Background
Bipolar disorder (BD) is a chronic and several psychiatric illness, characterized by recurrent episodes of mania and depression, and interspersed with remission period. Although many systems and pathways have been involved in the pathophysiology of BD together containing signal transduction pathways, various neurotransmitter abnormalities, mitochondrial and metabolic dysfunctions and so on [
1]; the definitive mechanisms of BD remains to be fully elucidated.
Mitochondria serve as key organelles in eukaryotic cells, well known for synthesizing adenosine triphosphate (ATP) from glucose by oxidative phosphorylation for energy production. Especially, the energy source of neurons relies primarily on mitochondrial oxidative phosphorylation. Additionally, mitochondria also play a vital role in calcium signaling, cell resilience, apoptosis, the regulation of reactive oxygen species (ROS) and DNA damage [
2‐
4]. Mitochondrial DNA (mtDNA) is a double-stranded DNA molecule with no introns and histones. Without protection of histones and due to limited DNA repair capacity, these mtDNA is very prone to oxidative or genotoxic injury. As a result, mitochondria may increase their own copy number to compensate for the defects or impairment. Excessive oxidative stresses in pathological conditions may induce structural and functional changes of mitochondrial [
5]. Postmortem, brain imaging, genetic and peripheral cells studies all support that mitochondrial dysfunction is associated with BD [
6‐
10]. Using magnetic resonance spectroscopy, some studies showed lower level of phosphocreatine, N-acetyl-aspartate and inorganic phosphate and inorganic phosphate, which are supposed to reflect impaired mitochondrial function in several brain regions of living BD patients [
6,
7]. There were lines of evidences revealing abnormal mitochondrial morphology in neurons of postmortem brain and peripheral cells of BD patients [
8,
9]. Additionally, many SNPs across the genome of mitochondria were associated with BD [
10,
11]. These observed changes maybe indicate an important role of mitochondria in the pathogenesis of BD.
Mitochondrial DNA copy number (mtDNAcn) that reflects the mitochondrial function and biogenesis could be measured per cell [
12]. Abnormal mtDNAcn has been suggested to associate with various psychiatric disorders [
13‐
15]. With regard to BD, few studies were conducted to address the possible role of mtDNAcn in BD. Vawter et al. firstly found the alteration of mtDNAcn in a post-mortem study [
16], but other three post-mortem studies got negative results [
17‐
19]. In peripheral blood leukocytes, Chang et al. revealed that euthymic BD patients had a lower mtDNAcn than control group [
20]; While de Sousa et al. found no significant differences in mtDNA content between depressed BD patients and health controls [
21]. Furthermore, the mtDNA content still did not have a significant decrease after lithium carbonate treatment.
Mitochondrial dysfunction may be a cause of BD symptoms. Given the heterogeneity of the disease itself and the inconsistent results from studies, it is necessary to find more evidence to interpret the effect of mitochondrial dysfunction in BD. Human leukocytes have been proved to reflect virtually the plasma oxidative damage to DNA [
22]. Heretofore, no studies have assessed the mtDNA content in leukocytes of patients suffering mania or different states of BD. In this study, mtDNAcn was measured in BD patients with three different states and healthy individuals to identify how mtDNA content is altered in different mood states. We hypothesized that mtDNAcn variation may associate with BD, and would differentiate mood states as well as bipolar euthymia from healthy controls.
Discussion
To our knowledge, this is the first study to examine mtDNA content across the depressed, manic and euthymic states in BD patients. This current research has three main findings. (1) The acute BD patients had markedly lower mtDNA amount than healthy controls, after controlling covariates; there were no differences in mtDNAcn between manic patients and depressed patients. (2) The euthymic group showed a comparable content of mtDNA with healthy control group. (3) The mtDNAcn was inversely correlated with the number of relapses in BD subjects during manic episodes.
Change of mtDNAcn is an important sensitive index of mitochondrial dysfunction and oxidative stress [
29]. Our findings indicated that there may be explicit mitochondrial impairments during episodes of BD. In the current study, we firstly reported decreased mtDNA amount in peripheral leukocytes of BD patients undergoing manic episodes. Our data suggested, manic group may display a similar level of mtDNA with depressive group. In contrast, there was nearly normal level of mtDNA in euthymic patients. Hence, we speculate that, leukocyte mtDNA content may change with mood states in BD. Nicod et al. suggested the mtDNA content of patients changed with time during 8-week treatment in major depression [
30]. Acute patients could recover after treatment, and gradually transit to euthymic states. According to the findings and clinical observations above, the mtDNAcn of acute patients in peripheral blood might be gradually restored to some extent as time goes on.
The results of prior data on mtDNAcn in peripheral leukocytes and brain tissues of mood disorder were not consistent. Compatible with our result, de Sousa et al. showed a slight decreased mtDNAcn in depressive patients with BD type I, although the differences just failed to be significant (
P = 0.05), due to a small sample (
N = 7) [
21]. Chang et al. described that mtDNAcn was still significantly lower in the BD euthymic patients [
20]. One plausible explanation is that mtDNAcn was negatively correlated with age in BD [
21], and their subjects were older than ours. Accumulated evidences demonstrate BD may be an accelerated aging disease [
31]. Aging was also reported to be associated with the downregulation of mtDNA-associated genes and a decline level of mtDNAcn [
32]. Therefor accelerated aging in BD may affect the levels of mtDNA. According to threshold hypothesis of mtDNAcn control [
12], low mtDNAcn triggers the upregulation of mtDNA replication and high mtDNAcn triggers the machinery to degradate mtDNA. Taken together, low threshold and accelerated aging may induce the decrease of mtDNAcn. Regarding mtDNAcn in postmortem brains, three out of four studies reported a negative result in BD [
16‐
19]. It was not clear about what kinds of mood states that those subjects suffered from. Thus, the comparison cannot be made with ours. Additional, another innegligible factor is using different tissues or cells in studies, which may confound the results. .
MtDNA polymerase gamma (POLG) is the only one polymerase responsible for the mtDNA replication. POLG is prone to oxidative damage,and its deleterious variants were suggested as a risk for BD [
33]. Munkholm et al. revealed downregulation of POLG expression in BD across mania, hypomania, depression or mixed states in peripheral leukocyte [
34]. We suppose that POLG downregulation in acute phases of BD may be another important point of resulting in decrease of mtDNAcn.
There are also several evidences for abnormal mtDNA amounts in major depression [
35‐
38]. A large sample multicenter study reported patients with major depression showed increased mtDNA level than controls in saliva and blood [
37]. Two studies found a lower peripheral mtDNA level in both depressive and euthymic states [
35,
38]. Nicod et al. suggested mtDNA content of patients changed with time in major depression [
30].Overall it is not consistent about changes of mtDNA content in major depression and bipolar depression. However all studies indicate the importance of mtDNA in mood disorder. Future research is warranted to explore the relationship in different states of mood disorders.
The relationship observed between mtDNAcn and the number of episodes in manic patients deserves attention. In the current study, we found that mtDNA content was negatively correlated with the number of episodes in BD patients. BD patients often experience recurrent episodes or relapses which appear to cause chronicity of this illness and vulnerability for high stress response [
39]. The subjects suffering manic episode are at a high stress level [
40]. The “sensitivity” state of oxidative stress may accelerate illness course, and cause cell deficits as well as rapid decline of mitochondrial function.. However, similar findings were not found in the depressive subjects of this study. Hence, the results should be interpreted cautiously, due to the heterogeneity and complexity of BD.
Although antipsychotic agents and mood stabilizers were reported to exert some effect on mitochondrial function [
4,
41‐
43]. de Sousa and our previous studies showed that six or eight weeks lithium or risperidone treatment have no significant effect on mtDNAcn in subjects with psychiatric disorders [
13,
21]. Accordingly, we suppose the mtDNAcn of the acute patients would make no significant changes after short–term treatment. Whereas our euthymic patients were medically stabilized for more than 6 months, and may have enough time to recover from various damage in cell.
There are several limitations to be considered. Since we didn’t have more clinical data, including smoking status, glucose intolerance, and serum lipid, the possible influence of these variables cannot be completely excluded. Although using peripheral blood is more convenient and less invasive to quantify mtDNAcn, these finding cannot be extrapolated mechanically to other tissues. In addition, only mtDNAcn was measured in this cross-sectional data, it does not indicate any information about causal relationship between oxidative stress, mtDNA variations, and BD. More biological parameters and longitudinal studies regarding mitochondria involved in the pathophysiology of BD are warranted in future.