Neuro-immunological investigation of auto-immune thyroid factors in bipolar disease.

Abstract

Introduction: The co-incidence of thyroid auto-immunity and neuro-psychiatric disorders is well-documented. However, the prevailing school of thought implicating auto-immune thyroid disease (AITD) in bipolar disorder is neither well-understood nor universally accepted. Thus, the lack of recent plausible data linking these two disorders provides the rationale for the present study and lends novelty to our results. We investigated the association between Hashimoto’s disease (the most common cause of auto-immune hypo-thyroidism) and bipolar disorder through the extra-thyroidal localisation of thyroid-specific proteins, thyroid-stimulating hormone receptor (TSH-R) and thyroglobulin (TG) in major limbic regions of bipolar human brain. The limbic system represents the cerebral control centres for human emotional behaviour responses. Interestingly, the extra-thyroidal localisation of TSH-R and TG has been well-documented. Thyroid-stimulating hormone receptors have been identified in mammalian heart, kidneys, bone, lymphocytes, thymus, pituitary, adipose, skin, hair follicles, fish testes and astrocyte culture. Thyroglobulin has been identified in mammalian skin, hair follicles, thymus and kidney. With particular reference to central nervous system (CNS) localisation, the most relevant to the present study are reports of immuno-reactive TSH-R in human anterior pituitary tissue and cultured astrocytes, as well as in rat brain tissue and cultured astrocytes. However, no previous studies have specifically detected TSH-R and TG in other human cerebral cortical regions. Recently, in a pilot study to the present project, we immuno-localised TSH-R and TG to cortical neurons and cerebral vasculature, respectively, within various human non-limbic brain regions including occipital lobe, parieto-occipitotemporal, primary motor cortex and primary sensory cortex. The present study extends this investigation into the distribution of thyroid-specific proteins, specifically in limbic regions of normal and bipolar human brain at both protein and molecular levels. We postulated that changes in thyroid protein expression in bipolar limbic regions may contribute to mood dysregulation associated with bipolar disorder, thereby implying a significant role for the thyroid system in the pathophysiology of bipolar disorder. Thus, we chose to compare the distribution of thyroid-specific mRNA and proteins in major limbic regions between normal and bipolar human adult brain and then use this data to infer whether any regulation of limbic thyroid protein expression could be related to the pathophysiology of bipolar disorder. Methods: Ethical approval for the present study was granted by the University of KwaZulu-Natal (UKZN), Biomedical Research Ethics Committee (Reference EXPO42/06). Forensic human brain limbic tissue samples (frontal cortices, amygdalae, cingulate gyrii, thalamii, hippocampi and hypothalamii) were obtained from five individuals whom had succumbed to causes unrelated to head injury and who had demonstrated no evidence of brain disease or psychological abnormality. Bipolar brain limbic tissue samples of frontal cortices, amygdalae and cingulate gyri were obtained from five confirmed cases of bipolar disorder from a commercial brain bank. In addition, normal thyroid gland tissue was collected for use as a control tissue. Three experimental techniques were used to fulfil the objectives of this project: Commercial polyclonal antibodies raised in rabbit against human thyroid-specific proteins, TSH-R and TG, were used to immuno-localise TSH-R and TG proteins in the major limbic regions of normal and bipolar human brain by standard immuno-histochemical techniques. Quantification of TSH-R and TG immuno-staining was determined by image analysis digitalisation followed by inter- and intra-group statistical comparisons. (ii) Using specific oligonucleotide primers, TSH-R mRNA in normal and bipolar limbic regions was detected by in-situ reverse-transcriptase polymerase chain reaction (in-situ RT-PCR). This cellular distribution of TSH-R mRNA was then graded according to visual intensity of labelling as well as the extent of cellular distribution. (iii) Solution-type reverse-transcriptase quantitative polymerase chain reaction (RTqPCR) was then used to determine fold-change in TSH-R gene expression in the bipolar limbic group, relative to normal controls, and normalised to the endogenous reference gene, glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The comparative threshold cycle (CT) method was used to calculate gene expression fold-differences between normal and bipolar groups.

Results: There were a number of novel results that were demonstrated in this study: Immuno-reactive TSH-R and TG was demonstrated within the neurons and vasculature, respectively, of normal and bipolar limbic regions. Specifically, we immuno-localised TSH-R to large motor neurons whilst TG was evident in smooth muscle cells of the tunica media and tunica adventitia of limbic vasculature. Other results included the presence of TG in limbic neurons of cingulate gyrus and frontal cortex in both normal and bipolar limbic regions. Thyroglobulin was also evident in bipolar amygdala neurons but not in the normal neurons. Further, TSH-R and TG proteins were not detected in any limbic neuronal support cells examined, such as astrocytes, neuroglia, oligodendrocytes and satellite cells. Image analysis intra-group statistical comparisons indicated a significant reduction of TSH-R and TG proteins in the bipolar limbic groups when compared to matched normal groups. In addition, we obtained inter-group statistical comparisons for TSH-R and TG proteins in all limbic regions examined. Inter-regional comparisons for TSH-R in normal limbic regions showed significant differences in the hypothalamus and thalamus groups when compared to the other four limbic regions but not when compared to each other. No appreciable difference was noted amongst the remaining four normal limbic groups. Inter-regional comparisons in all bipolar limbic regions examined, indicated no appreciable difference between any of these categories. Inter-regional comparisons for TG in normal limbic vasculature demonstrated that when compared to the amygdala, there was a significant difference in the cingulate gyrus, frontal cortex and thalamus. None of the other inter-regional comparisons showed statistical differences in these normal groups. Inter-regional comparisons for TG in bipolar limbic vasculature demonstrated that when compared to the frontal cortex, significant differences were noted in the cingulate gyrus and amygdala. The cellular localisation of TSH-R mRNA by in-situ RT-PCR demonstrated evident labelling for TSH-R mRNA in neurons within all matched normal and bipolar limbic regions. However, semi-quantitative comparisons indicated a down-regulated expression of TSH-R mRNA in all bipolar limbic regions when compared to normal controls. In other normal limbic regions examined, TSH-R mRNA was found exclusive to hypothalamic neurons, whilst the thalamus and hippocampus weredevoid of labelling. Further, TSH-R mRNA was not detected in cerebral vasculature or any neuronal support cells. RTqPCR determinations of TSH-R gene expression between normal and bipolar groups confirmed the down-regulated expression of TSH-R mRNA in the bipolar limbic regions. Bipolar inter-group statistical comparisons demonstrated TSH-R expression to be significantly reduced in the cingulate gyrus when compared to the amygdala. None of the other comparisons showed appreciable statistical differences. Discussion and Conclusion: The main study findings that demonstrate the extra-thyroidal expression of TSH-R and TG in limbic regions of normal and bipolar human brain are unique, and are suggestive of the translocation of thyroid-like proteins via the seemingly-impervious blood-brain barrier (BBB). Interestingly, multiple modes of egress beyond the BBB have been previously described, whereby the integrity of the BBB may be altered during neuro-pathological diseases including HIV-associated dementia, multiple sclerosis, Alzheimer’s disease and bipolar disorder. Alteration of the protective BBB increases its permeability to leukocytes and other circulating compounds and triggers signal transduction mechanisms that facilitate the loss of tight junction proteins that constitute the highly specialised cerebral endothelium. Thus, those reports indicating an altered BBB during neuro-pathological conditions, lends support to our findings of TSH-R and TG proteins in the bipolar brain. Further to these novel discoveries is the demonstration of reduced thyroid protein expression in major limbic regions of the bipolar brain. In particular, our findings of reduced TSH-R expression in limbic regions correlate with previous neuro-imaging reports that describe reduced physiological cortico-limbic tissue volumes and neuro-physiological activity during bipolar disorder. In addition, the presence of TG-like proteins, exclusive to bipolar amygdala neurons, may be associated with other neuro-imaging findings of amygdala neuro-physiological hyperactivity and enhanced emotional sensitivity in bipolar disorder. The importance of the thyroid hormones, T3 and T4, in mammalian CNS during various life stages has been reported. It is therefore reasonable to speculate that our findings of thyroid-related proteins, TSH-R and TG, in normal limbic neurons may also exhibit an alternate neuro-physiological role, specifically related to mood control. Further, the reduced expression of TSH-R/TG will likely result in reduced localised thyroid function in limbic neurons with consequent altered neuronal functioning and this may predispose or exhibit some involvement in the pathophysiology of mood dysregulation often observed in bipolar disorder. Interestingly, in the present study, we report the reduced expression of TSH-R and TG in limbic neurons of bipolar brain. Alternatively, we may attribute symptoms of mood dysregulation due to limbic-derived TSH-R providing potential targets for thyroid auto-immunity during Hashimoto’s disease. We speculate that inhibitory-type auto-antibodies associated with Hashimoto’s disease, may agonise TSH-R expressed in limbic neurons. This abnormal interaction may lead to the inactivation or cell-mediated atrophy of limbic neurons with consequential reduction in the levels of expressed TSH-R. Thus, the loss of limbic neurons or diminished neuronal activity may contribute towards mood dysregulation. The neuro-pathology of diminished neuronal functioning or neuronal atrophy, is suggestive of a neuro-degenerative aetiology in bipolar disorder. This is particularly controversial, since current evidence implicating neuronal structural and functional changes in the pathophysiology of bipolar disorder is limited and does not provide enough support to classify bipolar disorder as a neuro-degenerative type disorder. However, our study suggests that neuronal alterations may be due to changes in thyroidal status in bipolar disorder. Moreover, our correlation of reduced thyroid protein expression levels in bipolar limbic regions with previous neuro-imaging findings of reduced cortico-limbic volumes and activity during bipolar disorder, provides further supporting evidence indicating neuro-degeneration of mood-controlling limbic structures in bipolar disorder.