Parth V. Joshi

Research Paper Draft

Bipolar disorder is a major mood disorder showing manic and depressive episodes, which frequently accompany psychotic symptoms. Because genetic factors are known to contribute to this disorder based on twin studies among others, genetics-based animal models would be useful to understand its neurobiology. Because causative genes of bipolar disorder have not been identified yet, animal models of Mendelian genetic (Kato et al, 1992; Kato et al., 1994) diseases that frequently have comorbid bipolar disorder, such as Darier’s disease, Wolfram disease, and mitochondrial disease, would be one strategy.

In the 1990s, the scientists had studied the brain phosphorous metabolism in patients with bipolar disorder and found that phosphocreatine was decreased in the frontal lobe of patients with bipolar depression, decrease of phosphocreatine in the occipital cortex associated with stimulation was enhanced in a group of patients, intracellular pH in the frontal lobe and whole brain was decreased in the euthymic state. Although the results of recent 31P-MRS studies are not always consistent with each other, they also show some abnormalities of energy metabolism in bipolar disorder or improvement by treatment. After the proposal of mitochondrial dysfunction hypothesis of bipolar disorder, three groups performed structured interviews in patients with mitochondrial diseases and reported that the prevalence of bipolar disorder in mitochondrial diseases is 16–21%, nearly 20 times than general population. This suggests that having a mitochondrial disease is a strong risk factor for bipolar disorder.  (Kato et al, 1992; Kato et al., 1994)

Based on the initial findings, it could be seen that proposed mitochondrial dysfunction hypothesis of bipolar disorder was first discussed in 2000. In this hypothesis, the scientists proposed that mtDNA mutations, as well as common variations, may confer a risk of bipolar disorder by affecting intracellular calcium signaling systems. In spite that calcium levels are maintained low in cells, a higher concentration of calcium is accumulated in two organelles, mitochondria, and endoplasmic reticulum, where calcium released from endoplasmic reticulum is taken up by mitochondria. Elevation of calcium levels has been reported in blood cells. Lithium, the most established mood stabilizer, acts on intracellular calcium signaling by inhibiting inositol monophosphatase enzyme and upregulating mitochondrial Bcl-2 on the mitochondrial outer membrane. (Kato et al, 2000)

Exome or whole genome sequencing also suggested a possible role of genes related to calcium signaling, though they did not show mutations significantly associated with bipolar disorder at the genome-wide significance level. These findings altogether suggest a role of intracellular calcium signaling in bipolar disorder. Genetic factors, such as rare transmitted mutations and polymorphisms, may contribute to dysfunctional intracellular calcium signaling in bipolar disorder, and mitochondria-related genes affecting mitochondrial calcium signaling would be one of these factors.

Since then, numerous studies on mitochondrial dysfunction in bipolar disorder have been published. Regulation of mitochondria-related genes and regulation of a subset of mitochondria-related genes in postmortem brains, increase of lactate in the brain, decrease of complex I in postmortem brains, an abnormal mitochondrial structure in cells of bipolar patients, and elevation of isocitrate in cerebrospinal fluid associated with impaired function of isocitrate dehydrogenase. A recent study showed that neurons derived from induced pluripotent stem cells of patients with bipolar disorder had hyperexcitability associated with upregulation of mitochondrial genes, increased mitochondrial membrane potential, and smaller size of mitochondria.

To develop an animal model of bipolar disorder based on mitochondrial dysfunction hypothesis, the scientists focused on the accumulation of mtDNA in the brain, which is observed in patients with bipolar disorder and CPEO. (Parker and Brotchie, et al, 2010). To this end, they introduced a point mutation, Polg1, the mtDNA polymerase, to remove nuclease activity and generate ΔmtDNA. After they finished the generation of mice with neuron-specific mutant Polg1, it was discovered that Polg1 is one of the causative genes of CPEO with comorbid depression. Although knock-in mice of gene D257A mutation of Polg1 were also reported, these mice show muscular impairment and are not suitable for behavioral analysis. (Van Goethem, et al, 2001).  The neuron-specific transgenic mice of D181A mutation of Polg1 (mPolg1 Tg mice) showed accumulation of mtDNA specifically in the brain. They did not show gross abnormality in sensory and motor functions and learning and memory. The mPolg1 Tg mice were found to have intracellular calcium signaling abnormality, shown by attenuation of G-protein-coupled receptor-mediated calcium increase in hippocampal neurons.

By an extensive behavioral analysis, the scientists found that the mPolg1 Tg mice show altered intra-day wheel running activity rhythm. This was improved by electroconvulsive stimulation. Furthermore, by observing the wheel running activity for one month, the scientists found that female mPolg1 Tg mice show the fluctuation of wheel running activity associated with the estrous cycle. This was flattened by lithium treatment.

The scientists further extended the length of behavioral observation to more than half a year. The female mice showed recurrent spontaneous hypoactivity episodes, lasting approximately 2 weeks. This disappeared after ovariectomy, suggesting a role for female hormones. It is well known that the prevalence of depression is twice as high in females than males and women with bipolar disorder experience more depressive episodes than males. One of the mechanisms proposed to explain the female bias is “limbic system hyperactivity” where emotional dysregulation may be caused by high binding of gonadal steroids in the limbic system. The mechanism of female-only depression-like episodes and their potential relationship between hormonal control of depression episodes will require future study. (Diflorio, Jones et al, 2010)

There is no consensus on how one can assess whether the observed hypoactivity episode is “depressive episode”. Thus, it can simply be thought that the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5) is the de-facto standard of depressive episodes. Of course, there are numerous differences in behavior between humans and mice. Furthermore, scientists cannot communicate with mice by language and thus subjective symptoms such as depressive mood, guilty feeling, and suicidal ideation, can never be assessed in mice. However, other items of DSM-5 criteria of major depression refer to more objective signs such as sleep and appetite. Thus, the scientists assessed 6 out of 9 items of criteria A and criteria B of Major Depressive Episode.

Among criteria A, diminished interest or pleasure was assessed by comparing wheel running activity which is hedonic activity, and home cage activity. The mice showed normal home cage activity but showed reduced wheel running activity, suggesting that the mice had depression. During the episodes, the mice ate more food and showed weight gain. Electroencephalography (EEG) recording suggested that the mice show increased sleep during the dark period (active phase) and decreased total sleep. By measuring the speed of wheel running, the mice in the episodes showed slow movement, suggesting psychomotor retardation. A treadmill test showed that the mice showed fatigability during the episode.  (Kasahara et al, 2016)

Thus, 5 of the 9 items of A criteria of major depression were met. On the other hand, the other item, “diminished concentration” was assessed by five-choice serial reaction time test. The mice showed the same performance during the episodes. Thus, the mice in the episode did not have diminished concentration. This also indicates that the mice in the episode did not have consciousness disturbance (Kasahara et al, 2016).  Criterion B refers to the impaired social and occupational life. For this item, we performed pup retrieval assay. Whereas the majority of wild-type mice and the mPolg1 Tg mice in the euthymic state retrieved all three pups in the home cage, the mPolg1 Tg mice in the episode retrieved significantly a smaller number of pups. This finding suggests that the mPolg1 Tg mice in the episode showed impaired social function. Collectively, we concluded that the episode observed in mPolg1 Tg mice satisfies the DSM-5 criteria of Major Depressive Episode and is equivalent to the depressive episode in humans.

The scientists also found that a selective serotonin reuptake inhibitor, escitalopram, inhibited depressive episodes, and lithium withdrawal-induced depressive episodes. The scientists previously showed that administration of amitriptyline, a tricyclic antidepressant, induced manic like behavioral change. The scientists also examined the levels of corticosterone and found that excretion of corticosterone increased during the episode. By measuring the body temperature, the mPolg1 Tg mice was found to show the diminished intraday fluctuation of body temperature with higher average body temperature, which is like the findings in human depression.

Collectively, the scientists concluded that the mPolg1 Tg mice satisfy three validity criteria of animal model; construct validity, that are similarities of the mechanisms between the model and the disease, face validity, commonalities between the behavioral features of the model and those of the human disorder, and predictive validity, the efficacy of human drugs for the physical characteristics of the model animal. Though the mice did not show any spontaneous manic or hypomanic episodes, tricyclic antidepressant-induced manic like behavioral changes and atypical features such as body weight gain/increase of appetite and hypersomnia suggest that the mice might have “bipolar spectrum”. Though the scientists initially explained as if “mitochondrial dysfunction hypothesis” is specific to bipolar disorder, it is not at all true. Indeed, mitochondrial dysfunction is a risk factor not only for bipolar disorder but also for other medical diseases such as diabetes mellitus and Parkinson’s disease.

People tend to eat too much in the modern food-rich environment. Although there are several hormones to increase blood glucose level, only one, insulin, can decrease it. Thus, pancreatic islet is Achilles’ heel of humans in modern life. If a person has a latent mitochondrial dysfunction, the first symptom might become apparent at the vulnerable point, such as pancreatic islet. In the case of Parkinson’s disease, production of dopamine causes oxidative stress, and thus dopamine neurons by its nature should have the vulnerability. Associated with elongation of the human lifespan, mitochondrial dysfunction of such vulnerable point might cause disease in susceptible individuals.

The scientists identified that emotion-related neural circuit such as medial prefrontal cortex, nucleus accumbens, and amygdala, accumulated relatively high amount of ΔmtDNA. However, the highest level of ΔmtDNA was detected in the paraventricular thalamic nucleus (PVT). PVT receives input from hypothalamic CRH (corticotropin-releasing hormone) neurons, suprachiasmatic nucleus, and dorsal raphe serotonergic neurons, and send output to anterior cingulate cortex, nucleus accumbens, and amygdala. Although PVT itself has not been a strong candidate region of mood disorders, it has a connection to many candidate regions of mood disorders. The role of PVT in the regulation of emotion is recently drawing attention, and its role in the retrieval of fear memory and opiate withdrawal have been reported by optogenetic manipulation of specific circuits involving PVT.

To test whether there is a relationship between PVT dysfunction and depressive episodes, the scientists functionally knocked down PVT by expressing tetanus toxin that impairs neural transmission in PVT. The scientists found that the mice showed the fewer activity episodes than mPolg Tg mice. This suggests that accumulation of mtDNA in PVT has a pathophysiological role in depression-like behavior in mPolg Tg mice.

However, it is still not known whether PVT dysfunction plays a role in human disease. Thus, the scientists performed immunostaining of the thalamus sections of patients with CPEO and mood symptoms. The scientists found that COX (cytochrome oxidase) negative cells, lacking in COX, a mtDNA-derived protein, are abundant in paraventricular thalamus in those patients.

Collectively, these findings suggest that dysfunction of PVT associated with mitochondrial dysfunction is causative for mood episodes at least in patients with mitochondrial disease. It should be verified whether some patients with bipolar disorder or recurrent depression without mitochondrial diseases also have dysfunction of PVT. If PVT dysfunction, if any, can be detected by brain imaging, it would be useful for diagnosis. However, human PVT may be too small to be visible by current neuroimaging technologies. (Kasahara et al., 2016)

Because mitochondrial dysfunction is not specific to bipolar disorder, we should next focus on which neural circuit is damaged by mitochondrial dysfunction. Thus, the study of mitochondrial dysfunction of bipolar disorder should proceed to the next step. The genetics-based animal model generated in this study would be useful to develop new mood stabilizers. Previously, there has been no animal model to show spontaneous recurrent mood episodes, and thus there was no method to screen candidate mood stabilizer to show the prophylactic effect for mood episodes. The mPolg Tg mouse is the first animal model showing spontaneous recurrent depressive episodes, which responded to existing anti-depression treatments. Thus, the model mice would be helpful to develop new mood stabilizers or antidepressants. The scientists have already searched for candidate drug targets by gene expression analysis of the model mice and postmortem brains of patients with bipolar disorder. Further studies to identify promising lead compounds and search for companion diagnostic methods using the model mice will be needed to improve the clinical practice of bipolar disorder based on the neurobiology of this disorder.

 

 

 

Bibliography:

• Kato, Tadafumi. “Searching for the Molecular Basis of Bipolar Disorder.” American Journal of Psychiatry, vol. 172, no. 11, 2015, pp. 1057–1058.

Bipolar disorder is one of two major psychotic disorders together with schizophrenia and causes a severe psychosocial disturbance. Lack of adequate animal models hampers the development of new mood stabilizers. We proposed a mitochondrial dysfunction hypothesis and have been studying the neurobiology of bipolar disorder based on this hypothesis. We showed that deletions of mitochondrial DNA (ΔmtDNA) play a pathophysiological role at least in some patients with bipolar disorder possibly by affecting intracellular calcium regulation. Mutant polymerase γ transgenic mice that accumulate ΔmtDNA in the brain showed recurrent spontaneous depression-like episodes which were prevented by a serotonin-selective reuptake inhibitor and worsened by lithium withdrawal. The animal model would be useful to develop new mood stabilizers.

 

• Kasahara, and Kato. “What Can Mitochondrial DNA Analysis Tell Us About Mood Disorders?” Biological Psychiatry, vol. 83, no. 9, 2018, pp. 731–738.

Two lines of robust evidence supporting the importance of mtDNA mutations in brain tissues for mood disorders have come from clinical observation of mitochondrial disease patients who carry primary mtDNA mutations or accumulate secondary mtDNA mutations due to nuclear mutations and an animal model study. More than half of mitochondrial disease patients have comorbid mood disorders, and mice with a neuron-specific accumulation of mtDNA mutations show spontaneous depression-like episodes. In this review, they first summarize the current knowledge of mtDNA and its genetics and discuss what mtDNA analysis tells us about neuropsychiatric disorders based on an example of Parkinson’s disease. They also discuss challenges and future directions beyond mtDNA analysis toward an understanding of the pathophysiology of mood disorders.

 

• Parker, G., and H. Brotchie. “Do the Old Psychostimulant Drugs Have a Role in Managing Treatment‐Resistant Depression?” Acta Psychiatrica Scandinavica, vol. 121, no. 4, 2010, pp. 308–314.

Testing of tricyclic levels in some patients suggested that stimulant drugs may raise tricyclic levels in those who are rapid metabolizers. Although this study was not controlled, the high success rate in a diagnostically refined sample implies that the psychostimulants may be efficacious for patients with melancholic and bipolar depression who have failed to respond to orthodox antidepressant drugs.

 

• Diflorio, Arianna, and Ian Jones. “Is Sex Important? Gender Differences in Bipolar Disorder.” International Review of Psychiatry, 2010, Vol.22(5), P.437-452, vol. 22, no. 5, 2010, pp. 437–452.

Most studies, but not all, report an almost equal gender ratio in the prevalence of bipolar disorder but most of the studies do report an increased risk in women of bipolar II/hypomania, rapid cycling and mixed episodes. Important gender distinctions are also found in patterns of co-morbidity. No consistent gender differences have been found in a number of variables including rates of depressive episodes, age, and polarity of onset, symptoms, severity of the illness, response to treatment and suicidal behavior. Unsurprisingly, however, perhaps the major distinction between men and women with bipolar disorder is the impact that reproductive life events, particularly childbirth, have on women with this diagnosis.

 

• Kato, T., et al. “Animal Models of Recurrent or Bipolar Depression.” Neuroscience, vol. 321, no. C, 2016, pp. 189–196.

Animal models of mental disorders should ideally have construct, face, and predictive validity, but current animal models do not always satisfy these validity criteria. Additionally, animal models of depression rely mainly on stress-induced behavioral changes. These stress-induced models have limited validity, because stress is not a risk factor specific to depression, and the models do not recapitulate the recurrent and spontaneous nature of depressive episodes. Although animal models exhibiting recurrent depressive episodes or bipolar depression have not yet been established, several researchers are trying to generate such animals by modeling clinical risk factors as well as by manipulating a specific neural circuit using emerging techniques.