rodents in preclinical studies using NSAIDs.

The vast majority of the clinical studies used a mixture of men and women, but this was hardly ever analogized in mouse models.

- 113 - 4.5 Is Zinc a Good Biomarker?

Biomarkers, the portmanteau of "biological markers," are objective signs of a medical state, which can be measured reproducibly and accurately in a patient (Strimbu and Tavel, 2011).

The National Institutes of Health (NIH) define a biomarker as "a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention" (Strimbu and Tavel, 2011). Examples of biomarkers include a wide array of measures, such as blood sugar or hemoglobin A1c, which are used as diagnostic biomarkers to identify patients with

Type 2 diabetes mellitus (Strimbu and Tavel, 2011). Genetic markers are biomarkers – they are commonly used to distinguish responders and non-responders to cancer treatments

(Strimbu and Tavel, 2011).

Biomarkers could serve several purposes in BD research – they could function as prognosticators, diagnostic tools, or inform us as to the underlying pathophysiology of BD.

In Chapter 2 we investigated associations between neopterin as an inflammatory marker and zinc as a constitutive co-factor in the inflammatory response and mood severity in BD.

We found that zinc correlated with depression severity in symptomatic women with BD.

Therefore, zinc may hold some potential as a prognostic biomarker for women with BD, although our results require replication in a larger sample. There is currently no diagnostic, prognostic or theranostic biomarker for BD. The ultimate challenge to the development of zinc as a biomarker – or developing any biomarker for that matter – is our understanding of the pathophysiology of BD. In this way, development of a biomarker needs constant reevaluation as our understanding of the pathophysiology of this heterogeneous disorder evolves (Strimbu and Tavel, 2011).

- 114 - 4.6 Female Research Subjects

In 2010, the Institute of Medicine, a part of the National Academy of Sciences, published a consensus report “Women's Health Research: Progress, Pitfalls, and Promise”, which called for the inclusion of female subjects in scientific research. The report concluded that limitations in the design, analysis and scientific reporting of health research had slowed progress in women's health. Further, it found that there has been inadequate enforcement and recruitment of women and reporting data by sex in both clinical trials and other research

(Wizemann et al., 2012).

In the follow-up to this, Wizemann et al. (2012) authored “Sex-Specific Reporting of

Scientific Research” – the most current workshop summary related to gender inclusivity in research – by the Institute of Medicine. This report stated that women are still underrepresented in clinical trials and explored historical and current events that have hindered or helped progress to advance the study of women (Wizemann et al., 2012).

There are significant limitations to policies currently in place; for example, only NIH- funded research must adhere to the inclusivity criteria (Wizemann et al., 2012). Also, the

NIH cannot require journals and editors to mandate inclusion of analyses based on sex

(Wizemann et al., 2012). Few journals have formally accepted a policy requiring clinical and epidemiological studies to be analyzed based on gender (Wizemann et al., 2012).

There is an urgent need to identify biologic and physiologic differences between the sexes to inform diagnosis and treatments; there are more women (50.7%) in the US than men (US Census, 2010). Women live longer than men on average – 80.7 versus 74.8 years

(Wizemann et al., 2012). However, women respond more slowly and are less likely to achieve an optimal response to treatment for depression (Wizemann et al., 2012). The under-representation of females in clinical studies has resulted in the disparity in both the understanding and the treatment of diseases in the sexes (Fish, 2008).

- 115 - 4.7 Conclusions

My thesis has challenged preconceived notions, and generated novel hypotheses in an understudied area of mood disorders research. These ideas are discussed below.

1. Our work will provide insight into the pathophysiology of depressive episodes,

treatment for mood disorders, and the sex-based differences that may exist therein.

2. The mouse model and existing data suggest that males and females differ in their

behavioral responses to lipopolysaccharide. The data also indicate correlations

between behavior and hippocampal mRNA, suggesting perhaps a difference in

underlying pathophysiology between the sexes. These data may impact research in

mood disorders, plus a variety of fields where medical illnesses are tied to

depressive states.

3. The mouse studies suggest that females are resilient to LPS-induced depressive-

like behavior, which is supported by literature (Painsipp et al., 2009; Pitychoutis et

al., 2011). The mouse data also indicate that over the long-term, female mice are

not susceptible to hippocampal damage caused by LPS. Future work should aim to

elucidate the mechanism of this protective effect in females.

4. The longitudinal mouse studies indicated that markers of hippocampal apoptosis

were increased in males after LPS, suggesting that males may be susceptible to

cognitive impairment associated with systemic inflammation. A future study should

examine cognitive impairment between males and females in this model.

5. We have shown that immune/inflammatory markers correlate to mood severity in

BD differently between men and women. The data suggest that men and women

may benefit from the use of different biomarkers to predict mood severity in mixed

episodes. These data will provide insight into biomarker development, for theranostic

markers, which are essential for individualized treatment.

- 116 - 6. Our studies have also supported the importance of zinc in the pathophysiology in depressive moods, particularly in females. In our clinical work, we have shown that peripheral zinc is reduced in women and men with symptomatic BD, and that peripheral zinc positively predicted depression severity in women. Our mouse study revealed that males given LPS also reduce peripheral zinc, which is in line with literature. However, female mice given LPS appear to increase peripheral zinc.

Further, LPS caused labile zinc to be diminished in the female hippocampus, which was accompanied by a concomitant increase in metallothionein-II transcript levels in females. Therefore, we propose that zinc may have an important protective role in females. We hypothesize that peripheral inflammation may initiate an alteration in zinc utilization in females, perhaps to make this “antioxidant ion” more accessible.

Future work in this model should explore the potential protective mechanism of the zinc against oxidative stress.

- 117 - 4.8 References

Ader R, Kelley KW (2007). A global view of twenty years of Brain, Behavior, and Immunity. Brain Behav Immun 21(1): 20-22. Can A, Dao DT, Arad M, Terrillion CE, Piantadosi SC, Gould TD (2012). The mouse forced swim test. J Vis Exp(59): e3638. Cannizzaro C, Martire M, Steardo L, Cannizzaro E, Gagliano M, Mineo A, et al. (2002). Prenatal exposure to diazepam and alprazolam, but not to zolpidem, affects behavioural stress reactivity in handling-naïve and handling-habituated adult male rat progeny. Brain Res 953(1-2): 170-180. Choudhry MA, Schwacha MG, Hubbard WJ, Kerby JD, Rue LW, Bland KI, et al. (2005). Gender differences in acute response to trauma-hemorrhage. Shock 24 Suppl 1: 101-106. Commons KG, Cholanians AB, Babb JA, Ehlinger DG (2017). The Rodent Forced Swim Test Measures Stress-Coping Strategy, Not Depression-like Behavior. ACS Chem Neurosci 8(5): 955-960. Cray C, Zaias J, Altman NH (2009). Acute phase response in animals: a review. Comp Med 59(6): 517-526. Cryan JF, Holmes A (2005a). The ascent of mouse: advances in modelling human depression and anxiety. Nat Rev Drug Discov 4(9): 775-790. Curtis AL, Bethea T, Valentino RJ (2006). Sexually dimorphic responses of the brain norepinephrine system to stress and corticotropin-releasing factor. Neuropsychopharmacology 31(3): 544-554. Dalla C, Pitychoutis PM, Kokras N, Papadopoulou-Daifoti Z (2011). Sex differences in response to stress and expression of depressive-like behaviours in the rat. Curr Top Behav Neurosci 8: 97-118. Dantzer R (2009). Cytokine, sickness behavior, and depression. Immunol Allergy Clin North Am 29(2): 247-264. Drossopoulou G, Antoniou K, Kitraki E, Papathanasiou G, Papalexi E, Dalla C, et al. (2004). Sex differences in behavioral, neurochemical and neuroendocrine effects induced by the forced swim test in rats. Neuroscience 126(4): 849-857. Fish EN (2008). The X-files in immunity: sex-based differences predispose immune responses. Nat Rev Immunol 8(9): 737-744. Fiske A, Wetherell JL, Gatz M (2009). Depression in older adults. Annu Rev Clin Psychol 5: 363-389. Gaillard RC, Spinedi E (1998). Sex- and stress-steroids interactions and the immune system: evidence for a neuroendocrine-immunological sexual dimorphism. Domest Anim Endocrinol 15(5): 345-352. Galea LA, McEwen BS, Tanapat P, Deak T, Spencer RL, Dhabhar FS (1997). Sex differences in dendritic atrophy of CA3 pyramidal neurons in response to chronic restraint stress. Neuroscience 81(3): 689-697. Gandhi M, Aweeka F, Greenblatt RM, Blaschke TF (2004). Sex differences in pharmacokinetics and pharmacodynamics. Annu Rev Pharmacol Toxicol 44: 499- 523. Handa RJ, Burgess LH, Kerr JE, O'Keefe JA (1994). Gonadal steroid hormone receptors and sex differences in the hypothalamo-pituitary-adrenal axis. Horm Behav 28(4): 464-476. Husain MI, Chaudhry IB, Hamirani MM, Minhas FA, Kazmi A, Hodsoll J, et al. (2017a). Minocycline and celecoxib as adjunctive treatments for bipolar depression: a study

- 118 - protocol for a multicenter factorial design randomized controlled trial. Neuropsychiatr Dis Treat 13: 1-8. Husain MI, Chaudhry IB, Husain N, Khoso AB, Rahman RR, Hamirani MM, et al. (2017b). Minocycline as an adjunct for treatment-resistant depressive symptoms: A pilot randomised placebo-controlled trial. J Psychopharmacol 31(9): 1166-1175. KITAY JI (1961). Sex differences in adrenal cortical secretion in the rat. Endocrinology 68: 818-824. Klein SL (2000). The effects of hormones on sex differences in infection: from genes to behavior. Neurosci Biobehav Rev 24(6): 627-638. Kokras N, Antoniou K, Mikail HG, Kafetzopoulos V, Papadopoulou-Daifoti Z, Dalla C (2015). Forced swim test: What about females? Neuropharmacology 99: 408-421. Krishnan V, Nestler EJ (2011). Animal models of depression: molecular perspectives. Curr Top Behav Neurosci 7: 121-147. Köhler O, Benros ME, Nordentoft M, Farkouh ME, Iyengar RL, Mors O, et al. (2014). Effect of anti-inflammatory treatment on depression, depressive symptoms, and adverse effects: a systematic review and meta-analysis of randomized clinical trials. JAMA Psychiatry 71(12): 1381-1391. Lockshin MD (2006). Sex differences in autoimmune disease. Lupus 15(11): 753-756. Lucki I (2001). A prescription to resist proscriptions for murine models of depression. Psychopharmacology (Berl) 153(3): 395-398. Marriott I, Bost KL, Huet-Hudson YM (2006a). Sexual dimorphism in expression of receptors for bacterial lipopolysaccharides in murine macrophages: a possible mechanism for gender-based differences in endotoxic shock susceptibility. J Reprod Immunol 71(1): 12-27 Marriott I, Huet-Hudson YM (2006b). Sexual dimorphism in innate immune responses to infectious organisms. Immunol Res 34(3): 177-192. McKinney WT, Bunney WE (1969). Animal model of depression. I. Review of evidence: implications for research. Arch Gen Psychiatry 21(2): 240-248. Migeon BR (2006). The role of X inactivation and cellular mosaicism in women's health and sex-specific diseases. JAMA 295(12): 1428-1433. Moieni M, Irwin MR, Jevtic I, Olmstead R, Breen EC, Eisenberger NI (2015b). Sex differences in depressive and socioemotional responses to an inflammatory challenge: implications for sex differences in depression. Neuropsychopharmacology 40(7): 1709-1716. O'Connor JC, Lawson MA, André C, Moreau M, Lestage J, Castanon N, et al. (2009b). Lipopolysaccharide-induced depressive-like behavior is mediated by indoleamine 2,3-dioxygenase activation in mice. Mol Psychiatry 14(5): 511-522. O'Connor JC, Lawson MA, André C, Moreau M, Lestage J, Castanon N, et al. (2009c). Lipopolysaccharide-induced depressive-like behavior is mediated by indoleamine 2,3-dioxygenase activation in mice. Mol Psychiatry 14(5): 511-522. Offner PJ, Moore EE, Biffl WL (1999). Male gender is a risk factor for major infections after surgery. Arch Surg 134(9): 935-938; discussion 938-940. Painsipp E, Köfer MJ, Sinner F, Holzer P (2011). Prolonged depression-like behavior caused by immune challenge: influence of mouse strain and social environment. PLoS One 6(6): e20719. Palanza P, Parmigiani S (2017). How does sex matter? Behavior, stress and animal models of neurobehavioral disorders. Neurosci Biobehav Rev 76(Pt A): 134-143. Pinos H, Collado P, Rodríguez-Zafra M, Rodríguez C, Segovia S, Guillamón A (2001). The development of sex differences in the locus coeruleus of the rat. Brain Res

- 119 - Bull 56(1): 73-78. Pitychoutis PM, Nakamura K, Tsonis PA, Papadopoulou-Daifoti Z (2009). Neurochemical and behavioral alterations in an inflammatory model of depression: sex differences exposed. Neuroscience 159(4): 1216-1232. Porsolt RD, Bertin A, Jalfre M (1977a). Behavioral despair in mice: a primary screening test for antidepressants. Arch Int Pharmacodyn Ther 229(2): 327-336. Porsolt RD, Le Pichon M, Jalfre M (1977b). Depression: a new animal model sensitive to antidepressant treatments. Nature 266(5604): 730-732. Raison CL, Miller AH (2017). Pathogen-Host Defense in the Evolution of Depression: Insights into Epidemiology, Genetics, Bioregional Differences and Female Preponderance. Neuropsychopharmacology 42(1): 5-27. Ranjbar E, Kasaei MS, Mohammad-Shirazi M, Nasrollahzadeh J, Rashidkhani B, Shams J, et al. (2013). Effects of zinc supplementation in patients with major depression: a randomized clinical trial. Iran J Psychiatry 8(2): 73-79. Roberts CW, Walker W, Alexander J (2001). Sex-associated hormones and immunity to protozoan parasites. Clin Microbiol Rev 14(3): 476-488. Schmalstieg FC, Goldman AS (2002). Immune consequences of mutations in the human common gamma-chain gene. Mol Genet Metab 76(3): 163-171. Schröder J, Kahlke V, Staubach KH, Zabel P, Stüber F (1998). Gender differences in human sepsis. Arch Surg 133(11): 1200-1205. Seale JV, Wood SA, Atkinson HC, Bate E, Lightman SL, Ingram CD, et al. (2004a). Gonadectomy reverses the sexually diergic patterns of circadian and stress- induced hypothalamic-pituitary-adrenal axis activity in male and female rats. J Neuroendocrinol 16(6): 516-524. Seale JV, Wood SA, Atkinson HC, Harbuz MS, Lightman SL (2004b). Gonadal steroid replacement reverses gonadectomy-induced changes in the corticosterone pulse profile and stress-induced hypothalamic-pituitary-adrenal axis activity of male and female rats. J Neuroendocrinol 16(12): 989-998. Seedat S, Scott KM, Angermeyer MC, Berglund P, Bromet EJ, Brugha TS, et al. (2009a). Cross-national associations between gender and mental disorders in the World Health Organization World Mental Health Surveys. Arch Gen Psychiatry 66(7): 785-795. Seedat S, Scott KM, Angermeyer MC, Berglund P, Bromet EJ, Brugha TS, et al. (2009b). Cross-national associations between gender and mental disorders in the World Health Organization World Mental Health Surveys. Arch Gen Psychiatry 66(7): 785-795. Soczynska JK, Kennedy SH, Alsuwaidan M, Mansur RB, Li M, McAndrews MP, et al. (2017). A pilot, open-label, 8-week study evaluating the efficacy, safety and tolerability of adjunctive minocycline for the treatment of bipolar I/II depression. Bipolar Disord 19(3): 198-213. Tonelli LH, Holmes A, Postolache TT (2008). Intranasal immune challenge induces sex- dependent depressive-like behavior and cytokine expression in the brain. Neuropsychopharmacology 33(5): 1038-1048. Vogt MA, Mallien AS, Pfeiffer N, Inta I, Gass P, Inta D (2016). Minocycline does not evoke anxiolytic and antidepressant-like effects in C57BL/6 mice. Behav Brain Res 301: 96-101. Warner-Schmidt JL, Vanover KE, Chen EY, Marshall JJ, Greengard P (2011). Antidepressant effects of selective serotonin reuptake inhibitors (SSRIs) are attenuated by antiinflammatory drugs in mice and humans. Proc Natl Acad Sci U S

- 120 - A 108(22): 9262-9267. Whitacre CC (2001). Sex differences in autoimmune disease. Nat Immunol 2(9): 777-780. Willner P (1984c). The validity of animal models of depression. Psychopharmacology (Berl) 83(1): 1-16. Wizemann TM (ed) (2012). SEX-SPECIFIC REPORTING OF SCIENTIFIC RESEARCH: A WORKSHOP SUMMARY. Institute of Medicine of the National Academies. 500 Fifth Street, NW • Washington, DC 20001. THE NATIONAL ACADEMIES PRESS: 500 Fifth Street, NW • Washington, DC 20001.

Vita Caitlin E. Millett

Education • Ph.D. Neuroscience, Pennsylvania State University College of Medicine 2018 • B.S. Biology, Simmons College, Boston MA 2012

Publications 1. Millett, C.E., Mukherjee, D., Reider, A., Can, A., Groer, M., Fuchs, D., Postolache, T., Kelleher, S., Saunders, E.F.H. Peripheral zinc and neopterin concentrations are associated with mood severity in a gender-specific manner. Psychiatry Research, Volume 255, 52 - 58. http://dx.doi.org/10.1016/j.psychres.2017.05.022 2. Mukherjee, D., Krishnamurthy, V.B., Millett, C.E., Reider, A., Can, A., Groer, M., Fuchs, D., Postolache, T., Saunders, E.F.H. Total Sleep Time and Kynurenine Metabolism Associated With Mood Symptom Severity in Bipolar Disorder. Bipolar Disorders. DOI: 10.1111/bdi.12529.

Conference Presentations 1. Millett, C., Kelleher, S., Saunders, E.F.H.,” Short and Long-Term Effects of LPS on Behavior and Oxidative Stress in Male and Female C57bl/6 Mice.” Poster, American Society of Clinical Psychopharmacology (ASCP, 2017) 2. Millett, C., Mukherjee, D., Reider, A., Can, A., Groer, M., Fuchs, D., Postolache, T., Kelleher, S., Saunders, E.F.H. Neopterin and Zinc Differentially Predict Mood Severity in Men and Women with Bipolar Disorder.” Poster, Society of Biological Psychiatry (SOBP, 2017) 3. Millett, C., Kelleher, S., Saunders, E.F.H.,” The Intersection of Inflammation and Mood: Sex Differences in a Model of Depression.” Oral, The Graduate Student Research Forum, PSU College of Medicine (2017)

Published Abstracts 1. Millett, C.E., Mukherjee, D., Reider, A., Can, A., Groer, M., Fuchs, D., Postolache, T., Kelleher, S., Saunders, E.F.H. Neopterin and zinc differentially predict mood severity in men and women with Bipolar Disorder. Poster 297, Proceedings from the Society of Biological Psychiatry (2017), doi.org/10.1016/j.biopsych.2017.02.311 2. Millett, C., Reider, A., Phillips, B., Kelleher, S., Saunders, E.F.H. Zinc as a mediator of inflammation in the brain: implications for mood in bipolar disorder? Poster M110, Proceedings from the American College of Neuropsychopharmacology (2015), doi:10.1038/npp.2015.325

Awards • Karl H. Beyer Jr., M.D., Ph.D. Scholarship- for academic excellence & leadership (2017)