Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mitochondrial complex I functionality and protein oxidative damage in lymphoblasts from lithium responsive… Zborovszky, Camila Carolina 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_spring_zborovszky_camila.pdf [ 1.56MB ]
Metadata
JSON: 24-1.0071662.json
JSON-LD: 24-1.0071662-ld.json
RDF/XML (Pretty): 24-1.0071662-rdf.xml
RDF/JSON: 24-1.0071662-rdf.json
Turtle: 24-1.0071662-turtle.txt
N-Triples: 24-1.0071662-rdf-ntriples.txt
Original Record: 24-1.0071662-source.json
Full Text
24-1.0071662-fulltext.txt
Citation
24-1.0071662.ris

Full Text

MITOCHONDRIAL COMPLEX I FUNCTIONALITY AND PROTEIN OXIDATIVE DAMAGE IN LYMPHOBLASTS FROM LITHUIM RESPONSIVE PATIENTS WITH BIPOLAR DISORDER, AFFECTED AND UNAFFECTED RELATIVES by CAMILA CAROLINA ZBOROVSZKY B.Sc., The University of British Columbia, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Neuroscience)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) March 2011  © Camila Carolina Zborovszky, 2011  Abstract Background: Growing evidence has demonstrated that mitochondria underline the neurobiology of bipolar disorder (BD). The search for peripheral biomarkers for this debilitating disorder continues. Therefore, the objective of this study is to investigate the involvement of complex I functionality in transformed lymphoblasts from lithium responsive patients with BD and its relation with increased oxidative damage, as well as to verify the response of this factor to stress induced by low levels of glucose. Objective: Confirm the alterations in complex I activity as well as levels of the electron transport chain complex I subunit, NDUFS7, in addition to oxidative damage to mitochondrial proteins, by increased levels of protein carbonyls and 3-nitrotyrosine content, in lymphoblasts from lithium responsive patients with BD. The response of lymphoblasts to low glucose stress will be evaluated by measuring complex I activity as well as NDUFS7 levels and protein oxidative damage. Design: Complex I activity was measured by spectrophotometry and tyrosine-nitration induced damage was assessed by measuring levels of 3-nitrotyrosine using a competitive enzyme immunoassay. Immunoblotting was used in order to measure NDUFS7 protein levels in addition to protein carbonyl levels. All assays were carried out in cell pellet fractions from lymphoblasts, except for complex I activity which was carried out in mitochondrial fractions from lymphoblasts. Patients: We studied lymphoblasts from patients (14 with BD, and 14 relatives of BD patients affected with a psychiatric illness) and from non-psychiatric comparison controls (N=15) as well as 16 relatives of BD patients unaffected by a psychiatric illness. Results: Our results showed that levels of complex I activity in lymphoblasts differed significantly between the groups, with the lowest values for control subjects, and highest in unaffected and affected relatives of patients with BD, under normal glucose conditions. We did not find significant differences between the groups in NDUFS7, protein carbonyl, and 3-nitrotyrosine levels, nor did we find any correlation between complex I activity and NDUFS7, protein carbonyl, and 3-nitrotyrosine levels, for both treatment conditions (normal and low glucose). Conclusions: These findings suggest an up-regulation of peripheral complex I activity in patients with BD and their relatives, as a potential compensatory mechanism that is able to prevent protein oxidative damage induced by complex I dysfunctionality . Future studies, evaluating the potential role of lithium in peripheral cells targeting electron transport chain sites such as complex I, are necessary in order to gain further insight into the mechanisms in which cells are able to prevent the oxidative damage characteristic of the disorder.  ii  Preface The lymphoblast cell lines used in this study, transformed from lymphocytes taken from human subjects’ blood samples, had the approval of UBC’s clinical research ethics board (CREB). The lymphoblasts are part of a genetic study on lithium responsive bipolar disorder patients whose principal investigators are Dr. Martin Alda M.D., F.R.C.P.C. (Dalhousie University) and my supervisor, Dr. Trevor Young M.D., Ph.D., F.R.C.P.C. (University of British Columbia, University of Toronto). UBC’s clinical research ethics board found the project to be acceptable on ethical grounds for research involving human subjects, and was given the following UBC CREB number: H07-01926. The current UBC CREB approval for this study expires on August 10, 2011.  iii  Table of Contents Table of Contents Abstract..........................................................................................................................................ii Preface .........................................................................................................................................iii Table of Contents .........................................................................................................................iv List of Tables ................................................................................................................................ v List of Figures ...............................................................................................................................vi Acknowledgements......................................................................................................................vii Dedication ................................................................................................................................... viii 1 Introduction ............................................................................................................................... 1  2  3  1.1  Bipolar disorder: description and pathology ..................................................................... 1  1.2  The genetics and neurobiology of bipolar disorder .......................................................... 4  1.3  Mitochondrial dysfunction and oxidative stress in bipolar disorder .................................. 6  1.4  Lithium neuroprotective effects ...................................................................................... 15  1.5  Objectives and hypothesis ............................................................................................. 21  Methods ................................................................................................................................ 24 2.1  Lymphoblast samples .................................................................................................... 24  2.2  Cell culture ..................................................................................................................... 26  2.3  Mitochondrial and cell pellet extraction .......................................................................... 27  2.4  Complex I activity ........................................................................................................... 30  2.5  NDUFS7 levels............................................................................................................... 33  2.6  Protein oxidation ............................................................................................................ 34  2.7  Tyrosine nitration-induced damage ................................................................................ 37  2.8  Statistical methods ......................................................................................................... 40  Results .................................................................................................................................. 41 3.1  Demographic variables .................................................................................................. 41  3.2  Mitochondrial complex I functionality and protein damage ............................................ 42  4 Discussion ............................................................................................................................. 48 References ................................................................................................................................. 66  iv  List of Tables Table 1  Volume of reagents added for complex I activity measurements……………31  Table 2  Demographic and clinical data for lymphoblasts in patients and controls…. 41  v  List of Figures Figure 1  Cellular processes leading to oxidative damage………………………………10  Figure 2  Kinetic activity measured in order to determine complex I functionality…….32  Figure 3  Complex I activity…………………………………………………………………42  Figure 4  NDUFS7 levels……………………………………………………………………43  Figure 5  Protein carbonyl levels…………………………………………………………...44  Figure 6  3-Nitrotyrosine content…………………………………………………………...45  Figure 7  Correlations for normal glucose treatment……………………………………..46  Figure 8  Correlations for low glucose treatment………………………………………....47  vi  Acknowledgements Great opportunities are made possible by wonderful people coming together at the right time and place in life. Under the supervision of Dr. L. Trevor Young, I have been granted the opportunity to broaden my knowledge of research in the field of psychiatry, inspiring me to continue working towards a better understanding of bipolar disorder, making the past two years of my life a memorable experience. Therefore, I extend my gratitude towards Dr. Young and his wonderful lab team, in addition to Dr. Jun-Feng Wang and Dr. Martin Alda. Dr. Ana C. Andreazza, you were my graduate school survival guide with the magic solution to all problems. Thank you for your great smile, kindness and understanding. It was a pleasure working with you. Dr. Li Shao, I thank for your invaluable help, and for your kind words of advice. Lastly, to J. Felipe Montenegro S., you give me the strength and motivation to be a better person every day. Thank you my love.  vii  Dedication  To JFMS  viii  1 Introduction 1.1 Bipolar disorder: description and pathology Bipolar disorder (BD) is a chronic psychiatric illness affecting around 1 to 5% of the world population (Konradi et al., 2004), associated with impaired social function and decreased quality of life. The onset of BD is typically during young adulthood, although it is not uncommon to have people displaying characteristics of the disorder as adolescents. BD is characterized by recurrent and persistent episodes of mania and depression. The Diagnostic and Statistical Manual of Mental disorders, fourth edition (DSM-IV), describes some of the principal characteristics of a manic episode as having feelings of grandiosity, flight of ideas, elated mood, hyperactivity, talkativeness, increased energy, lack of sleep, as well as excessive involvement in pleasurable activities with a high risk for negative consequences. The mood disturbances associated with mania can impair occupational and social functioning and even damage existing relationships with others. In extreme cases, this state of elevated mood can induce psychotic symptoms such as hallucinations or delusions that may require hospitalization in order to prevent self-harm, or harm to others, due to irrational or dangerous behaviour, affecting multiple aspects of the patient’s life. A major depressive episode, on the other side of the manic spectrum of symptoms, is characterized by feelings of worthlessness, loss of interest, difficulty in concentrating, restlessness, appetite loss, sleep disturbance, psychomotor retardation, and in some cases, patients will manifest suicidal thoughts and tendencies (DSM-IV). It is estimated that about 80% of all suicide victims are severely depressed (Rosenzweig et al., 2005). 1  As with mania, depression is also capable of affecting an individual’s quality of life, impairing social function due to isolation and withdrawal from pleasurable activities usually shared with others. The duration of a manic or a major depressive episode varies across individuals, and many variables influence the cycling between these mood states, such as the type of BD diagnosis, response to medication, social support, effective psychotherapy and pharmacological treatment. Some patients exhibit rapid cycling in the disorder, where there can be at least four distinct cycles between mania and depression in one year. Some individuals even experience this rapid cycling several times in one day. Alcohol and drug abuse tend to complicate the treatment of patients with BD, in addition to the clinical course of the disease, as patients experiencing a major depressive episode may use alcohol or drugs to relieve the pain, and patients in the manic phase may crave drugs and alcohol as part of the arousal characteristic of a manic episode (Belmarker, 2004). The DSM-IV lists four groups of bipolar disorders: cyclothymia, bipolar disorder I (BDI), bipolar disorder II (BDII), and a variety of bipolar disorders that don’t fit the criteria for any of the other three (DSM-IV). A cyclothymic disorder can be described as a chronic state of mood disturbance lasting for at least two years, presenting episodes of hypomania and dysthymia, which are milder versions of mania and depression respectively. On the other hand, a patient with BDI will experience severe episodes of mania and depression. A patient can be diagnose with BDI after only one manic episode, although the most recent episode can be either a major depressive episode, a hypomanic episode or a mixed episode where both manic and depressive episodes occur almost daily for at least a week. BDII differs 2  from BD I in that the manic behaviour is manifested to a lesser degree, what is referred to as hypomania. A hypomanic state is characterized by the presence of elevated or irritable mood, without psychotic symptoms or any other behaviour that can be dangerous to oneself or to others, although the persistent and sudden mood changes in BDII can eventually affect the person’s ability to cope with everyday life. In turn, the mood cycles observed in all types of BD can severely impair an individual’s quality of life, depending on the severity of symptoms, ranging from out of character behaviours impairing social functioning to actions capable of threatening the individual’s life. Psychotic symptoms such as delusions and hallucinations that can be observed in patients with BD, can be indistinguishable from those observed in other psychiatric illnesses such as schizophrenia (SCZ), yet the nature of such psychosis is different: delusions and hallucinations in SCZ are associated with impairment of self awareness, whereas those of BD are associated with mood disturbances (Kato et al., 2007). Even after the mood disturbances subside with pharmacological treatment, recent studies have shown impairment in a variety of cognitive functions in patients with BD that continue to be present even after symptoms’ remission (Martinez-Aran et al., 2004; Robinson et al., 2006). Findings such as these ones suggest that BD could be linked to persistent cognitive impairment (Andreazza et al., 2008). Pharmacological treatments available for BD have augmented over the past two decades and include a diverse group of pharmacological agents spanning from lithium, the standard treatment for BD for over 50 years, to novel and effective antipsychotic agents (Bezchlibnyk and Young, 2002).  3  1.2  The genetics and neurobiology of bipolar disorder  Genetic factors are believed to play a role in the aetiology of BD, supported by twin, adoption and family studies (Kato et al., 2005), yet the results are not consistent enough to establish a genetic risk factor for the manifestation of the disorder (Kato, 2007). Nevertheless, some of the genes associated with BD that have generated research interest over the years have been those coding for cyclic-AMP responsive element binding (CREB) proteins. Cyclic-AMP, or cyclic adenosine monophosphate (cAMP), is part of a signal transduction pathway, responsible for transcription regulation. Ligand activation of G-protein coupled receptors on the cell surface triggers this cAMP pathway, a cascade of biochemical signalling events that terminate when CREB proteins are phosphorylated. In their phosphorylated state, CREB proteins bind to the cAMP responsive element (CRE) recognition sequence of cAMP responsive genes, in order to regulate gene transcription (Mamdani et al., 2008). A number of studies have concluded that lithium decreases CREB phosphorylation, which can lead to decreased DNA binding, and in turn, to altered expression of cAMP responsive genes (Bezchlibnyk and Young, 2002). Investigating CREB genes from a sample of lithium responsive patients with BD lead to the conclusion that, in fact, two single nucleotide polymorphisms in genes belonging to the CREB family are believed to be associated with BD and/or response to lithium (Mandami et al., 2008). Therefore, changes in gene expression are likely to play a role in the mechanism of action of lithium treatment, aiding in the possibility of understanding lithium’s mechanism of action as a mood stabilizer and the genetic factors influencing its pharmacological efficacy, in the near future. In addition, searching for the genetic basis of BD has been the focus of linkage 4  and genetic association studies for the past two decades (Kato, 2008), and will remain a priority since the lack of homogeneity of the disease makes it hard to identify the causative genes or genetic risk factors for the disorder (Kato, 2007). There are many questions that continue to be unanswered regarding the neurochemical mechanisms and neuropathologies that underlie the disorder, and more so, on how abnormalities in the brain translate to mood disturbances. Structural neuroimaging and post-mortem studies have highlighted anatomical and neuropathological abnormalities in patients with BD, such as amygdala enlargement, bilateral reduction in prefrontal cortex gray matter volume, decreased levels of markers for neuronal integrity such as N-acetyl-aspartate in the dorsolateral prefrontal cortex, decreased neuronal density in dorsolateral and orbital prefrontal cortical regions, and reduced glial cell density in frontolimbic brain regions (Rajkowska, 2002; Strakowski et al., 2005). BD is associated with a reduced number or density of neurons, in multiple brain regions such as the anterior cingulate cortex, prefrontal cortex, thalamic ventromedial nucleus, hypothalamic paraventricular nucleus, and ventrolateral subnucleus of the dorsal raphe (Cotter et al., 2002; Benes et al., 2001). Neuronal vulnerability to cellular stress is a powerful candidate that can account for the observed reduction in cell number and density observed in BD (Kato, 2008). The molecular basis for this vulnerability could be related to impaired calcium and neurotrophin signalling as well as endoplasmic reticulum stress response dysfunction and mitochondrial dysfunction (Kato, 2008). The neurobiology of BD is an emerging field of research, and accumulated evidence suggests that mitochondrial dysfunction and oxidative stress play an important role in the characterization of the disorder. Evidence for BD as a mitochondrial energy 5  metabolism disease includes decreased cellular pH as well as the presence of highenergy phosphates in the frontal and temporal lobes of patients with bipolar disorder (Kato T. et al., 2004). The mitochondrial dysfunction observed in these patients is hypothesized by some to be due to abnormal expression of nuclear or mitochondrial genes that code for mitochondrial proteins (Konradi et al., 2004). In fact, decreased expression of genes coding for the enzymatic complexes responsible for oxidative phosphorylation in mitochondria, was reported in the hippocampus of brain samples taken from patients with BD, as well as downregulation of proteasome degradation, a process dependent on the generation of ATP by mitochondria (Konradi et al., 2004). Combined lines of evidence point at a variety of neuropathologies, genetic studies and metabolism abnormalities in the attempt to explain the underlying cause for BD. Growing evidence has demonstrated that mitochondria underline the neurobiology of BD, and the search for improved pharmacological therapy, in addition to peripheral biomarkers for this debilitating disorder continues.  1.3  Mitochondrial dysfunction and oxidative stress in bipolar disorder  Mitochondria are membrane-enclosed organelles whose cellular role is to generate adenosine triphosphate (ATP) through oxidative phosphorylation, a process involving a set of reactions collectively known as the citric acid cycle or the Krebs cycle. Mitochondria require a large number of proteins which form different complexes (I-V) embedded in their inner membrane, in order to perform their dominant role in the cell: to oxidize the major products of cellular respiration. This oxidation is based on the transfer 6  of electrons through complexes I through V (complex V is also called ATP-synthase) of the mitochondrial electron transport chain (ETC), creating a proton gradient and generating energy in the form of ATP, essential for brain functioning. Electron carriers present in mitochondria, such as nicotinamide adenine dinucleotide (NAD+), capture the electrons lost when food materials are oxidized, in order to be reoxidized by oxygen in the ETC, producing large amounts of ATP (Halliwell and Gutteridge, 2007). In other words, after the flow of electrons through the complexes, ATP synthase (complex V) uses the energy lost by protons re entering the mitochondrial matrix, to create an electrochemical gradient in order to generate ATP. Oxidative phosphorylation begins in complex I. This large protein complex catalyzes the transfer of electrons from reduced nicotinamide adenine dinucleotide (NADH) to coenzyme Q (also called ubiquinone) (Halliwell and Gutteridge, 2007). Alongside this reaction, the translocation of protons across the mitochondrial inner membrane takes place, an event contributing to the electrochemical gradient required to produce ATP (Wang, 2007). Because NADH is oxidized to NAD+ by this large enzymatic complex, it is given the name NADH dehydrogenase. Complex I, or NADH dehydrogenase, consists of up to 45 different subunits, making it one of the largest and most complicated membrane protein complexes known (Brandt et al., 2003). In eukaryotic cells there are 14 principal complex I subunits , half of which are highly hydrophobic proteins, while the remaining seven subunits contain redox prosthetic groups consisting of a molecule of flavin adenine mononucleotide (FMN) and eight to nine iron-sulfur clusters (Brandt et al., 2003). Very little is known about the function of the 31 other subunits forming this important enzymatic complex. FMN is where electrons come in 7  from NADH, and serves as an electron converter between the electron donor (NADH) and the electron transferring iron-sulfur clusters of complex I. The mechanism outlining the conversion of redox energy for proton transport across the membrane into ATP is limited. Nevertheless, over the years the interest in understanding how this large subunit complex does this has lead to the proposal of three main mechanisms: redox linked proton pump, ligand conduction mechanisms, and energy transfer due to conformational changes in the enzymatic complex (Brandt et al., 2003). Therefore, mitochondrial complex I plays a crucial role in the generation of the ETC reduction potential gradient, by having electrons travel through the different complexes and translocating protons from the mitochondrial matrix to the intermembrane space. This, in turn, creates a pH difference and a charge difference across the inner mitochondrial membrane (Halliwell and Gutteridge, 2007), essential for the generation of the electrochemical gradient required to produce ATP. During the transfer of electrons through the different ETC complexes, single electrons can escape, and in the presence of molecular oxygen (O2), can result in a singleelectron reduction of O2, leading to the formation of a superoxide anion (O2•-). Inhibitors of electron transport such as rotenone, a compound targeting complex I, can increase mitochondrial O2•- production, by facilitating electron transfer to molecular oxygen and by increasing the levels of reduced electron carriers. Therefore, leakage of electrons to molecular oxygen to form O2•- can be due to a decrease in complex I activity and a dysfunctional ETC. Uncoupling agents such as 2,4-dinitrophenol with the ability to decrease proton gradient and accelerate electron transport are able to decrease O2•production (Halliwell and Gutteridge, 2007). 8  A dysfunctional ETC will generate excessive reactive oxygen species (ROS) such as O2•-, leading to oxidative damage (Wang, 2007). The major sites of O2•- formation in the ETC are within complexes I and III (Lenaz, 2001), as proven by previous work describing the involvement of complex I in ROS production (Takeshige and Minakami, 1979). Free radicals such as O2•- produced when electrons are leaked from complex I and complex III (Figure 1), are generated under physiological conditions during aerobic metabolism to serve important roles in physiological processes, and are inactivated by antioxidant enzyme systems after the body has used the required amount of such reactive molecules (Kuloglu et al., 2002). Most O2•- is released into the matrix and its most likely fate is dismutation to hydrogen peroxide (H2O2) by syperoxide dismutase (SOD), an antioxidant enzyme responsible for neutralizing the free radicals present in cells (Halliwell and Gutteridge, 2007). When free radicals are overproduced due to oxidative stress, as in the case of mitochondrial dysfunction at the level of the ETC, or when antioxidant defence systems such as SOD are not working properly, devastating chain reactions causing cellular atrophy or even death, are activated (Kuloglu et al., 2002). In addition, free radicals have been implicated in the pathogenesis of treatment complications in psychiatric disorders (Mahadik and Mukherjee, 1996), and increased SOD activity has been linked to SCZ and BD (Kuloglu et al., 2002). In the presence of ferrous iron (Fe2+), H2O2, the product generated by SOD from O2•-, can react to generate highly reactive hydroxyl radicals (OH•), a chemical process known as the Fenton reaction (H2O2 + Fe2+ → Fe3+ + OH- + OH•) (Andreazza et al., 2010).  9  Figure 1. Cellular processes leading to oxidative damage  CELL  •-  Abbreviations: CI, complex I; CII, complex II; CIII, complex III; CIV, complex IV; CV, complex V; O2 , superoxide; SOD, superoxide dismutase; NO•, nitric oxide; H2O2, hydrogen peroxide; ONOO–, 2+ peroxinitrite; Fe , ferrous iron; 1, amino acids lysine, proline, arginine and threonine; 2, amino acid tyrosine; OH•,
hydroxyl
radical;
3‐NT,
3‐nitrotyrosine;
C=O,
carbonyl
functional
groups.
  Proteins can react with OH•, and in the presence of catalysts such as iron (Fe2+) and copper (Cu2+), carbonyl groups can be introduced to lysine, proline, arginine and threonine residues, causing oxidative damage to proteins (Figure 1). These  10  modifications can inactivate not only key enzymes but various membrane signalling pathways (Naoi et al., 2005). Oxidative modifications of proteins can affect DNA binding related to transcription factors, enzymatic activity, and proteins’ susceptibility to proteolytic degradation (Rivett, 1986). Therefore, protein carbonyl groups have become the most widely studied marker for protein oxidation, and for identifying the role of ROS in the pathogenesis of aging and neurodegenerative diseases (Beal, 2002). Carbonyl functional groups can also be introduced into proteins by addition of 4-hydroxynonenal, a product of lipid peroxidation, via the Michael reaction (Beal, 2002). The rate of oxidation of proteins increases as a function of age, with dramatic effects in the last third of life, where about one in three proteins are affected. This increased oxidative damage could lead to physiological implications that can explain the pathogenesis of several neurodegenerative disorders (Beal, 2002), and could play a role in the characterization of BD (Wang, 2007). Alternatively, overproduction of O2•- can lead to the generation of reactive nitrogen species (RNS) and subsequent protein and DNA modifications, outside the mitochondria (Figure 1). Nitric oxide (NO•) is believed to be synthesized in microglia and astrocytes, and subsequently transported to neurons where it can react with O2•-, producing peroxinitrite (ONOO-), a very reactive nitrogen radical (Naoi et al., 2005). This RNS in turn, nitrates sulfhydryl (R-SH) and hydroxyl residues (R-OH) in the side chains of the amino acids cysteine, methionine, phenylalanine, and tyrosine. This modification inactivates not only cellular membrane function, but key enzymes (Beckman, 1996). Nitration of tyrosine residues by reaction with ONOO- yields 3-nitrotyrosine (3-NT), an additional marker for oxidative stress (Figure 1) (Halliwell, 1997). 11  Accumulation of ROS and RNS as a result of overwhelmed mitochondrial cytoplasm enzymatic and nonenzymatic antioxidant systems, are believed to be the cause of neuronal cell death, as they are capable of modifying biomolecules such as lipids, proteins, DNA and carbohydrates with the peroxidation products of lipid or carbohydrates (Naoi et al., 2005). Cell injury caused by damage to lipids, proteins or DNA, can trigger cell death by either apoptosis or necrosis. Apoptosis or programmed cell death usually ensues as a result of extensive DNA damage in order to avoid the risk of becoming a cancerous cell. Mitochondrial dysfunction and overproduction of ROS and RNS have emerged as powerful candidates for neurodegeneration, contributing to the energy depletion and oxidative damage observed in devastating disorders such as Parkinson’s disease and Alzheimer’s disease (Papa and Rockwell, 2008). Oxidative stress, the term used to describe and excess of ROS in relation to the available antioxidants, has even been linked to cancer, diabetes, heart failure and aging (Ricci et al., 2007). A variety of functions can be altered due to the accumulation of modified proteins, which can change the enzymatic and regulatory activity of specific proteins, as well as the balance between the generation of ROS-RNS and the degradation of modified protein, monitored by the levels of oxidized proteins. These modifications are believed to involve proteosome degradation systems. In addition, RNS can activate pro-apoptotic signalling pathways by triggering the opening of the mitochondrial permeability transition pore (mPTP) (Naoi et al., 2005). This pore has a voltage-dependent anion channel located between the mitochondrial outer and inner membrane, whose opening is regulated by both bcl-2 and the adenine nucleotide translocator, located in the outer and inner membrane respectively (Kroemer 12  et al., 1998). Bcl-2 is a member of the family of proteins that regulate apoptosis induced by stimuli such as oxidative stress. The maintenance of mitochondria-cytosolic coupling of oxidative phosphorylation, which prevents the opening of mPTP, is an event key in mediating the antiapoptotic function of bcl-2, which functions as an antioxidant both directly and indirectly (Naoi et al., 2005). There are many events that can cause the opening of the mPTP by ONOO-, such as nitration of tyrosine residues in the mPTP in order to release cytochrome-c (Okada et al., 1992). Alternatively, ONOO- might modify the bcl-2 family of proteins to change its localization and function (Cassina et al., 2000), by triggering the opening of mPTP, activating caspase-3, which initiates the proapoptotic pathway, inducing the translocation of glyceraldehydes-3-phosphate dehydrogenase (GPDH) into the cell nucleus, and the fragmentation of nuclear DNA (Naoi et al., 2005). The end result is irreversible DNA damage. After reviewing the mechanisms by which increased ROS and RNS lead to oxidative stress, and knowing that decreased complex I activity in mitochondria ETC is an important source for free radicals that can lead to the overproduction of O2•-, several lines of evidence have given mitochondrial dysfunction an important role in the pathogenesis of BD. Mitochondrial DNA mutations or polymorphisms have been hypothesized to alter brain energy metabolism leading to the onset of manic and depressive symptoms in BD. A defective metabolism can cause stress-induced energy failure in the brain, lactate accumulation, in addition to decreased intracellular pH and high-energy phosphates (Kato and Kato, 2000). Abnormal expression of mitochondrial genes coding for mitochondrial proteins can be attributed to mitochondrial dysfunction. DNA microarray analysis for the expression of nuclear messenger RNA (mRNA) that 13  codes for mitochondrial proteins such as the enzymatic complexes responsible for oxidative phosphorylation, is significantly decreased in BD, and proteosome degradation, an ATP-dependent process, is downregulated at the level of gene expression (Konradi et al., 2004; Iwamoto et al., 2004), strengthening the idea that a decrease in pH and high-energy phosphate levels in BD are due to mitochondrial dysfunction. Therefore, mitochondrial dysfunction associated with decreased oxidative phosphorylation, alters cellular metabolism, shifting towards anaerobic energy production, increasing lactate levels as well as pH. This metabolic shift can lead to the overproduction of mitochondrial ROS, whose major source are the ubiquinone sites in complexes I and III (Maher and Schubert, 2000). Using high-density complementary DNA spot microarrays, Sun et al reported alterations in the expression of 23 mitochondria-related genes in subjects with BD, including downregulation of 8 mitochondrial genes coding for proteins in the ETC such as NDUFS7 and NDUFS8 in complex I, UQCRC2 in complex III, COX5A and COX6C in complex IV, and ATP5C1, ATP5J and ATP5G3 in complex V. (Sun et al., 2006). In addition, DNA array analysis and real-time PCR data showed decreased expression of complex I subunit, NDUFS7, in the frontal cortex in bipolar disorder (Sun et al, 2006). The important role mitochondria play in energy metabolism influences the regulation of cell death and survival, and can be the cause of various diseases and disorders. Mitochondrial DNA has several characteristics that can contribute to its pathophysiological role in various diseases and neurodegenerative disorders such as diabetes mellitus, cardiomyopathy, Parkinson’s disease, Alzheimer’s disease, and other diseases related to age (Kato, 2000). These characteristics include a poor DNA 14  repairing system and susceptibility to somatic mutations (Ozawa, 1998). In addition, the brain damage observed in BD leading to neuronal death can be attributed to mitochondrial dysfunction. As a result, a properly functioning mitochondrion is essential for cellular function, and mitochondrial dysfunction has gained tremendous interest over the years in the field of BD research and will remain the focus of future studies.  1.4  Lithium neuroprotective effects  In addition to the genetic factors important in the aetiology of BD, patients with mood disorders have well characterized neuroplastic changes, such as reductions in hippocampal and cortical volume, glial and neuronal cell density in brain areas such as the anterior cingulate cortex and prefrontal cortex, as well as reduction in levels of brainderived neurotrophic factor (BDNF) in the hippocampus and prefrontal cortex (Sanacora, 2008). These changes can vary depending on the type of mood stabilizer used to treat the disorder. Clinical studies of patients with BD under pharmacological treatment that did not include lithium, showed significant reductions in glial cell density in the entorhinal cortex of the medial temporal lobe, when compared to patients under lithium treatment (Bowley et al., 2002). In addition to clinical analyses, in vitro and in vivo studies have concluded that lithium is neuroprotective, and one of the explanations for this phenomenon is related to the induction of the anti-apoptotic signalling pathway to promote cell survival, which includes upstream factors such as BDNF (Rowe and Chuang, 2004).  15  Besides its important role in transcription regulation, CREB plays a role in other cellular processes such as apopotosis. CREB is sensitive to various stimuli such as growth factors and stress, and can be regulated by BDNF (Bonni et al., 1999), as well as by MEK/ERK (Kopnisky et al., 2003) and GSK-3β (Hansen et al., 2004), regulatory enzymes in the cell survival signalling pathway. CREB is considered neuroprotective as it is capable of increasing the expression of anti-apoptotic proteins, including those of the BDNF family of proteins (Shieh et al., 1998). CREB regulates BDNF expression, and BDNF is capable of inducing CREB activation, events that generate a feedback loop throughout the cell survival signalling pathway (Rowe and Chuang, 2004). Activation of BDNF promotes anti-apoptotic events crucial for cell maintenance and survival. Chronic social stress is one of many factors that decrease the expression of BDNF in the prefrontal cortex, decreasing the occurrence of anti-apoptotic cascades, and key regulatory enzymes and kinases, shifting the balance toward cell atrophy or even cell death (Sanacora, 2008). Lithium enhances the expression of BDNF, promoting cell survival. In addition, chronic lithium treatment changes CREB activity in exitotoxic situations by preventing loss of phosphorylated CREB and CRE-induced gene expression (Kopnisky et al., 2003). In addition, in vitro lithium treatment has given neuroprotection against a large variety of insults such as anticonvulsants, potassium deprivation, and β-bungarotoxin, among others (Rowe and Chuang, 2004). Lithium’s mood-stabilizing properties have been investigated for over 50 years, showing great efficacy in treating BD, in addition to reported evidence on its neuroprotective effects, yet the answers as to why lithium has mood-stabilizing properties remain largely unknown. If new discoveries on lithium’s mechanisms of action are made, potential  16  causes of the disease can be hypothesized and further investigated. If lithium is believed to be neuroprotective, it raises the possibility that neuroprotection is crucial for the treatment of bipolar disorder (Rowe and Chuang, 2004). In order to gain insights into the mechanism of action of mood stabilizers such as lithium, the inositol depletion hypothesis has been proposed and investigated for a number of decades. Studies over the past 30 years have shown that lithium inhibits some of the enzymes that are important for the generation of inositol inside the cell. Inositol mono-phosphate (IP) is the substrate of inositol-monophosphatase (IMPase), the enzyme involved in the synthesis of inositol from IP. Inositol bis-phosphate (IP2) is dephosphorylated by inositol-1,4 bisphosphate 1-phosphatase (IPP), in order to produce IP (Harwood 2005). Both IMPase and IPP have been reported to be inhibited by lithium (Allison et al., 1976; Inhorn and Majerus, 1987), leading many to believe that lithium plays a role in the reduction of intracellular inositol, and to suggest the inositoldepletion hypothesis as a possible mechanism of lithium action inside the cell. Inositol tris-phosphate’s (IP3) synthetic pathway begins with inositol. If there are depleting concentrations of inositol in the cell, calcium (Ca2+) intracellular release will be impaired due to decreased concentrations of IP3. Intracellular Ca2+ concentrations are crucial for modulating neuronal activity. Changes in the response and kinetics of Ca2+ release through changes in IP3 signalling would be expected to affect neuronal function (Harwood 2005). Conclusions regarding the clinical relevance of inositol depletion due to lithium treatment in BD are yet to be determined, as well as the mechanism of action of lithium and other mood stabilizers for the treatment of BD.  17  Over the years, multiple pharmacological treatments for bipolar disorder have shown tremendous efficacy in delaying the recurrence of mania or depression. Lithium is among those pharmachotherapies proven effective, and there is evidence that the response to lithium is familial and related to a family history of bipolar disorder (Perlis et al., 2009). The response to long-term lithium treatment clusters in families of responders to a significant extent, and findings report that the majority of relatives of lithium responders benefit from long-term administration of lithium as well (Grof et al., 2002). This suggests that there might be a genetic component or trait that can explain the efficacy of lithium as a mood stabilizer, yet genome wide association studies of lithium response in BD have not yielded single-nucleotide polymorphisms (SNPs) in the DNA that would meet the threshold for genomewide association (Perlis et al., 2009). Since lithium is the most common and readily available pharmacological treatment for BD, and some patients tend to have an excellent response to lithium treatment, based on the lithium response scale outlined in Turecki et al., 2001, the Consortium on Lithium Genetics (ConLiGen) was formed, as an international effort to elucidate the genetic factors of lithium response in bipolar disorder (Schulze et al., 2010). Findings from the literature emphasize the familiality in lithium treatment response, suggesting that genetic variation may contribute to differences among individuals in terms of treatment response. Identifying such differences, can aid in the development of novel treatments for BD specific to each individual (Schulze et al., 2010). The evidence thus far on lithium-induced alterations in gene expression is not sufficient to hypothesize the underlying mechanism of this mood stabilizer. Therefore, isolating genes that are  18  regulated by lithium is crucial for the understanding of this disorder, and more so if they play a role in the aetiology of BD (Sun et al., 2004). The characterization of lithium targets in peripheral cells can be studied using DNA array analysis in order to elucidate the effect of lithium on gene expression in lymphoblasts from patients with BD. Indeed, DNA array analysis in lymphoblasts identified several genes as potential targets for future studies in patients with BD, such as the alpha 1B-adrenoceptor (α1B-AR) gene, which activates a signalling pathway for the release of intracellular calcium, and the phosphodiesterase (PDE 4D) gene, important for cAMP signalling, both of which were found to have decreased expression due to lithium treatment in vitro (Sun et al., 2004). Most of the existing knowledge of the neurobiology of BD and other psychiatric illnesses comes from studies using post-mortem brain tissue, limiting the amount of variables that can be manipulated if live tissue was used instead. The use of cultured lymphoblasts from transformed lymphocytes has several advantages over the use of post-mortem brain tissue. Lymphocytes are easy to obtain from live patients’ blood, and their response to lithium treatment can be isolated. In addition, several studies on transformed lymphoblasts have reported similarities in biological responses, comparable to central nervous system cells (Sun et al., 2004). Many factors such as the dose-dependent response to lithium, and length of treatment among different samples can influence the outcome of the results gathered from lymphoblasts. However, reports on BDNF levels in lymphoblasts from lithium responsive patients with BD, found decreased BDNF protein levels, a similar biological response as that observed in brain tissue samples from patients with BD (Tseng et al., 2008). The fact that BDNF 19  expression can be measured in transformed lymphoblasts, and that the results could be attributed to pharmacological treatment suggests that they can be used as a diagnostic marker for the predisposition to develop BD, in lithium responsive patients with the disorder. Therefore, the search for peripheral biomarkers that can aid in the pharmacological treatment for this debilitating disorder continues. In addition, pharmacological treatment with mood-stabilizing drugs, such as lithium, may have antioxidant properties by preventing protein damage induced by overproduction of ROS and RNS (Frey et al., 2007). In addition to reported findings on mood stabilizers’ neuroprotective effects due to activation of cell survival signalling pathways (Rowe and Chuang, 2004), accumulated evidence points to the neuroprotective effects of moodstabilizing drugs against oxidative damage. Lithium, at clinically relevant concentrations, protects cells against oxidative cytotoxicity as a result of oxidative stress (Lai et al., 2006). Oxidative stress can be induced in neuronal cultured cells by mitochondrial complex I inhibition with rotenone or by direct application of hydrogen peroxide (H2O2) (Wang et al., 2007). Rotenone and H2O2 induce cytochrome c and caspase-3 activation, events critical for the activation of pro-apoptotic cascades. Cells under chronic lithium administration, and clinically relevant concentrations, are able to attenuate cell death induction by both rotenone and H2O2 (King and Jope, 2005; Lai et al., 2006). Findings such as these ones provide convincing evidence that lithium is able to prevent the oxidative damage caused by ETC complex I dysfunctionality found in BD. In the last few years the neuroprotective effects of mood stabilizers against oxidative damage are beginning to be recognized, despite the fact that there is not enough data directly linking those effects to mood-stabilizers’ treatment for BD. If the pharmacological treatment for 20  BD can be linked to the neuroprotective effects of drugs such as lithium, patients under long-term lithium treatment should manifest some of the antioxidant effects attributed to lithium, and be able to prevent the oxidative damage linked to BD. However, reports on complex I activity and levels of its principal subunit NDUFS7 in post mortem brain tissue samples showed a significant decrease in complex I activity and NDUFS7 expression in patients with BD compared to healthy controls and patients with major depressive disorder (MDD) and SCZ, suggesting this difference is specific to BD (Andreazza et al., 2010). Since complex I is one of the principal sites where electrons can be leaked to oxygen leading to the production of ROS and RNS, and in turn, to oxidative and nitration-induced damage to proteins, decreased complex I activity can lead to protein damage through excessive production of ROS and RNS. In support of this mechanism, increased protein carbonyl and 3-nytrotyrosine levels were reported in patients with BD, implying a correlation between decreased complex I activity and NDUFS7 expression with protein oxidation in the brain (Andreazza et al., 2010). In turn, mitochondrial dysfunction, at the level of complex I, and oxidative damage may play a role in the pathophysiology of BD.  1.5  Objectives and hypothesis  As the search for peripheral biomarkers for BD continues, in order to gain insight into the pathogenesis and genetics of BD, the objectives of this study are as follows: 1. Investigate the involvement of complex I functionality in lymphoblasts from lithium responsive patients with BD, their unaffected and affected relatives, in addition to 21  nonpsychiatric control subjects, by testing for complex I activity and the levels of its subunit, NDUFS7, under basal conditions. 2. Look at the relationship between complex I functionality and protein oxidative damage, by measuring carbonyl and 3-nitrotyrosine levels in lymphoblasts, across the groups in the study, under basal conditions. 3. Study the response to stress induced by low glucose treatment on complex I activity and NDUFS7 levels in lymphoblasts, across the groups in the study. 4. Study the response to stress induced by low glucose treatment on protein carbonyl and 3-nitrotyrosine levels in lymphoblasts, across the groups in the study. We hypothesize: 1. Decreased complex I activity and NDUFS7 levels in lymphoblasts from lithium responsive patients with BD, under basal conditions. In addition, we expect to find a difference in affected relatives when compared to unaffected relatives and nonpsychiatric controls, in both of these variables. 2. Increased protein carbonyl and 3-nitrotyrosine content in lymphoblasts from lithium responsive patients with BD, under basal conditions. In addition, we expect to find a difference in affected relatives when compared to unaffected relatives and nonpsychiatric controls, in both of these variables. 3. Lithium responsive patients with BD, in addition to affected relatives with BD, will not respond effectively to stress induced by low glucose treatment, determined by a further decrease in complex I activity and NDUFS7 levels, compared to basal conditions. However, unaffected relatives and control subjects will be able 22  to respond effectively to stress induced by low glucose treatment, by maintaining the complex I activity and NDUFS7 levels measured under basal conditions. 4. Lithium responsive patients with BD, in addition to affected relatives with BD, will not respond effectively to stress induced by low glucose treatment, determined by a further increase in protein carbonyl and 3-nitrotyrosine levels, compared to basal conditions. However, unaffected relatives and control subjects will be able to respond effectively to stress induced by low glucose treatment, by maintaining the protein carbonyl and 3-nitrotyrosine levels measured under basal conditions.  23  2 Methods 2.1  Lymphoblast samples  Lymphoblasts from patients (14 with BD, and 14 relatives of BD patients affected with a psychiatric illness) and from non-psychiatric comparison controls (N=15) as well as 16 relatives of BD patients unaffected by a psychiatric illness were made available for this project by Dr. Martin Alda M.D., F.R.C.P.C. and Dr. Trevor Young M.D., Ph.D., F.R.C.P.C. The research was approved by the Clinical Research Ethics Board (CREB) from the University of British Columbia, and all subjects gave informed consent. The subjects’ demographic and clinical data are given in Table 2 presented in the results section on page 41. All BD patients met Research Diagnostic Criteria (Spitzer et al., 1978) as well as the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM IV) criteria for BD, and they all responded unequivocally to long term lithium treatment. The samples of patients with BD were recruited from prospectively followed patients at specialized clinics at McMaster University, Hamilton, and at University of Ottawa. Patients with BD were followed prospectively, and fully stabilized on lithium monotherapy. Their response to lithium was documented by long-term stability on lithium monotherapy (8.4 +/- 5.5 years). The full criteria used to define the response to lithium have been described previously by Turecki et al., 2001. In order to diagnose patients with BD as excellent lithium responders, three different criteria need to be met. The first one is meeting the diagnosis of primary episodic BD based on the Schedule for Affective Disorders and Schizophrenia, Lifetime version (SADS-L) interview and  24  Research Diagnostic Criteria (RDC). The second is based on the patients’ high recurrence risk for episodes of mania and depression prior to treatment with lithium. Lastly, the patient has an unequivocal response to lithium where no other mood stabilizer is required to prevent the occurrence of an episode (Turecki et al., 2001). None of the patients with BD had any other Axis I or Axis II disorder. The patients and their relatives have been part of a genome-wide linkage study of BD patients responsive to lithium, representing a subset of a larger sample of 36 families comprising 275 interviewed and genotyped individuals (Alda et al., in press). The affected relatives met criteria for BDI and BDII, or recurrent unipolar depression; the unaffected relatives had no personal history of any psychiatric illness. Control subjects had no personal history of any psychiatric illness and had family history negative for bipolar disorder, major depression, and schizophrenia. After providing written informed consent, all subjects were interviewed using the SADS-L (Endicott and Spitzer, 1978) by pairs of clinicians (two psychiatrists or a psychiatrist and an experienced research nurse) blind with respect to their status. All interview information was reviewed by a panel of clinical researchers, blinded to the patients’ status as well. Family history information was determined using personal interviews with multiple family members and, in the case of those who could not be interviewed in person, the Family History Research Diagnostic Criteria (FH RDC) was used (Andreasen et al., 1977). All investigators were blinded to the group’s identity and demographic variables of the participants throughout all experiments and measures.  25  2.2  Cell culture  Blood collected from subjects was processed to generate lymphoblastoid cell lines by transformation with Epstein-Barr virus (EBV). In vitro transformation of human lymphocytes by EBV is the standard method used to generate stable human cell lines for research purposes. Based on the protocol outlined by Anderson and Gusella, 1984, upon blood collection in sodium herapin tubes, samples are centrifuged at 1500 rpm for 10 minutes at room temperature, to begin the process of virus transformation of lymphocytes to lymphoblastoid cell lines. Next, a Pasteur pipette is used to remove the layer of nucleated cells, which will be resuspended in RPMI 1640 medium. After centrifugation at 1800 rpm for 45 minutes, the discrete band of lymphoid cells produced is removed with a Pasteur pipette. Subsequently, the cells are resuspended in RPMI 1640 medium and washed twice by pelleting (1500 rpm, 10 min) and resuspension. The final cell pellet is resuspended in complete RPMI 1640 medium containing 1X penicillinstreptomycin, 15% fetal bovine serum, 4 mM glutamine, and 20 mM HEPES, with a pH of 7.2, followed by placement in tissue culture flasks with virus stock. B-lymphocyte cell line (B95-8) is used to generate the EBV. According to Alspaugh et al, B95-8 cells are cultured in RPMI 1640 medium containing 10% fetal calf serum, supplemented with glutamine and nonessential amino acids. After the B95-8 cells reached their peak growth of 2.5 x 106 cells/ml, they are spun down, resuspended at 2 x 107 per ml in RPMI 1640 and sonicated at 4°C. Centrifugation follows, and the supernatant is filtered through 0.45 µM Millipore filter (Millipore Corp., Bedford, Mass), giving the virus stock. Infected cultures are then incubated for 7 days at 37°C, and replenished with variable amounts of complete medium, depending on the apparent cell density determined by 26  the number and size of visible clumps of transformed cells. After reaching an ideal cell density of around 106 cells/ml, in a period of time ranging from 3 to 4 weeks, the cells are harvested by centrifugation (1500 rpm, 10 min) and resuspended in complete medium. Dimethyl sulfoxide is then added and the suspension is aliquoted into plastic freezing vials, placed overnight at -70°C overnight, before transferring to a liquid nitrogen freezer for long term storage (Anderson and Gusella, 1984). For our study on transformed lymphoblasts from lithium responsive patients with BD, the cell lines were incubated in RPMI-1640 medium, for two to three weeks depending on the sample, a period enough for the cells to reproduce to significant numbers required for experimental measures. After this incubation period, the cells were divided into two batches, and were cultured in either regular RPMI-1640 medium, to mimic normal glucose conditions, or low-glucose RPMI-1640 medium (25% normal glucose content) to induce cellular stress due to glucose deprivation, for a period of 5 days. As a result, two groups of cells were generated from each subject, normal and low glucose treated lymphoblasts’ samples. Therefore, each sample from the 4 groups in our study, was cultured in normal glucose medium and low glucose medium.  2.3  Mitochondrial and cell pellet extraction  Mitochondrial extraction from lymphoblasts began with the centrifugation of 50ml of cells from culture (approximately 5 x 107 cells) at 600g for 10 minutes at a temperature of 4°C. The supernatant was discarded and the remaining cell pellet was washed twice with cold PBS and centrifuged at 600g again for 10 minutes at 4°C. Cells were  27  subsequently resuspended in 3ml of ice-cold IBc (10ml of 0.1M Tris-MOPS and 1ml of 0.1M EGTA/Tris and 20ml of 1M sucrose, bringing the final volume up to 100ml with distilled water, and adjusting the pH to 7.4) and stroked with a Dounce grinder on ice for 30 seconds, for tissue homogenization. 0.1M Tris-MOPS was prepared by dissolving 12.1 g of Tris in 500ml of distilled water; pH was adjusted to 7.4 using MOPS powder; the solution’s final volume was 1L and was stored at 4°C. 0.1M EGTA/Tris was prepared by dissolving 38.1g of EGTA in 500ml of distilled water; pH was adjusted to 7.4 using Tris powder; the solution’s final volume was 1L and was stored at 4°C. 1M sucrose was prepared by dissolving 342.3g of sucrose in 1L of distilled water, and aliquots of 20ml were stored at -20°C. After tissue homogenization, samples were centrifuged, and the collected supernatant was centrifuged at 7000g for 10 minutes at 4°C. The resulting supernatant was discarded and the cell pellet was washed with 200µl of ice-cold IBc. Following another round of 7000g centrifugation, the resulting supernatant was discarded once again. The mitochondria-enriched pellets were dissolved in 200µl of storage buffer (0.21M mannitol, 0.07M sucrose, 10mM Tris-HCl, pH 7.4), and stored at -80°C in aliquots of 150µl, before determining protein concentration and complex I activity measurements. Mitochondrial protein concentrations were determined using the Bradford assay which is a colorimetric protein assay based on an absorbance shift of the dye Coomassie Brilliant Blue G-250, a dye that under acid conditions, changes color from red/brown to blue upon binding to proteins from the sample. This reaction is based on the formation of a complex when the Coomassie dye first donates its free electron to the ionisable groups on the protein, causing a disruption of the proteins’ native state, which exposes its hydrophilic pockets.  28  These readily available pockets can in turn bind non-covalently through van der Waals interactions, positioning the positively charged amino acid residues close to the negatively charged molecules of the dye. This interaction between the dye and the protein stabilizes the blue form of the Coomassie dye, where the amount of dye-protein complex present in the solution is a measure of the protein concentration of the sample, which can be calculated by measuring the absorbance spectrum of the complex at a wavelength of 595nm with a spectrophotometer plate reader. Samples with a protein concentration of less that 1µg/µl were recollected prior to testing for complex I activity. After complex I activity measurements were gathered, we began the extraction of the cell pellet from the cultured lymphoblasts in order to test for NDUFS7, protein carbonyl, and 3-nitrotyrosine levels. This extraction was carried out by the centrifugation of 50ml of cells from culture (approximately 5 x 107 cells) at 600g for 10 minutes at a temperature of 4°C. The supernatant was discarded and the remaining cell pellet was washed twice with cold PBS, before storing at -80°C. Protein concentration measurements before testing for NDUFS7 and protein carbonylation were obtained using the Dc protein assay, which is a colorimetric assay to determine the protein concentration of a sample upon detergent solubilisation. This assay is based on the reaction of protein with an alkaline copper tartrate solution and Folin reagent. Color development is due to the reaction between protein and copper in the alkaline medium and by the subsequent reduction of Folin reagent by the copper-treated protein. This causes Folin to lose up to 3 oxygen atoms, producing several reduced species with a characteristic blue color with a maximum absorbance of 750nm. Absorbance in turn, was read with a spectrophotometer plate reader at a wavelength of 750nm. The 29  Bradford assay method described previously was used in order to calculate the protein content of the cell pellet samples destined for 3-nitrotyrosine testing. All samples were tested in triplicates for all measures obtained (complex I activity, NDUFS7 levels, protein carbonyl levels and 3-nitrotyrosine content).  2.4  Complex I activity  Complex I activity was measured using a spectrophotometric technique adapted from Estronell et al., 1993, and Andreazza et al., 2010. First, mitochondria-enriched pellets extracted from lymphoblasts underwent three cycles of freeze and thaw in order to expose the inner mitochondrial membrane for the determination of complex I activity kinetics. The mitochondria-enriched pellets were diluted to 20µg/µl in the assay buffer (50mM potassium chloride, 10mM Tris hydrochloride, 1mM EDTA, and 2mM potassium cyanide; pH 7.4), as this was the concentration required for optimal results during preliminary testing. 500µM NADH was prepared by dissolving a 500mM stock solution of NADH with distilled water, and stored in aliquots of 800µl at -80°C. CoQ1 was dissolved in ethanol to achieve a concentration of 500µM, and stored in aliquots of 200µl at -20°C. Rotenone at a concentration of 100µM was prepared fresh on the day when complex I measurements will take place, by dissolving in ethanol, and protecting the inhibiting agent with aluminum foil, as rotenone is known to be extremely sensitive to light. NADH, CoQ1 and rotenone were placed in a water bath at a temperature of 30°C for 5 to 10 minutes, until the reagents felt warm upon touching against the skin. After adding the corresponding amounts of assay buffer and sample to the wells of a 96 well plate, the plate was placed in the water bath set at 30°C, and the amounts of 30  reagents outlined in Table 1 were added to the plate. The oxidation of NADH by the enzymes in complex I, which depends on the presence of the coenzyme Q1, is known to work in an optimal temperature and pH of 30°C and 7.4 respectively.  Table 1. Volume of reagents added for complex I activity measurements Assay Buffer  MTC Sample  NADH  Coenzyme Q1  Rotenone  Final Volume  Blank  20µl  65µl  15µl  -  -  100 µl  Overall Rate  10µl  65µl  15µl  10µl  -  100 µl  Inhibited Rate  -  65µl  15µl  10µl  10µl  100 µl  Read for 5 minutes at 30°C  The reaction procedure is summarized in Figure 2 on the following page. The oxidation of NADH started after the addition of 50µM coenzyme Q1, the last reagent to be added to the reaction mixtures for each sample in the plate, except for those corresponding to the blank. Immediately after the addition of CoQ1, the plate was read at 340nm, which is the absorbance wavelength for NADH, in 20 second intervals for a period of 5 minutes. Complex I is inhibited by agents such as rotenone. Complex I inhibition by rotenone is an effective control as it rules out other enzymes and agents present in the mitochondrial enriched pellets that are capable of oxidizing NADH in the presence of CoQ1, so the rate of disappearance of NADH obtained when reading the absorbance of NADH at 340nm, is due to complex I activity converting NADH into NAD+. The end result, is a plot of absorbance over time, with a linear equation in the form Y = mX + b,  31  where m corresponds to the slope which in this case, it is the measurable rate needed to calculate complex I activity by subtracting the rate of inhibition from the overall rate mentioned in Figure 2. As the rate of disappearance of NADH is negative, the absolute value of the slope was used for all calculations.  Figure 2. Kinetic activity measured in order to determine complex I functionality Blank  Overall Rate  Inhibited Rate  SAMPLE  SAMPLE  SAMPLE  +  +  +  NADH → NAD+  NADH →

NAD+  NADH →

NAD+  340nm  340nm  340nm  ↑  ↑  +  +  CoQ1  Rotenone  5 minutes  5 minutes  5 minutes  + CoQ1 No complex I activity No rate [ ] NADH = 75 µM  Complex I activity ↑ Rate [ ] CoQ1 = 50 µM  Complex I activity ↓ Rate [ ] Rotenone = 10 µM  Complex I activity (µmol NAD+/min/mg of protein) = [(Overall Rate – Blank) – Inhibited Rate] * 1000  32  2.5  NDUFS7 levels  Immunoblotting was used in order to measure NDUFS7 protein levels. Cell pellet fractions were used instead of mitochondrial fractions. Before use, the cell pellets underwent cell lysis using Nestler’s buffer (distilled water, 100mM hepes, 1M NaCl, 100% glycerol, 1M MgCl, 0.5M EDTA, 0.1M EGTA, nonidet P40, and a protease inhibitor cocktail). The lysate was centrifuged at 10000g for 20 minutes at 4°C. Following centrifugation, the supernatant was collected in aliquots of 100µl and stored at -80°C, until samples were needed for testing. After determining the samples’ protein concentrations with the Dc protein assay, a western blot test curve was performed with varying concentrations ranging from 10µg to 30µg, in order to see how much protein is required to achieve optimal band thickness. It was determined that 15µg of lysed lymphoblasts’ cell pellet extracts would give protein bands in the 31kDa position according to the loading marker, with the desired thickness and size. Complex I subunit NDUFS7 has an approximate size of 31kDa. In turn, 20µl of sample with the desired concentration were loaded on each of the wells of 12% acrylamide sodium dodecyl sulphate-polyacrylamide gels followed by electrophoresis, applying a voltage of 110V until the protein samples migrated to the bottom of the gel. This process took from 90 to 100 minutes. Immediately after, the gels were transferred to polyvinylidene difluoride membranes for no more than 70 minutes. The membranes were left to dry overnight before soaking in 100% methanol and subsequently blocking with 0.05% polysorbate 20 in 0.01M phosphate-buffered solution containing 5% nonfat milk (PBS-T) at room temperature with gentle shaking for 1 hour. The blots were independently probed using either rabbit anti-human NDUFS7 antibody (Santa Cruz 33  Biotechnology), 1:500, or mouse β-Actin antibody (Santa Cruz Biotechnology), 1:2000, as loading control in PBS-T at room temperature with gentle shaking for 2 hours. The membranes were washed with PBS-T 3 times, each wash lasting 5 minutes, in order to incubate with goat anti-rabbit horseradish peroxidise conjunct secondary antibody (Ab Cam), 1:1000, and goat anti-mouse horseradish peroxidase conjunct secondary antibody (AbCam), 1:3000, for loading control, in PBS-T for 1 hour at room temperature. Before applying electrochemiluminescence reagents (Western lightning Plus-ECL, PerkinElmer Inc.), the membranes were washed with PBS-T 3 times, each wash lasting 5 minutes, and one last wash with PBS for 5 minutes. The luminol in the chemiluminescent solution is converted to a light-emitting form at a wavelength of 428 nm by the antigen/primary antibody/secondary antibody/peroxidase complex, which is bound to the membrane, in an H2O2 catalized oxidation reaction. The light is detected by short exposure to blue-light sensitive films, such as the one in the molecular imaging system used (LAS-3000). The resulting protein bands showing at 31kDa were quantified densitometrically and normalized to the β-actin signal using a molecular imaging system.  2.6  Protein oxidation  The levels of protein carbonyl groups in protein residues reflect the extent of protein oxidation of the samples, and were measured using the OxyBlot™ Protein Oxidation Detection Kit (catalog No. S7150; Chemicon, Kankakee, Illinois). As proteins are one of the major targets of oxygen free radicals, metal catalyzed oxidation of proteins  34  introduces carbonyl groups (aldehydes and ketones) to the protein side chains of lysine, proline, arginine and threonine. In the presence of Fe3+ or Cu2+, O2, and an electron donor, several metal-catalyzed oxidation systems such as enzymatic systems including NADH dehydrogenase and other flavoproteins, can oxidize amino acid residues of proteins. Flavoproteins serve as electron carriers in cellular metabolism, but when their normal electron acceptors are present at low concentrations, the reduced forms of these flavoproteins can undergo auto oxidation, forming H2O2, or can transfer electrons to Fe3+, by either superoxide-dependent or independent pathways (Stadtman, 1993). Therefore, the ability of these metal-catalized oxidation systems to catalize the oxidative modification of proteins and other macromolecules, can be due to interactions with H2O2 and Fe2+, what is commonly known as the Fenton reaction. The most common sites of oxidation by metal-catalized oxidation systems are the amino acid residues mentioned above. This metal ion-catalyzed oxidation of proteins converts lysine into 2-amino adipicsemialdehyde residues; proline residues are converted to either glutamate, pyroglutamate, cis/trans-4-hydroxyproline, 2-pyrrolidone, glutamic semialdehyde, or gamma-aminobutyric acid residues; arginine residues are converted to glutamic semialdehyde residues; and threonine residues are converted to 2-amino-3-ketobutyric acid residues (Stadtman, 1993). The OxyBlot™ Protein Oxidation Detection Kit gives a sensitive and straight forward methodology for detection and quantification of proteins modified by oxygen free radicals and other reactive species. Protein oxidation measurements are accomplished by monitoring the reaction between carbonyl groups with 2,4-dinitrophenylhydrazine (DNPH), which yields 2,4-dinitrophenylhydrazone (DNPhydrazone) - derivatized protein samples.  35  After determining the samples’ protein concentrations with the Dc protein assay, a western blot test curve was performed following the instructions given in the kit for sample preparation, with varying lymphoblasts’ concentrations ranging from 10µg to 30µg, in order to know the amount of protein required to achieve optimal results. It was determined that 15µg of lysed lymphoblasts’ cell pellet extracts would be used to test for protein oxidation by allowing the reaction between carbonyl groups in the samples and DNPH to happen, yielding derivatized samples of DNP-hydrazone. The DNPH solution was provided by the kit (No. 90448). The DNP-derivatized protein samples were loaded on each of the wells of 12% acrylamide sodium dodecyl sulphate-polyacrylamide gels followed by electrophoresis, applying a voltage of 110V until the protein samples migrated to the bottom of the gel. This process took from 90 to 100 minutes. Immediately after, the gels were transferred to polyvinylidene difluoride membranes for no more than 70 minutes. The membranes were left to dry overnight before soaking in 100% methanol and subsequently blocking with 0.05% polysorbate 20 in 0.01M phosphate-buffered solution containing 5% nonfat milk (PBS-T) at room temperature with gentle shaking for 1 hour. The blots were independently probed with the primary antibody, rabbit anti-DNP antibody (No. 90451) specific to the DNP moiety of the proteins, in a 1:150 dilution in PBS-T, at room temperature with gentle shaking for 2 hours. The membranes were washed with PBS-T 3 times, each wash lasting 5 minutes, in order to incubate with a goat anti-rabbit horseradish peroxidase-antibody conjugate, goat anti-rabbit IgG (No. 90452), directed against the primary antibody, in a 1:300 dilution in PBS-T, for 1 hour at room temperature. The membranes were then treated with electrochemiluminescent reagents (Western lightning Plus-ECL, PerkinElmer Inc.)  36  after washing with PBS-T for 3 times, for 5 minutes each, and one last wash with PBS for another 5 minutes. The luminol in the chemiluminescent solution is converted to a light-emitting form at a wavelength of 428 nm by the antigen/primary antibody/secondary antibody/peroxidase complex, which is bound to the membrane, in an H2O2 catalized oxidation reaction. The light is detected by short exposure to bluelight sensitive films, such as the one in the molecular imaging system used (LAS-3000). The protein bands were analyzed densitometrically and normalized against a control sample without 2,4-dinitrophenylhydrazine, in order to measure the extent of protein oxidation in the samples due to the presence of carbonyl groups. The kit is able to detect as little as 5 fentomoles of carbonyl residue present in the samples.  2.7  Tyrosine nitration-induced damage  The levels of 3-nitrotyrosine in amino acid protein residues reflect the extent of tyrosine nitration-induced damage, and were measured using the OxiSelect Nitrotyrosine ELISA [enzyme-linked immunoabsorbent assay] Kit (catalog No. STA-305; Cell Biolabs Inc, San Diego, California). The nitrotyrosine quantitation kit is a competitive ELISA, a method useful for quantifying a specific antigen present in a protein sample. In this case, the antigen is 3-nitrotyrosine or nitrated bovine serum albumin (BSA). A competitive ELISA is based on the principle of incubating free antigen and antibody in order to form an antigen-antibody complex, and any free antibody left will bind to an antigen-coated surface of the assay plate. In the nitrotyrosine ELISA kit, the enzyme immunoassay (EIA) plate has already been coated with the antigen, BSA. To the  37  preabsorbed EIA plate, we proceeded to add 50µl of diluted protein sample from lymphoblasts, in a 1:100 dilution factor (previously determined by a dilution test curve to be the ideal combination of sample and diluent needed to achieve optimal results) in the Assay Diluent (No. 310804) provided by the kit, into each well. In addition to the lymphoblast samples, the EIA plate should have 50µl of the nitrated BSA standards added to the EIA plate wells. The nitrated BSA standards, obtained by serial dilution of 60µl of BSA standard in 240µl of Assay Diluent, will provide not only a standard curve of the absorbance value obtained at the end of the procedure (OD 450nm), but the nitrotyrosine concentration (nM) determined by the serial dilution, in order to obtain a linear equation that will help us determine the amount of 3-nitrotyrosine in our samples. After addition of the samples and nitrated BSA standards to the EIA plate, a 10 minute incubation period at room temperature on an orbital shaker followed. We then proceeded to add 50µl of anti-nitrotyrosine antibody (No. 230502), in a 1:1000 dilution in Assay Diluent, into each well, followed by 1 hour incubation at room temperature on an orbital shaker. At this point, there were varying amounts of antigen-antibody complex in each sample, depending on the amount of antigen (3-nitrotyrosine) present in the samples. The antigen-antibody complexes forming at the surface of the EIA plate is dependent on the amount of 3-nitrotyrosine is present in the samples, that will allow for the binding of free antibody to the nitrated BSA on the surface of the plate. After the incubation time, the plates were washed 3 times with 250 µl of 10 x Wash Buffer (No. 310806). During this process, unbound antibody was removed. The more antigen in the samples, the less antibody was available to bind to the antigen in the plate’s surface, hence the term “competitive” ELISA. Then, 100 µl of secondary antibody-enzyme  38  conjugate (No. 231003) against the primary antibody, in a 1:1000 dilution in Assay Diluent, was added to each well, incubating at room temperature for 1 hour on an orbital shaker. After the incubation time, the plate was washed 3 times with 250 µl of 10 x Wash Buffer per well. Afterwards, 100 µl of Substrate Solution (No. 310807) was added to each well, incubating at room temperature on an orbital shaker for at least two minutes. During this time, the reaction between the enzyme added in the secondary antibody and the substrate tetramethylbenzidine (TMB) can be monitored by the change in color, from blue to yellow, in the wells of the plate. When this occurs, the enzymatic reaction has happened and it needs to be stopped to prevent saturation, by adding 100 µl of Stop Solution (No. 310808) into each well. Immediately after, the plate was read using a spectrophotometer at a wavelength of 450 nm. The antigen concentration was determined by the signal strength elicited by the enzymatic-substrate reaction. The higher the original content of 3-nitrotyrosine present in the samples, the weaker the signal read by the spectrophotometer. The protein content of nitrotyrosine in the unknown samples was determined by comparing absorbance values of the samples with the standard curve prepared with nitrated BSA standards with known nitrotyrosine concentrations, as previously mentioned. The resulting concentration of 3-nitrotyrosine obtained (nM), was subsequently divided by the samples’ protein concentration (mg/ml), previously determined using the Dc protein assay, and the results of 3-nitrotyrosine content in the samples were expressed in nmol/mg protein. The kit has a nitrotyrosine detection sensitivity range of 20 nM to 8.0 µM.  39  2.8  Statistical methods  Statistical analysis of the data was performed using a computer software program (BMDP for Windows version 8.1; BMDP Statistical Software Inc, Saugus, Massachusetts). Normal distribution of data was determined using the KolmogorovSmirnov test. Data was subjected to analysis of variance and covariance (ANCOVA) with repeated measures, followed by the least significant difference post hoc test, in order to determine significant differences on the measured variables (Complex I activity, NDUFS7, protein carbonyl, or 3-nitrotyrosine levels) across the four groups in our study. Age and gender were added as covariates, and persistence of the significant difference in main effect between groups was reported. Correlations were analyzed using the Pearson product moment correlation test. Data are given as mean and standard deviation.  40  3 Results  3.1  Demographic variables  Demographic and clinical information of patients and controls are given in Table 2. Controls (n=15) and unaffected relative (n=16) were matched with patients with BD (n=14) and those belonging to the group of affected relatives (n=14) for age and gender. As a result, no significant differences were noted among the groups on these measures. All lithium responsive patients with BD were euthymic, under lithium treatment at the time of sampling, and had no history of drugs or alcohol abuse.  Table 2. Demographic and clinical data for lymphoblasts in patients and controls Variable Control  Lithium responsive BD  Unaffected Relative  Affected Relative  N 15  14  16  14  9:6  9:5  6 : 10  3 : 11  46.1 +/- 14.2  43.5 +/- 11.9  39.0 +/- 18.1  47.1 +/- 11.1  N/A  8:6:0  N/A  3:4:7  N/A  25.6 +/- 8.5  N/A  29.3 +/- 9.2  men : women Age (mean +/- SD)  Years Diagnosis (BD I : BD II : Unipolar depression) Age at onset (mean +/- SD)  Years  41  3.2  Mitochondrial complex I functionality and protein damage  We found that complex I activity under normal (basal) glucose conditions was significantly different between the groups in our study (F3,55=4.91; P=0.0043). Complex I activity increased (23.30%) in patients with BD (P=0.0975), and to a significant extent in unaffected (55.6%; P=0.005) and affected relatives (48.54%; P=0.0021), in relation to controls (Figure 3A). We did not find significant differences between the groups under low glucose treatment (F3,55=0.14; P=0.9339) (Figure 3B),  Figure 3. Complex I activity  Complex I activity for normal (A) and low (B) glucose treatment in lymphoblasts of control (CTL), bipolar disorder (BD), unaffected relatives (UR), and affected relatives (AR) subjects. Complex I, or nicotinamide adenine dinucleotide [NADH]-ubiquinone oxidoreductase, is a multisubunit integral membrane protein complex, part of the mitochondrial electron transport chain catalyzing the transfer of electrons from NADH + to ubiquinone. The activity of complex I was determined by monitoring the oxidation of NADH to NAD at 340nm. Differences among the groups were determined using 1-way analysis of variance, followed by the least significant difference test comparing the patient groups and unaffected relatives group with the control group. Horizontal lines indicate the group mean. (A) *P=0.0975; †P=0.005; ‡P=0.0021  42  nor did we find significant differences in complex I activity under low glucose stress conditions, compared to those reported under basal conditions, in each of the 4 groups in the study (F3,55=0.10; P=0.7566). We did not find that NDUFS7 levels were significantly different when comparing the 4 groups, under normal (F3,55=0.03; P=0.9940) (Figure 4A) and low (F3,55=1.80; P=0.1572) (Figure 4B) glucose treatment. The low glucose treatment had no significant effects on NDUFS7 levels across all groups, compared to the value obtained under normal glucose treatment (F3,55=0.00; P=0.9675).  Figure 4. NDUFS7 levels  NDUFS7 levels for normal (A) and low (B) glucose treatment in lymphoblasts of control (CTL), bipolar disorder (BD), unaffected relatives (UR), and affected relatives (AR) subjects. The immunocontent of NDUFS7, one of the principal subunits of complex I, was analyzed by means of Western blotting. Differences among the groups were determined using 1-way analysis of variance, followed by the least significant difference test comparing the patient groups and unaffected relatives group with the control group. Horizontal lines indicate the group mean. AU indicates arbitrary units.  43  Next, we analyzed protein carbonyl and 3-nitrotyrosine levels in lymphoblast samples in order to assess oxidative and nitrosative damage to mitochondrial proteins, respectively. We did not find that protein carbonyl levels were significantly different when comparing the 4 groups, under normal (F3,55=0.24; P=0.8656) (Figure 5A) and low (F3,55=0.71; P=0.5504) (Figure 5B) glucose treatment. The low glucose treatment had no significant effects on protein carbonyl levels across all groups, compared to the value obtained under normal glucose treatment (F3,55=2.10; P=0.1532).  Figure 5. Protein carbonyl levels  Cellular protein carbonyl immunocontent for normal (A) and low (B) glucose treatment in lymphoblasts of control (CTL), bipolar disorder (BD), unaffected relatives (UR), and affected relatives (AR) subjects. Western blotting was used to analyze cellular protein carbonyl levels. Horizontal lines indicate the group mean. AU indicates arbitrary units.  Furthermore, there were no significant differences in 3-nitrotyrosine levels between the groups under both normal (F3,55=0.29; P=0.8341) (Figure 6A) and low (F3,55=1.57;  44  P=0.2084) (Figure 6B) glucose treatment. The low glucose treatment had no significant effects on 3-nitrotyrosine levels across all groups, compared to the value obtained under normal glucose treatment (F3,55=0.03; P=0.8536).  Figure 6. 3-Nitrotyrosine content  Cellular 3-nitrotyrosine levels for normal (A) and low (B) glucose treatment in lymphoblasts of control (CTL), bipolar disorder (BD), unaffected relatives (UR), and affected relatives (AR) subjects. 3Nitrotyrosine levels were determined by means of competitive enzyme-linked immunoabsorbent assay. Horizontal lines indicate the group mean.  Next, we wanted to see how complex alterations reflected on the levels of NDUFS7, protein carbonyl and 3-nitrotyrosine, under both normal and low glucose treatment conditions. As the only significant difference between the groups reported in our study was on measurements of complex I activity, and not followed by significant differences in NDUFS7, protein carbonyl, and 3-nitrotyrosine levels, we did not expect to find any correlation between complex I activity and NDUFS7, protein carbonyl, and 3-  45  nitrotyrosine levels. As expected, we did not find any correlation between complex I activity and NDUFS7 (n=58; r=-0.164; P=0.218) (Figure 7A), protein carbonyl (n=58; r=0.027; P=0.84) (figure 7B), and 3-nitrotyrosine (n=57; r=-0.163; P=0.226) (Figure 7C) levels under normal glucose conditions.  Figure 7. Correlations for normal glucose treatment  Correlations among complex I activity and NDUFS7 levels (A), protein oxidation (carbonyl levels) (B), and tyrosine nitration-induced damage (3-nitrotyrosine) (C), for normal glucose treatment. Results were assessed using the Pearson product moment correlation test. AU indicates arbitrary units.  In addition, we did not find any significant correlation between complex I activity and NDUFS7 (n=58; r=-0.237; P=0.073) (Figure 8A), protein carbonyl (n=58; r=-0.061; P=0.652) (figure 8B), and 3-nitrotyrosine (n=57; r=-0.052; P=0.701) (Figure 8C) levels under low glucose conditions. The lack of correlation across these measures was expected, since we did not find a significant difference in complex I activity, and levels  46  of NDUFS7, protein carbonyl, and 3-nitrotyrosine across the groups, that would be indicative of a correlation across these measures under these conditions.  Figure 8. Correlations for low glucose treatment  Correlations among complex I activity and NDUFS7 levels (A), protein oxidation (carbonyl levels) (B), and tyrosine nitration-induced damage (3-nitrotyrosine) (C), for low glucose treatment. Results were assessed using the Pearson product moment correlation test. AU indicates arbitrary units.  47  4 Discussion The objective of this study was to investigate the involvement of complex I functionality in transformed lymphoblasts from lithium responsive patients with BD, their unaffected and affected relatives, and its relation to protein oxidative damage, in addition to the response of mitochondrial complex I to stress induced by glucose deprivation. Complex I functionality was determined by measuring its enzymatic activity and the levels of its subunit, NDUFS7. Oxidative damage was measured by the levels of protein carbonyl and 3-nitrotyrosine content present in our samples. In the present study, we found significant differences in complex I activity between groups, with the highest values for unaffected and affected relatives of lithium responsive patients with BD, and lowest for control subjects, under normal glucose conditions. This up-regulation in complex I activity was not followed by alterations in NDUFS7 levels, as we did not find significant differences between the groups on this measure. Furthermore, there were no significant differences in protein carbonyl and 3nitrotyrosine levels across the groups in our study. We measured the response of lymphoblasts from lithium responsive patients with BD, their unaffected, and affected relatives, to stress induced by glucose deprivation. Our results showed that the levels of complex I activity, NDUFS7, protein carbonyl, and 3-nitrotyrosine were not affected by stress induced by glucose deprivation, across the groups in our study, as we did not find significant differences between the groups on these measures. Lastly, no correlation between complex I activity and NDUFS7, protein carbonyl, or 3-nitrotyrosine levels were found, under both normal and low glucose treatment. Together our results suggest an up-regulation of peripheral complex I activity in relatives of patients with BD, and to a 48  lesser extent in patients with BD. This may be indicative of a compensatory mechanism present in lithium responsive patients with BD and their relatives, at the level of complex I, that may be able to prevent protein oxidative damage. There is growing evidence about mitochondrial impairment and its relationship with BD, placing complex I dysfunctionality as an important contributor to the pathogenesis of BD (Iwamoto et al., 2005; Konradi et al., 2004; Sun et al., 2006; Wang, 2007). However, the trend towards increased complex I activity measured in our samples of lithium responders, unaffected and affected relatives, in relation to nonpsychiatric control subjects, suggests that this complex I alteration is indicative of a mitochondrial compensatory mechanism, able to prevent the complex I dysfunctionality characteristic of the disorder. The lack of significant differences between the groups with respect to NDUFS7 levels, further support the absence of mitochondrial impairment, at the level of complex I, in our group of lithium responders, their unaffected and affected relatives. Real-time quantitative polymerase chain reaction data from lymphoblastoid cell lines derived from patients with BD show a significant down-regulation in the expression of complex I subunit NDUFV2 in Japanese patients with BD, compared with control subjects (Washizuka et al., 2009). Similarly, microarray analysis measuring gene expression levels in lymphocytes from patients with BD, reported differences in complex I subunits from those of nonpsychiatric controls, such as down-regulation of NDUFS2 and up-regulation of NDUFA5 (Naydenov et al., 2007). The up-regulation in gene expression of complex I subunit NDUFA5 observed in lymphocytes from patients with BD, in addition to other unknown alteration in complex I subunits’ expression, could explain the trend towards increase in complex I activity in lithium responsive BD 49  patients, reported in our study, to some extent. In addition to testing for complex I functionality in our samples under basal conditions, we wanted to test complex I activity in lymphoblasts under low glucose treatment, as Naydenov et al reported a downregulation of complex I subunits NDUFA5, NDUFA6, and NDUFB1 in lymphocytes from patients with BD treated with low glucose, compared to normal glucose controls, suggesting an impairment in complex I activity in BD under those conditions. Our results under normal glucose treatment did not show impairment of complex I activity in patients with BD, and our measurements on NDUFS7 levels failed to show any significant alterations when comparing the 4 groups in our study. Interestingly, when our samples were subjected to glucose deprivation stress, the absence of complex I impairment and NDUFS7 alterations remained, across all the groups in our study. The reports (Naydenov et al., 2007; Washizuka et al., 2009) on complex I alterations in patients with BD highlight the differences in gene expression levels of additional complex I subunits in peripheral cells, that could play a crucial role in explaining not only the observed increase in complex I activity across our lymphoblast samples of lithium responders and their relatives, but the lack of dysfunctionality of this important enzymatic complex. Several lines of evidence gathered from microarray and real-time quantitative polymerase chain reaction data from post mortem brain tissue, have suggested an association of BD with decreased expression of multiple messenger ribonucleic acids (mRNAs) coding form complexes I through V subunits (Iwamoto et al., 2005; Konradi et al., 2004; Sun et al., 2006). Decreased mRNA levels of subunits NDUFS1, UQCRC2, and COX15, from complexes I, III, and V respectively, in the prefrontal cortex of patients 50  with BD has been reported (Iwamoto et al., 2004). In addition, down-regulation of 8 genes coding for subunits of ETC complexes I (NDUFS7 and NDUFS8), III (UQCRC2), IV (COX5A and COX6C), and V (ATP5C1, ATP5J, and ATP5G3) was observed in the prefrontal cortex of patients with BD, as well as decreased mRNA levels of subunits NDUFS7 and COX6C (Sun et al., 2006). Furthermore, evidence from post-mortem brain tissue studies measuring complex I activity in patients with BD (Andreazza et al., 2010) reported a significant decrease in activity of this large ETC multi subunit complex, as well as the levels of its subunit, NDUFS7. These complex I alterations were found to be important contributors to the characterization of BD. In addition, Andreazza et al reported a positive correlation between complex I activity and NDUFS7 levels in prefrontal cortex from patients with BD (Andreazza et al., 2010). Our results did not show complex I impairment in lithium responders, corroborated by the absence of alteration in NDUFS7 levels, in addition to a lack of correlation between complex I activity and levels of NDUFS7 in our samples. Together, our findings on complex I functionality could be an important contributor towards our understanding of the mechanisms in which peripheral cellular metabolism differs from brain metabolism. Future studies are necessary in order to continue our search for peripheral biomarkers that can give further insight into the pathogenesis of BD, outside of the central nervous system. Complex I, consisting of up to 45 different subunits, is divided into 3 modules according to function: the dehydrogenase, hydrogenase, and transporter modules. The dehydrogenase module is the one responsible for the oxidation of NADH via FMN onto electron transferring iron-sulfur clusters. The hydrogenase module directs the electrons 51  lost by NAHD to electron acceptors in mitochondria. Lastly, the transporter module makes the translocation of protons across the mitochondrial membrane possible (Brandt et al., 2003). In BD, abnormalities in complex I functional modules have been reported, such as decreased expression of the subunit NDUFS1 (Iwamoto et al., 2004) in the dehydrogenase module, as well as decreased expression of subunits NDUFS7 and NDUFS8 in the hydrogenase module (Sun et al., 2006), suggesting that patients with BD may have an impaired ability to oxidize NADH, and in turn, to transfer electrons lost by NADH onto ubiquinione. In fact, complex I impairments reflected as a decrease in the oxidation rate of NADH, and a decrease in levels of its subunit NDUFS7, are observed in BD (Andreazza et al., 2010). As a result, the free electrons leaked from complex I persist long enough in mitochondria that they can react with the readily available molecular oxygen, producing superoxide (O2•–), contributing to the susceptibility of mitochondrial proteins to oxidative (Beal, 2002) and nitrosative damage (Murray et al., 2003). The findings highlighted herein provide strong evidence to show down-regulation of various complex I subunits in patients with BD. These abnormalities may be associated with the susceptibility to cellular damage observed in BD, through oxidative stress as a consequence of mitochondrial dysfunction. The generation of O2•– is a direct consequence of electrons leaking from ETC sites in mitochondria such as complex I, as well as from complex III (Figure 1). As our study, along with the reported findings from the literature, focuses on complex I functionality, we will discuss the accumulation of O2•– in relation to complex I activity, and leave complex III details for further discussion based on future studies. The generation of O2•–, and its subsequent reaction with superoxide dismutase (SOD), forming hydrogen 52  peroxide (H2O2), is a physiological process and part of normal cellular metabolism. SOD’s cellular role is to clear ROS such as O2•–, the highly reactive radical anion, converting it into H2O2, in order to prevent cellular damage due to free radicals. The H2O2 generated, is further transformed into 2 molecules of water by enzymes such as catalase (CAT) and glutathione peroxidase (GSH-Px) (Kuloglu et al., 2002). Cellular metabolic and catalytic processes can be altered by abnormal molecular turnover that can overwhelm antioxidant systems, such as those utilizing CAT and GSH-Px. In pathological situations, such as excess production of O2•– by dysfunctionality in complex I, or not enough antioxidant enzymes present to clear the excess O2•– and H2O2 generated, can lead to the production of hydroxyl radicals (OH•), and in turn, to protein oxidative damage. One of the mechanisms by which impaired complex I activity could be associated with increased protein carbonylation, is the overproduction of OH•. As described in Figure 1, this highly reactive radical is produced by the reaction between excess O2•–, and the enzyme SOD, through a H2O2 intermediate (Beal, 2002). In turn, OH• can react with lysine, proline, arginine and threonine residues of proteins, introducing carbonyl groups such as aldehydes and ketones, modifications that can alter protein structure and function. Not all proteins are susceptible to oxidative damage to the same degree. Studies on aging and disease have shown that mitochondrial proteins such as aconitase and adenine nucleotide translocase are more vulnerable to oxidative damage than others (Yan et al., 1997; Yan and Sohal, 1998). Mitochondrial aconitase participates in ribosomal binding and mRNA turnover and degradation, and its ironsulfur clusters are highly sensitive to oxidation by O2•–, while mitochondrial adenine 53  nucleotide translocase catalyzes the exchange of ADP for ATP across the inner mitochondrial membrane at complex V of the ETC. Oxidative modifications of these proteins, as part of the aging process, can lead to the inactivation of various catalytic functions, and in turn, to an abnormal energy metabolism. Protein modifications can lead to detrimental intermolecular aggregates, making their degradation by the proteosomal system difficult, if not impossible. In BD, besides the suggested mitochondrial dysfunction mechanism that can lead to increased protein carbonylation and accumulation through decreased complex I activity, the normal process of protein degradation may be impaired, leading to further accumulation. Findings on proteosomal system abnormalities from prefrontal cortex from patients with BD, found decreased expression of several genes involved in the proteosome degradation process (Konradi et al., 2004), putting into perspective the deleterious consequences of an abnormally functioning ETC, and the importance of a properly functioning mitochondrion and proteosomal degradation system, as part of a healthy cellular metabolism. In fact, accumulation of modified proteins by carbonylation has been implicated in the aetiology, and in some cases, the progression of various chronic neurodegenerative disorders such as Alzheimer’s disease (AD). Protein oxidation products can affect cell growth and differentiation. Increased ROS production and oxidative alterations of brain proteins are characteristic of AD, suggesting a link between protein oxidation and neurodegeneration, as the oxidatively modified proteins can ultimately lead to neuronal death (Castegna et al., 2002). In BD, there have been numerous reports (Kato, 2008; Ongur et al., 1998; Rajkowska et al., 2001) of decreased neuronal and glial density that could perhaps be attributed to these protein modifications, highlighting the importance  54  of preventing oxidative damage to proteins and neuronal vulnerability to cellular stress that could ultimately result in cell death. Future studies will be needed in order to identify if specific mitochondrial proteins can be targets of protein oxidation in BD. Furthermore, defining the relationship between oxidative protein modification, such as carbonylation, and cellular function, will be critical for our better understanding of the disorder, as well as to establish the link between oxidative modifications and neuronal death in BD. An alternate pathway leading to protein damage in pathological situations, is one in which excess amounts of O2•– generated in cellular mitochondria, can react with the readily available nitric oxide (NO•). NO•, a diffusible gas molecule involved in neurotransmission, regulation of vascular relaxation, and part of the inflammatory response (Beal, 2002), is generated from the amino acid arginine by the enzyme nitric oxide synthase. NO• becomes detrimental to cells when it finds its way to react with excess amounts of O2•– that manage to escape from mitochondria, due to electrons leaking from complex I reacting with molecular oxygen, leading to the generation of peroxinitrite (ONOO–). This compound has the capacity to act in a similar manner as the hydroxyl radical, OH•, to induce lipid and protein oxidation, and in the presence of the amino acid residue tyrosine, it can form 3-nitrotyrosine, a pathological cellular process leading to protein nitrosative damage (Figure 1). In addition to having the capacity to induce nitrosative damage to tyrosine residues, ONOO– plays a role in neuronal damage associated with excitotoxicity (Baranano and Snyder, 2001). Experiments have shown that when ONOO– reacts with mitochondrial membranes from bovine heart, the reaction occurs predominantly in complex I subunits, underlying the importance of nitration and complex I functionality (Murray et al., 2003). In turn, ONOO– is able to 55  significantly inhibit complex I activity, implying a functional relationship between complex I activity and nitration (Andreazza et al., 2010). In addition, incubation of SHSY5 cells with ONOO– is able to increase 3-nitrotyrosine levels in mitochondrial complex I subunits, and in no other mitochondrial proteins (Naoi et al., 2005). Recent reports show a negative correlation between complex I activity and levels of 3nitrotyrosine in prefrontal cortex of patients with BD (Andreazza et al., 2010). All together, these findings suggest a susceptibility of mitochondrial complex I to nitration, contributing to the mitochondrial dysfunction present in BD. ROS and RNS generated in mitochondria are able to modify bioactive molecules, such as proteins and DNA, either directly or indirectly, with lipid or carbohydrate peroxidation products (Naoi et al., 2005). Furthermore, mitochondria are considered to play a critical role in apoptosis (Kroemer et al., 1998) or programmed cell death, from evidence gathered on changes in energy charge and redox, disruption of membrane potential and release of cytochrome-c that occur in mitochondria, prior to DNA fragmentation, the key event in apoptosis morphology (Naoi et al., 2005). Overall, oxidative stress is associated with a great number of pathological conditions that include cancer (Halliwel, 2007), diabetes (Robertson et al., 2006), neurodegenerative diseases such as Parkinson’s and Alzheimer’s disease (Naoi et at., 2005), in addition to heart failure and disease (Ide et al., 2000). There is growing evidence that oxidative stress is increased in patients with BD. Furthermore, increased levels of carbonylated proteins have been shown to be negatively correlated with complex I activity, and increased levels of 3-nitrotyrosine have been reported in the prefrontal cortex of patients with BD (Andreazza et al., 2010). 56  Examining the levels of oxidative damage and antioxidant system responses in blood samples from patients with BD, Kuloglu et al reported a significant increase in SOD levels, in addition to increased levels of malondialdehyde (MDA), a lipid peroxidation product. Increase levels of SOD, the enzyme responsible for clearing O2•– from the cell and generating H2O2, suggest that increased amounts of O2•– are being generated by mitochondria, in addition to excess amounts of H2O2, that can contribute to lipid peroxidation by OH•. This in turn, is supported by findings showing a lack of increase in GSH-Px levels, an increase that is necessary in order to respond efficiently to the increase in turnover rate of H2O2 that can lead to oxidative damage (Kuloglu et al., 2002). Furthermore, increased oxidative stress has been reported in the anterior cingulate cortex of subjects with BD by measuring levels of 4-HNE, a lipid peroxidation product (Wang et al., 2009), suggesting that lipid peroxidation is increased in BD. The complex I alterations observed in the group of lithium responders from our study indicate a compensatory mechanism by which these patients might be able to decrease the turnover rate of O2•–, leaked from ETC sites such as complex I, in order to prevent oxidative and nitrosative damage to peripheral cells. The absence of mitochondria dysfunction at the level of complex I in this subset of patients with BD, could provide a measure of the patients’ response to lithium and their success in managing the disorder, or how lithium in the long term, is able to prevent oxidative cellular stress. Lithium has been reported to increase levels of antioxidant enzymes such as catalase and SOD, enzymes important in the prevention of protein oxidative damage (MachadoVieira et al., 2007). In addition, chronic lithium treatment at therapeutically relevant concentrations inhibits glutamate-induced lipid peroxidation and protein oxidation, by 57  preventing the increase of intracellular calcium concentrations, in addition to DNA fragmentation and cell death, in rat cerebral cortical neurons (Shao et al., 2005). Finding such as these ones suggest that mood stabilizers might have neuroprotective effects against excitotoxicity due to oxidative stress inhibition. In addition, antioxidative stress may play an important role in the pharmacological treatment of BD (Shao et al., 2005). Other reports showed increased expression of the antiapoptotic factor bcl-2 in both rat brain and PC-12 cells under chronic lithium treatment (Chen et al., 1999; Hiroi et al., 2005). Bcl-2 stabilizes the mitochondrial inner transmembrane potential, inhibiting neuronal cell death by decreasing the production of ROS (Kane et al., 1993). All together, these findings suggest that lithium is able to stabilize the mitochondrial inner membrane at multiple sites, and reduce the concentration of intracellular calcium, decreasing the accumulation of glutamate-induced ROS (Shao et al., 2005). Mood stabilizers are also believed to restore the balance among aberrant signalling pathways in certain brain regions and prevent degeneration, by stimulating bcl-2 expression and inhibiting GSK-3β activity, a regulatory enzyme in the cell survival signalling pathway (Andreazza et al., 2007). Furthermore, lithium can exert protective effects against cell death in cultured human cells, induced by oxidative stress, a protective effect that seems to depend on bcl-2 up-regulation (Lai et al., 2006). Therefore, one possible explanation for the lack of significant differences in oxidative and nitrosative damage to proteins reported in our study, across all groups, might have to do with the lithium response characteristic of our group of patients with BD. In addition to the evidence on lithium’s neuroprotective and antioxidant properties already reported, that provide strong evidence to support our results, there are other studies on tyrosine nitration-induced  58  damage in serum samples from patients with BD that can provide additional support for the antioxidant lithium response in our subset of BD patients. Andreazza et al reported a significant increase in 3-nitrotyrosine content in serum samples from patients with BD in both early and late stages of the illness, compared to control subjects (Andreazza et al., 2009). The levels of 3-nitrotyrosine reported in our samples of lithium responsive BD patients did not differ to a significant extent from those reported in affected and unaffected relatives, and nonpsychiatric controls. These results are indicative of an association between the lithium response and the lack of oxidative and nitrosative damage. There are reports showing that the response to long term lithium treatment clusters in families of lithium responders to a significant extent, reports that can give way into future studies on whether this clustering is due to the lithium response, the illness itself, or other factors (Grof et al., 2002). Studies such as these ones are necessary in order to understand the similarities and differences in the genetics of relatives of BD, and the mechanisms by which they are able to prevent the manifestation of the disorder. The findings on complex I alterations and protein oxidative damage from transformed lymphoblasts from lithium responsive patients with BD reported in our study, differ from those gathered from prefrontal cortex samples of patients with BD (Andreazza et al., 2010). Interestingly, using the same set of lymphoblast samples as those from this study, our lab reported a similar trend when looking at pCREB levels of lithium responsive patients with BD, their unaffected and affected relatives, from what we observed in our measurements on complex I activity. The phosphorylated CREB protein, pCREB, is an active molecule initiating binding to specific sites in promoter 59  regions of target genes, leading to the production of mRNA, the blueprint for protein synthesis (Bezchlibnyk and Young, 2002). The process of CREB phosphorylation begins at the level of G-protein couple receptors’ activation, through the G protein α subunit (Gαs). Lower Gαs levels have been reported in post mortem brain tissue samples from subjects with BD treated with lithium, in addition to decreased CREB levels in BD subjects taking anticonvulsants, at the time of death (Dowlatshahi et al., 1999). Recently, our lab completed a measure on pCREB levels in lymphoblasts from lithium responsive BD patients, and found a significant increase in basal pCREB levels when compared to controls subjects (Alda et al., in press), suggesting that CREB phosphorylation may play an important role in the pathophysiology of BD. However, the outcome is markedly different in peripheral cells, when compared to central nervous system tissues. In addition, Alda et al found that basal pCREB levels were also increased in unaffected and affected relatives of lithium responders when compared to control subjects, suggesting a genetic association between CREB phosphorylation and lithium responsive patients with BD. Similarly, our results showed an increase in complex I activity in unaffected and affected relatives from lithium responders, compared to controls. Again, the possibility of potential endophenotypes in complex I activity exists, as in the case of pCREB, for lithium responders. Findings such as these ones highlight the importance of mood stabilizer effects on the treatment of BD, and how medication can excert its effects in different ways in brain and peripheral tissues. In the future, when treating our samples with lithium, could lead to a further increase in complex I activity across all groups, providing evidence into the mechanisms of action of lithium in BD. A compensatory biochemical mechanism, such as increased levels of  60  pCREB to up-regulate gene transcription, including the genes coding for complex I subunits responsible for the measured increase in complex I activity, may be occurring in patients that respond well to mood stabilizers such as lithium. Furthermore, Sun et al., 2006 reported the possibility of complex I as a target for lithium, evidence gathered from experiments looking at the effects of lithium-treated cells on the expression of ETC subunits such as NDUFS7 at the time of death in subjects with BD. Lithium increased the expression of NDUFS7 to a significant degree in subjects with BD, suggesting that this mood stabilizer could target complex I as part of its pharmacological efficacy in treating BD (Sun et al., 2006). Under chronic lithium treatment, our lymphoblast samples may show an increase in NDUFS7 levels, as previously reported by Sun et al, that could account for the expected increase in complex I activity. These biochemical alterations associated with lithium, could play a role in preventing the occurrence of mania and depression, the mood states characteristic of the disorder. It would be crucial to explore the up-regulation of complex I subunit NDUFA5 reported by Naydenov et al, and its relationship with increased complex I activity, as well as other possible subunits’ genes, such as those for NDUFS7, that may be up-regulated during gene transcription in our lymphoblast samples under chronic lithium treatment. Experiments such as these ones could explain the alterations in complex I observed in lymphoblasts from lithium responders and their relatives, in order to give insights into the processes responsible for the prevention of complex I dysfunctionality in BD. In psychiatric research, studies on cultured lymphoblasts from patients with BD have some advantages over studies on post mortem brain tissue, in the sense that lymphocytes are easily obtainable from live patients whose response to lithium can be 61  isolated, and the possibility for comparisons before and after drug treatment in the same sample can lead to interesting findings. Nevertheless, the biological response when comparing untransformed lymphocytes and Epstein-Bar virus (EBV) transformed lymphoblasts, may not represent the changes that can occur in particular brain regions (Sun et al., 2004). Yet, over the years, a number of studies on transformed lymphoblasts from patients with BD have been published, reporting similarities in biological responses between central nervous system tissue and peripheral cells (Schreiber et al., 1991; Young et al., 1993; Young et al., 1994). More specifically, BDNF levels in serum and cortex have been shown to be positively correlated in rats (Karege et al., 2002) in addition to reports of its association with the expression of Nacetylaspartate, a neuronal integrity marker, in the cerebral cortex of human subjects (Lang et al., 2007). Despite some of the limitations of using lymphoblasts in psychiatric research, they have the incredible advantage of allowing for the study of supposed biomarkers in their native state, free of the influence of medications, hormones or other factors that can potentially lead to misinterpretations and inaccurate results. We reported a trend towards up-regulation of complex I activity in lithium responders and a significant up-regulation of complex I activity in unaffected relatives compared to both lithium responders and nonpsychiatric controls, findings that can be a crucial factor in understanding the metabolic differences that can prevent these subjects from developing the disorder. In this respect, we expected to find similar results between the unaffected relatives and the nonpsychiatric controls, yet the highly significant difference in complex I activity between these group of subjects with no history of psychiatric illnesses, opens the door to argue for a compensatory mechanism where unaffected 62  relatives of lithium responsive BD patients are able to up-regulate peripheral complex I activity, one of the ways in which their cells are able to protect against the expression of the bipolar phenotype, in this genetically predisposed group of subjects. It is worth mentioning that the significant increase in complex I activity observed in the affected relatives, compared to both nonpsychiatric controls and their BD counterparts, should be interpreted with caution, as these relatives were afflicted by other psychiatric illnesses such as major depression, besides BD. In addition to the compensatory mechanism that may be present in the group of unaffected relatives, in terms of complex I activity, that may prevent the manifestation of the disorder’s phenotype, the trend towards increased complex I activity seen in our group of lithium responders, could also be seen as a compensatory mechanism at play, for the prevention of protein oxidative and nitrosative damage, as this group of lithium responders may be able to reduce the levels of ROS and RNS normally produced by mitochondria dysfunction, by preventing complex I dysfunctionality. The possible mechanism at play here is a decrease in the generation of O2•– by mitochondrial complex I due to an increase in its activity. There are several limitations in our study, the results need to be interpreted with caution. Peripheral cells such as lymphocytes are very approximate models of central nervous system processes. However, they do not necessarily express the same proteins as those found in neurons, and different brain regions may differ in terms of complex I functionality, subunit expression, and levels of oxidative stress. Furthermore, it is not entirely clear what the impact of cell transformation into lymphoblasts and its subsequent cultivation may have on the measured variables, and whether this 63  transformation affects cells in the same way across subject groups. Nevertheless, cell lines are the one of the few model tissues with the same DNA sequence as neurons from specific and well-characterized living subjects, to date (Alda et al., in press). In conclusion, our results provide evidence of a compensatory mechanism for the upregulation of complex I activity in unaffected relatives of patients with BD, and to a lesser extent in lithium responsive patients with BD, that may prevent protein oxidative and nitrosative damage observed in BD, by our lab and others. With increased knowledge of the neuroprotective effects of lithium and its antioxidant properties (Wang et al., 2007), future experiments measuring complex I activity and NDUFS7 levels in transformed lymphoblasts under chronic lithium treatment, may provide insight into the possibility of lithium’s ability to up-regulate complex I activity, decreasing ROS and RNS production in mitochondria, preventing oxidative damage to proteins. We would expect to find a significant increase in complex I activity across all groups after lithium treatment, compared to the complex I measurements reported in our study without mood-stabilizer treatment. In addition, it would be possible to find a significant decrease in protein carbonyl and 3-nitrotyrosine levels after lithium treatment, confirming lithium’s antioxidant properties, by targeting complex I in order to reduce the production of ROS and RNS, preventing protein oxidative damage in BD. Furthermore, it would be interesting to investigate other complex I subunits besides NDUFS7 that could be potential targets for lithium, in order to see if their levels are increased and correlate positively with complex I activity, after chronic lithium treatment. Since mitochondrial complex III is another major source of O2•– in the cell, measuring complex III activity would be crucial in order to gain further insight into the metabolic characterization of 64  lithium responsive patients with BD that can contribute to their pharmacological treatment success. Understanding the underlying mechanisms of lithium treatment for BD in this group of lithium responders is of extreme importance. Increasing our knowledge with respect to mitochondrial functionality at the level of complex I in this subset of patients, can lead into the characterization of a complex I phenotype of lithium responsive BD patients.  65  References Allison, J.H., Blisner, M.E., Holland, W.H., Hipps, P.P., Sherman, W.R. Increased brain myo-inositol 1-phosphate in lithium-treated rats. Biochem Biophys Res Commun. 71, 664-670 (1976). Alda, M., Shao, L., Wang, J.F., Lopez de Lara, C., Jaitovich-Griosman, I., Lebel, V., Sun, X., Duffy, A., Grof, P., Rouleau, G., Turecki, G., Young, L.T. Alterations in pCREB signalling: an endophenotype of lithium-responsive bipolar disorder? Int J Neuropsychopharmacol. In press, (2011). Alspaugh, M.A., Jensen, F.C., Rabin, H., Tan, E.M. Lymphocytes transformed by Epstein-barr virus: Induction of nuclear antigen reactive with antibody in rheumatoid arthritis. J Exp Med. 147, 1018-1027 (1978). Anderson, M.A., Gusella, J.F. Use of cyclosporine H in established Epstein-Barr virustransformed human lymphoblastoid cell lines. In Vitro. 20, 856-858 (1984). Andreasen, N.C., Endicott, J., Spitzer, R.L., Winokur, G. The family history method using diagnostic criteria. Arch Gen Psychiatry. 34, 1229-1235 (1977). Andreazza, A.C., Frey, B.N., Erdtmann, B., Salvador, M., Rombaldi, F., Santin, A., Gonҫalves, C.A., Kapczinski, F. DNA damage in bipolar disorder. Psychiatry Res. 153, 27-32 (2007). Andreazza, A.C., Kapczinski, F., Kauer-Sant’Anna, M., Walz, J.C., Bond, D.J., Gonҫalves, C.A., Young, L.T., Yatham, L.N. 3-Nitrotyrosine and glutathione antioxidant system in patients in the early and late stages of bipolar disorders. J Psychiatry Neurosci. 34, 263-71 (2009). Andreazza, A.C., Shao, L., Wang, J.F., Young, L.T. Mitochondrial complex I activity and oxidative damage to mitochondrial proteins in the prefrontal cortex of patients with bipolar disorder. Arch Gen Psychiatry. 67, 360-368 (2010). Baranano, D.E., Snyder, S.H. Neural roles for heme oxygenase: contrasts to nitric oxide synthase. Proc Natl Acad Sci. USA. 98, 10996-11002 (2001). Beal, M.F. Oxidatively modified proteins in aging and disease. Free Radic Biol Med. 32, 797-803 (2002). Beckman, J.S. Oxidative damage and tyrosine nitration from peroxynitrite. Chem Res Toxicol. 9, 836-844 (1996).  66  Benes, F.M., Vincent, S.L., Todtenkopf, M. The density of pyramidal and nonpyramidal neurons in anterior cingulated cortex of schizophrenic and bipolar subjects. Biol Psychiatry. 50, 396-406 (2001). Bezchlibnyk, Y., Young, L.T. The neurobiology of bipolar disorder: focus on signal transduction pathways and the regulation of gene expression. Can J Psychiatry. 47, 135-148 (2002). Bonni, A., Brunet, A., West, A.E., Datta, S.R., Takasu, M.A., Greenberg, M.E. Cell survival promoted by the Ras-MAPK signalling pathway by transcription-dependent and –independent mechanisms. Science. 286, 1358-1362 (1999). Bowley, M.P. Drevets, W.C., Ongür, D., Price, J.L. Low glial numbers in the amygdale in major depressive disorder. Biol Psychiatry. 52, 404-412 (2002). Brandt, U., Kerscher, S., Dröse, S., Zwicker, K., Zickermann, V. Proton pumping by NADH:ubiquinone oxidoreductase. A redox driven conformational change mechanism? FEBS Lett. 545, 9-17 (2003). Cassina, A.M., Hondara, R., Souza, J.M., Thomson, L., Castro, L., Ischiropoulos, H., Freeman, B.A., Radi, R. Cytochrome c nitration by peroxynitrite. J Biol. Chem. 275, 409415 (2000). Castegna, A., Aksenov, M., Aksenova, M., Thongboonkerd, V., Klein, J.B., Pierce, W.M., Booze, R., Markesbery, W.R., Butterfield, D.A. Proteomic identification of oxidatively modified proteins in Alzheimer’s disease brain. Part I: Creatine kinase BB, glutamine synthase, and ubiquitin carboxy-terminal hydrolase L-1. Free Radic Biol Med. 33, 562-571 (2002). Chen, G., Zeng, W.Z., Yuan, P.X., Huang, L.D., Jiang, Y.M., Zhao, Z.H., Manji, H.K. The mood-stabilizing agents lithium and valproate robustly increase the levels of the neuroprotective protein bcl-2 in the CNS. J Neurochem. 72, 879-882 (1999). Cotter, D., Landau, S., Beasley, C., Stevenson, R., Chana, G., MacMillan, L., Everall., I. The density and spatial distribution of GABAergic neurons, labelled using calcium binding proteins, in the anterior cingulate cortex in major depressive disorder, bipolar disorder, and schizophrenia. Biol Psychiatry. 51, 377-386 (2002). Dowlatshahi, D., MacQueen, M., Wang, J.F., Reiach, J.S., Young, L.T. G proteincoupled cyclic AMP signalling in post-mortem brain of subjects with mood disorders. J Neurochem. 73, 1121-1126 (1999). Endicott, J., Spitzer, R.L. A diagnostic interview: the schedule for affective disorders and schizophrenia. Arch Gen Psychiatry. 35, 837-844 (1978). 67  Estornell, E., Fato, R., Palloti, F., Lenaz, G. Assay conditions for the mitochondrial NADH:coenzyme Q oxidoreductase. FEBS Lett. 332, 127-131 (1993). Frey, B.N., Andreazza, A.C., Kunz, M., Gomes, F.A., Quevedo, J., Salvador, M., Gonҫalves, C.A., Kapczinski,F. Increased oxidative stress and DNA damage in bipolar disorder: a twin-case report. Prog Neurophychopharmacol Biol Psychiatry. 31, 283-285 (2007). Grof, P., Duffy, A., Cavazzoni, P., Grof, E., Garnham, J., MacDougall, M., O’Donovan, C., Alda, M. Is response to prophylactic lithium a familial trait? J Clin Psychiatry. 63, 942-947 (2002). Halliwell, B. Oxidative stress and cancer: have we moved forward? Biochem J. 401, 111 (2007). Halliwell, B. What nitrates tyrosine? Is nitrotyrosine specific as a biomarker of peroxynitrite formation in vivo? FEBS Lett. 411, 157-160 (1997). Halliwell, B., Gutteridge, J.M.C. Free radicals in biology and medicine. Oxford University Press. (2007). Hansen, T.O., Rehfeld, J.F., Nielsen, F.C. GSK-3beta reduces cAMP-induced cholecystokitin gene expression by inhibiting CREB binding. Neuroreport. 15, 841-845 (2004). Harwood, A.J. Lithium and bipolar mood disorder: the inositol-depletion hypothesis revisited. Mol Psychiatry. 10, 117-126 (2005). Hiroi, T., Wei, H., Hough, C., Leeds, P., Chuang, D.M. Protracted lithium treatment protects against the ER stress elicited by thapsigargin in rat PC12 cells: Roles of intracellular calcium, GRP78 and bcl-2. Pharmacogenomics J. 5, 102-111 (2005). Ide, T., Tsutsui, H., Kinugawa, S., Seumatsu, N., Hayashidani, S., Ichikawa, K., Utsumi, H., Machida, Y., Egashira, K., Takeshita, A. Direct evidence for increased hydroxyl radicals originating from superoxide in the failing myocardium. Circ Res. 86, 152-157, (2000). Inhorn, R.C., Majerus, P.W. Inositol polyphosphate 1-phosphatase from calf brain. Purification and inhibition by Li+, Ca2+, and Mn2+. J Biol Chem. 262, 15946-15952 (1987). Iwamoto, K., Bundo, M., Kato, T. Altered expression of mitocondrial-related genes in postmortem brains of patients with bipolar disorder or schizophrenia, as revealed by large-scale DNA microarray analysis. Hum Mol Genet. 14, 241-253 (2005). 68  Iwamoto, K., Kakiuchi, C., Bundo, M., Ikeda, K., Kato, T. Molecular characterization of bipolar disorder by comparing gene expression profiles of postmortem brains of major mental disorders. Mol Psychiatry. 9, 406-416 (2004). Kane, D.J., Sarafian, T.A., Anton, R., Hahn, H., Gralla, E.B., Valentine, J.S, Ord. T., Bredesen, D.E. Bcl-2 inhibition of neural death: Decreased generation of reactive oxygen species. Science. 262, 1274-1277 (1993). Karege, F., Schwald, M., Cisse, M. Postnatal developmental profile of brain-derived neurotrophic factor in rat brain and platelets. Neurosci Lett. 328, 261-264 (2002). Kato, T., Inubushi, T., Kato, N. Magnetic resonance spectroscopy in affective disorders. J Neuropsychiatry Clin Neurosci.10, 133-147 (1998). Kato, T., Kato, N. Mitochondrial dysfunction in bipolar disorder. Bipolar Disord. 3, 180190 (2000). Kato, T., Kuratomi, G., Kato, N. Genetics of bipolar disorder. Drugs Today (Barc.). 41, 335-344 (2005). Kato, T. Molecular genetics of bipolar disorder and depression. Psychiatry Clin Neurosci. 61, 3-19 (2007). Kato, T. Molecular neurobiology of bipolar disorder: a disease of ‘mood stabilizing neurons’? Trends Neurosci. 10, 495-503 (2008). King, T.D., Jope, R.S. Inhibition of glycogen synthase kinase-3 protects cells from intrinsic but not extrinsic oxidative stress. Neuroreport. 16, 597-601 (2005). Konradi, C., Eaton, M., MacDonald, M.L., Walsh, J., Benes, F.M., Heckers, S. Molecular evidence for mitochondrial dysfunction in bipolar disorder. Arch Gen Psychiatry. 61, 300-308 (2004). Kopnisky, K.L., Chalecka-Franaszek, E., Gonzalez-Zuleta, M., Chuang, D.M. Chronic lithium treatment antagonizes glutamate-induced decrease of phosphorylated CREB in neurons via reducing protein phosphatise 1 and increasing MEK activities. Neuroscience. 116, 425-435 (2003). Kroemer, G., Dallaporta, B., Resch-Rigon, M. The mitochondrial death/life regulator in apoptosis and necrosis. Annu Rev Physiol. 60, 619-642 (1998). Kuloglu, M., Ustundag, B., Atmaca, M., Canatan, H., Tezcan, E., Cinkilinc, N. Lipid peroxidation and antioxidant enzyme levels in patients with schizophrenia and bipolar disorder. Cell Biochem Funct. 20, 171-175 (2002).  69  Lai, J.S., Zhao, C., Warsh, J.J., Li, P.P. Cytoprotection by lithium and valproate varies between cell types and cellular stress. Eur J Pharmacol. 539, 18-26 (2006). Lang, U.E., Hellweg, R., Seifert, F., Schubert, F., Gallinat, J. Correlation between serum brain-derived neurotrophic factor level and an in vivo marker of cortical integrity. Biol Psychiatry. 62, 530-535 (2007). Lenaz, G. The mitochondrial production of reactive oxygen species: mechanisms and implications in human pathology. Life. 52, 159-164 (2001). Mahadik, S.P., Mukherjee, S. Free radical pathology and antioxidant defense in schizophrenia: a review. Schizophrenia Res. 19, 1-17 (1996). Maher, P., Schubert, D. Signalling by reactive oxygen species in the nervous system. Cell Mol Life Sci. 57, 1287-1305 (2000). Mamdani, F., Alda, M., Grof, P., Young, L.T., Rouleau, G., Turecki, G. Lithium response and genetic variation in the CREB family of genes. Am J Med Genet B Neuropsychiatr Genet. 147B, 500-504 (2008). Murray, J., Taylor, S.W., Zhang, B., Ghosh, S.S., Capaldi, R.A. Oxidative damage to mitochondrial complex I due to peroxynitrite: identification of reactive tyrosines by mass spectrometry. J Biol Chem. 278, 37223-37230 (2003). Naoi, M., Marayama, W., Shamoto-Nagai, M., Yi, H., Akao, T., Tanaka, M. Oxidative stress in mitochondria: decision to survival and death of neurons in neurodegenerative disorders. Mol Neurobiol. 31, 81-93 (2005). Öngür, D., Drevets, W.C., Price, J.L. Glial reduction in the subgenual prefrontal cortex in mood disorders. Proc Natl Acad Sci USA. 95, 13290-13295 (1998). Okada, K., Wangpoengtrakui, C., Osawa, T., Toyokuni, S., Tanaka, K., Ucida, K. 4Hydroxy-2-nonenal-mediated impairment of intracellular proteolysis during oxidative stress. J Biol Chem. 274, 787-793 (1992). Ozawa, T. Mitochondrial DNA mutations and age. Ann New York Acad Sci. 854, 128154 (1998). Papa, L., Rockwell, P. Persistent mitochondrial dysfunction and oxidative stress hinder neuronal cell recovery from reversible proteasome inhibition. Apoptosis. 13, 588-599 (2008). Perlis, R.H., Smoller, J.W., Ferreira, M.A.R., McQuillin, A., Bass, N., Chir, B., Lawrence, J., Sachs, G.S., Nimgaonkar, V., Scolnick, E.M., Gurling, H., Sklar, P., Purcell, S. A  70  genomewide association study of response to lithium for prevention of recurrence in bipolar disorder. Am J Psychiatry. 166, 718-725 (2009). Rajkowska, G., Halaris, A., Selemon, L.D. Reductions in neuronal and glial density characterize the dorsolateral prefrontal cortex in bipolar disorder. Biol Psychiatry. 49, 741-752 (2001). Rajkowska, G. Cell pathology in bipolar disorder. Semin Clin Neuropsychiatry. 4, 105116 (2002). Ricci, C., Pastukh, V., Leonard, J., Turrens, J., Wilson, G., Schaffer, D., Schaffer, S.W. Mitochondrial DNA damage triggers mitochondrial-superoxide generation and apoptosis. Am J Physiol Cell Physiol. 294, 413-422 (2008). Rivett, AJ. Regulation of intracellular protein turnover: covalent modification as a mechanism of marking proteins for degradation. Curr Top Cell Regul. 28, 291-337 (1986). Robertson, R.P., Harmon, J.S. Diabetes, glucose toxicity, and oxidative stress: a case of double jeopardy for the pancreatic islet beta cell. Free Rad Biol Med. 41, 177-184 (2006). Rosenzweig, M.R., Breedlove, S.M., Watson, N.V. Biological psychology: an introduction to behavioural and cognitive neuroscience. Sinauer Associates Inc. (2005). Rowe, M.K., Chuang, D.M. Lithium neuroprotection: molecular mechanisms and clinical implications. Expert Rev Mol Med. 18, 1-18 (2004). Sanacora, G. New understanding of mechanisms of action of bipolar medications. J Clin Psychiatry. 69, 22-27 (2008). Schreiber, G., Avissar, S., Danon, A., Belmaker, R.H. Hyperfunctional G proteins in mononuclear leukocytes of patients with mania. Biol Psychiatry. 29, 273-280 (1991). Schulze, T.G., Alda, M., Adli, M., Akula, N., Ardau, R., Bui, E.T., Chillotti, C., Cichon, S., Czerski, P., Del Zompo, M., Detera-Wadleigh, S.D., Grof, P., Gruber, O., Hashimoto, R., Hauser, J., Hoban, R., Iwata, N., Kassem, L., Kato, T., Kittel-Schneider, S., Kliwicki, S., Kelsoe, J.R., Kusumi, I., Laje, G., Leckband, S.G., Manchia, M., MacQueen, G., Masui, T., Ozaki, N., Perlis, R.H., Pfennig, A., Piccardi, P., Richardson, S., Rouleau, G., Reif, A., Rybakowski, J.K., Sasse, J., Schumacher, J., Serevino, G., Smoller, J.W., Squassina, A., Turecki, G., Young, L.T., Yoshikawa, T., Bauer, M., McMahon, F.J. The international consortium on lithium genetics (ConLiGen): an initiative by the NIMH and IGSLI to study the genetic basis of response to lithium treatment. Neuropsychobiology. 62, 72-78 (2010). 71  Shao, L., Young, L.T., Wang, J.F. Chronic treatment with mood stabilizers lithium and valproate prevents excitotoxicity by inhibiting oxidative stress in rat cerebral cortical cells. Biol Psychiatry. 58, 879-884 (2005). Shieh, P.B., Hu, S.C., Bobb, K., Timmusk, T., Ghosh, A. Identification of a signaling pathway involved in calcium regulation of BDNF expression. Neuron. 20, 727-740 (1998). Spitzer, R.L., Endicott, J., Robins, E. Research diagnostic criteria: rationale and reliability. Arch Gen Psychiatry. 35, 773-782 (1978). Stadtman, E.R. Oxidation of free amino acids and amino acid residues in proteins by radiolysis and by metal-catalyzed reactions. Annu Rev Biochem. 62, 797-821 (1993). Strakowski, S.M., Delbello, M.P., Adler, C.M. The functional neuroanatomy of bipolar disorder: a review of neuroimaging findings. Mol Psychiatry. 10, 105-116 (2005). Sun, X., Young. L.T., Wang, J.F., Grof, P., Turecki, G., Rouleau, G.A., Alda, M. Identification of lithium-regulated genes in cultured lymphoblasts of lithium responsive subjects with bipolar disorder. Neuropsychopharmacology. 29, 799-804 (2004). Sun, X., Wang, J.F., Tseng, M., Young, L.T. Downregulation in components of the mitochondrial electron transport chain in the post-mortem frontal cortex of subjects with bipolar disorder. J Psychiatry Neurosci. 31, 189-196 (2006). Takeshige, K., Minakami, S. NADH and NADPH-dependent formation of superoxide anions by bovine heart submitochondrial particles and NADH-ubiquinone reductase preparation. Biochem. 180, 129-135 (1979). Tseng, M., Alda, M., Xu, L., Sun, X., Wang. J.F., Grof, P., Turecki, G., Rouleau, G., Young, L.T. BDNF protein levels are decreased in transformed lymphoblasts from lithium-responsive patients with bipolar disorder. J Psychiatry Neurosci. 33, 449-453 (2008). Turecki, G., Grof, P., Grof, E., D’Souza, V., Lebuis, L., Marineau, C., Cavazzoni, P., Duffy, A., Bétard, C., Zvolský, P., Robertson, C., Brewer, C., Hudson, T.J., Rouleau, G.A., Alda, M. Mapping susceptibility genes for bipolar disorder: a pharmacogenetic approach based on excellent response to lithium. Mol Psychiatry. 6, 570-578 (2001). Wang, J.F. Defects in mitochondrial electron transport chain in bipolar disorder: implications for mood-stabilizing treatment. Can J Psychiatry. 52, 753-762 (2007). Wang, J.F., Shao, L., Sun, X., Young, L.T. Increased oxidative stress in the anterior cingulated cortex of subjects with bipolar disorder and schizophrenia. Bipolar Disord. 11, 523-529 (2009). 72  Washizuka, S., Iwamoto, K., Kakiuchi, C., Bundo, M., Kato, T. Expression of mitochondrial complex I subunit gene NDUFV2 in the lymphoblastoid cells derived from patients with bipolar disorder and schizophrenia. Neurosci Res. 63, 199-204 (2009). Yan, L.J., Levine, R.L., Sohal, R.S. Oxidative damage during aging targets mitochondrial aconitase. Proc Natl Acad Sci USA. 94, 11168-11172 (1997). Yan, L.J., Sohal, R.S. Mitochondrial adenine nucleotide translocase is modified oxidatively during aging. Proc Natl Acad Sci USA. 95, 12896-12901 (1998). Young, L.T., Li, P.P., Kish, S.J., Siu, K.P., Kamble, A., Hornykiewick, O., Warsh, J.J. Cerebral cortex Gsα protein levels and forskolin-stimulated cyclic AMP formation are increased in bipolar affective disorder. J Neurochem. 61, 890-898 (1993). Young, L.T., Li, P.P., Kamble, A., Siu, K.P., Warsh, J.J. Mononuclear leukocyte levels of G-proteins in depressed patients with bipolar disorder or major depressive disorder. Am J Psychiatry. 151, 594-596 (1994).  73  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0071662/manifest

Comment

Related Items