識別情緒疾患之雙極性和季節模式的臨床和遺傳特徵研究

Abstract

情緒疾患是複雜且高度異質性的慢性精神疾病，迄今，我們對於情緒疾患高度異質性特徵了解仍有限，釐清不同表現型亞型、探討其臨床特徵、遺傳因子和可能的致病機制對疾病能夠精準治療十分重要。躁鬱症中有部分患者從未經歷過鬱症發作，稱為單極性躁症，其在全球光照充足的地區報告有較高的比例，可能是受季節變化影響。然而，單極性躁症佔台灣躁鬱症的比例尚不清楚，對於精神與其他身體疾病的共病也缺乏系統性的了解。此外，部分情緒疾患的患者的復發和季節變化具有高度相關，而季節適應性受到許多生理機制和生物節律時鐘基因所調控。晝夜節律障礙或時鐘基因突變和患者伴隨季節性特徵有關，並增加復發或睡眠障礙的風險。 綜合上述，本論文針對單極性躁症和季節情感性特徵進行研究，研究目標為：(1) 估計台灣躁鬱症患者中單極性躁症的比例；(2) 檢視單極性躁症的臨床與心理社會特徵，和終身共病狀況；(3) 檢視季節評估問卷結構和情緒疾患患者四季重複測量的信度，以及情緒疾患中具有季節情感特徵的盛行率；(4) 識別情緒疾患季節情感性特徵的潛在遺傳基因座；(5) 探索季節性情感特徵患者其候選時鐘基因表現量的四季變化。 本論文使用臨床回溯型研究和全國人口之世代研究分別估算單極性躁症在躁鬱症中的比例，並比較單極性躁症和雙極性躁鬱症之特徵差異。在季節情感性特徵方面，使用中文版季節評估問卷結構，並進行四季追蹤研究，以估計患者四季重複評估季節情感性特徵的信度，以及季節情感特徵患者的盛行率。研究也進行全基因體關聯性分析，使用連續分數以及類別型季節情感性特徵定義，探討與季節情感性特徵相關的遺傳因子。此外，利用四季追蹤資料，探索季節情感性特徵患者其候選時鐘基因表現量以及睡眠參數的四季變化。 研究結果發現單極性躁症在躁鬱症中的比例約為12.91%至14.87%，臨床特性有較多精神症狀、較少自殺行為，晝夜節律多屬於清晨型、有較好的睡眠品質，終身共病模式也不相同。季節情感性特徵方面，季節評估問卷屬於雙因子結構，受試者於四季重複評估季節情感性特徵具有良好的信度，大約四分之一的情緒疾患患者有季節情感性特徵。全基因體關聯性研究發現和季節連續分數最相關的位點為rs146370530位於USH2A基因上，和類別季節情感性特徵最相關的的位點為rs10810641位於BNC2基因上，其他與季節情感性特徵也有相關的基因包含LDB2、NPAS3、SHC4和SGCZ。時鐘基因的四季表現量變化方面，除了RORA以外，ARNTL、CLOCK、CSNK1D和NPAS2基因均有四季變化，其中CLOCK和CSNK1D四季變化在驗證樣本也有四季變化。我們進一步發現季節情感性特徵患者NPAS2的春季基因表現量顯著高於健康對照組，並在驗證樣本中有相同的結果。 透過不同的研究設計和資料分析，此論文探討兩種情緒疾患的亞型，並提供其臨床特性和實徵遺傳資料。然單極性躁症的潛在致病機轉尚不清楚，仍缺乏遺傳研究，未來應可以進一步進行研究。論文的結果有助於進一步了解情緒疾患的異質性，對於臨床預防復發和探討致病機制提供指引。

Mood disorder (MD) is a complex and highly heterogenous chronic mental disorder. To date, our understanding of the highly heterogeneous characteristics of MD is still limited. It is crucial to clarify different phenotypes and subtypes and to explore their clinical characteristics, genetic factors, and possible pathogenic mechanisms for precision treatment. A certain proportion of bipolar disorder (BD) patients have never experienced depressive episodes, namely unipolar manic (UM). UM has a higher proportion in areas with sufficient light worldwide, which may be affected by seasonal changes. However, the proportion of UM among BD in Taiwan is unclear, and there is a lack of systematic understanding of the comorbidity of psychiatric and physical diseases. In addition, a high correlation between recurrent and seasonal changes was found in partial MD patients. Seasonal adaptability is regulated by many physiological mechanisms and biological rhythm genes. Circadian rhythm disruption or circadian clock gene mutations are associated with patients with seasonal affective features (SA) and an increased risk of relapse or sleep disturbance. This dissertation focuses on UM and SA, and has several aims: (1) To estimate the proportion of UM among BD patients in the Taiwanese population; (2) To examine relevant clinical features, psychosocial characteristics, and lifetime comorbidity patterns between the UM and depressive-mania (D-M) groups; (3) To examine the factor structure of the Chinese seasonal pattern assessment questionnaire (SPAQ) and the reliability of SPAQ measurement across four seasons, and further to investigate the proportion of SA among MD patients; (4) To identify susceptible genetic loci for SA in MD patients; (5) To explore the gene expression profile of candidate circadian clock genes in response to seasonal changes in MD patients with SA. In the dissertation, we utilized two datasets, the clinical retrospective dataset and the whole population-based cohort dataset, to examine the proportion of UM and differences in clinical features between UM and the D-M group. Regarding SA, we first examined the structure of Chinese SPAQ and estimated the proportion of UM. We then conducted a four-season follow-up study to evaluate the reliability of SPAQ measurement across four seasons. We performed Genome-wide association analyses (GWAS) using continuous scores and categorical SA definitions to explore genetic factors associated with seasonal affective traits. In addition, we used the four-season follow-up data to analyze seasonal changes in the expression of candidate clock genes and sleep parameters in patients with SA. We found the proportion of UM in BD ranged from 12.91% to 14.87%. Clinical features of UM patients were more psychotic symptoms, fewer suicidal behaviors, more morningness chronotype preference, better sleep quality, and different lifetime comorbidity patterns compared with the D-M group. In the aspect of SA, the SPAQ was a two-factor structure. There was good reliability in SPAQ measurements across four seasons. There were around one-quarter of MD patients with SA. The findings of GWAS indicated the most significant loci of continuous scores was rs146370530 in the USH2A gene. The most significant loci of categorical SA definitions was rs10810641 in the BNC2 gene. Other genes associated with SA included LDB2, NPAS3, SHC4, and SGCZ. In terms of seasonal gene expressions, ARNTL, CLOCK, CSNK1D, and NPAS2 were presented with seasonal variation, while RORA was not. Results of the CLOCK and CSNK1D gene expressions were replicated in the validation samples. We further found that NPAS2 showed higher expression in SA patients in spring than that in HC Similar results were found in the validation samples. Through different study designs and data analyses, this dissertation focused on two subtypes of MD. Our results contributed to a better understanding of the clinical characteristics and genetic variants for subtypes of MD. So far, the underlying pathogenic mechanism of UM is still unclear, and more empirical genetic research is warranted for seasonal patterns in MD. Future research in this regard can further our knowledge on this. The results of clinical and genetic studies in the dissertation would help to tackle the heterogeneity of MD, and provide guidelines for recurrent prevention and exploration of pathogenic mechanisms.