- 學位論文
探討情緒障礙之憂鬱特徵與腸道微生物相的相關性
Exploration of the Relationship Between Depressive Features and Gut Microbiota in Affective Disorders
本文將於2025/09/02開放下載。若您希望在開放下載時收到通知，可將文章加入收藏。
摘要
情感性疾患主要包含重度憂鬱症以及雙極性疾患。兩種疾患在所有疾病負擔中佔有很大的比例。高疾病負擔部分來自情感性疾患患者症狀無法依賴現有藥物達到有效的緩解，部分來自於疾病反覆發作的特性，以致病患在生活上嚴重的失能。至今，我們仍對疾病的制病機制了解有限，為了降低疾病負擔，找出疾病致病機制對疾病的在治療以及介入預防相當重要。 許多的證據顯示微生物相-腸-腦軸影響著憂鬱情緒的調控，不管是微生物相多樣性的差異、各個分類學層級發現的微生物相標的或是微生物相的功能。過去幾篇文獻在比較重度憂鬱症患者與健康對照組時，找到數個與憂鬱相關的微生物相標的，然而這些研究並沒有很完善的控制飲食或者其他干擾因子，導致找到的微生物相標的可能受到干擾因子影響。此外雙極性疾患也會表現出憂鬱的症狀，過去研究表示重度憂鬱症與雙極性疾患不管在基因抑或是其他生物性的指標有著相似性，因此有些研究開始了解雙極性疾患內憂鬱與微生物相的相關性，目前已分別在重度憂鬱症以及雙極性疾患內發現數個與憂鬱相關的微生物相標的，然而仍缺乏比較重度憂鬱症與雙極性疾患共享的微生物相標的以及他們的微生物相功能。過去的文獻多利用病例對照研究找到與疾病相關的微生物相標的，疾病相關的微生物相標的可能參與疾病的制病機轉，然而作為生物標記而言，仍有侷限性，因此我們額外想像，究竟微生物相標的是否可能作為憂鬱嚴重度的標記，並且隨著疾病嚴重度不同改變，同時反映其他與憂鬱相關的特徵，例如：焦慮或壓力感知的程度。 結合上述，本研究有五個研究目標：(1) 在考慮到飲食以及其他干擾因子的影響，釐清重度憂鬱症與雙極性疾患各自與憂鬱相關的微生物相標的；(2) 比較重度憂鬱症與雙極性疾患內與憂鬱相關的微生物相標的是否共享；(3) 透過兩種研究設計探討微生物相是否與疾病嚴重度，以及疾病急性緩解狀態有顯著相關；(4) 探討前述找到的憂鬱標的是否額外與憂鬱相關的焦慮情緒、壓力感知程度的表現相關；(5) 利用現有微生物相功能分析軟體，剖析在重度憂鬱症以及雙極性疾患的微生物相功能。 本論文分為三個部分，第一部分利用病例對照設計探討重度憂鬱症內與憂鬱相關的微生物相標的；討論與憂鬱、焦慮以及壓力感知程度的相關性；以及剖析重度憂鬱症相關的微生物相功能。第二部分與第一部分呼應，同樣使用病例對照設計，在雙極性疾患患者內探討上述描述的標的、相關性以及功能，並且額外探討重度憂鬱症以及雙極性疾患是否有共享的微生物相標的以及功能。第三部分則採用追蹤研究設計，收集重度憂鬱症以及雙極性疾患患患者憂鬱急性發作以及緩解兩個時間點的資料，深入探討與憂鬱嚴重程度相關的微生物相標的。 本研究考量微生物相的干擾因子，排除兩個月內有服用益生菌或抗生素、進行腸胃道手術以及腸胃道感染的受試者。第一部分共收集36位重度憂鬱症患者以及37位健康對照組；第二部分共收集33位雙極性疾患、47位重度憂鬱症以及53位健康受試者，並且包含第一部分的27位重度憂鬱症患者，以及26位健康受試者；第三部分則額外收集11位憂鬱急性發作的情感性疾患患者。第一、二部分，每位受試者填寫貝氏憂鬱量表收集過去兩周的憂鬱情形，以及填寫壓力感知問卷以及貝氏焦慮量表。此外也訪問飲食頻率問卷並且收集糞便檢體。糞便檢體經過DNA萃取後進行16S核醣體定序 (16S ribosomal RNA gene sequencing)。在微生物相的分析，我們利用QIIME軟體 (Quantitative Insights Into Microbial Ecology)，將定序後的序列進行微生物分類。分析策略中，我們利用ANCOM (Analysis of composition of microbiomes)，在校正飲食資訊以及定序資訊後找到與重度憂鬱症相關之標的，再利用斯皮爾曼等級相關係數探討微生物相標的與焦慮、感知壓力以及憂鬱嚴重度的相關性，最後使用PICRUST軟體 (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States) 分析與憂鬱相關的功能。 第三部分，我們於受試者急性發作與緩解兩個時間點，使用漢氏憂鬱量表 (HAMD-17) 評估兩個時間點憂鬱症狀的嚴重程度，並且定序微生物相組成，分析憂鬱嚴重度相關的微生物相標的。此外我們將憂鬱症狀拆解成六個因子，利用斯皮爾曼等級相關係數釐清微生物相與憂鬱因子群的相關性。 比較重度憂鬱症患者與健康對照組後，發現12個與憂鬱相關的屬，而在比較雙極性疾患患者與健康對照組後，則發現11個與憂鬱相關的屬，其中巨單胞菌屬 (Megamonas)、普雷沃氏菌屬 (Prevotella)、布勞特氏菌屬 (Blautia) 以及薩特氏菌屬 (Sutterella) 皆在兩部分的比較中發現。額外將雙極性疾患患者合併重度憂鬱症患者與健康對照組比較，發現除了上述四個屬外，還有腸球菌屬 (Enterococcus)、小桿菌屬 (Dialister)、顫螺菌屬 (Oscillospira) 以及脫硫弧菌科 (Desulfovibrionaceae) 內一個無法分類的屬。與憂鬱、焦慮以及壓力感知程度進行相關性分析，在第一、二部分皆可發現薩特氏菌屬與憂鬱和壓力感知程度呈現顯著的負相關。第三部分，比較病患於急性與緩解的微生物相組成顯示有兩個屬別有顯著不同，分別為柔嫩梭菌屬 (Faecalibacterium) 以及鏈球菌屬 (Streptococcus)，與憂鬱的六個因子進行相關性分析，柔嫩梭菌與¬失眠有高度的正相關。從PICRUST結果來看，我們也觀察到躁鬱症患者、重度憂鬱症患者與健康對照組的細菌組成在代謝相關的功能有顯著不同。 本研究顯示腸道微生物相標的不僅與情感性疾患相關，也與憂鬱嚴重程度相關，兩個部分找到的菌相標的不同，顯示致病與嚴重度牽涉的制病機轉可能不同。由於微生物相與憂鬱特徵的異質性皆高，因此未來需收集更大的樣本，並且結合更精密的霰彈槍定序法 (Shotgun sequencing) 、代謝體學抑或是動物實驗模型，找出比屬更精確的分類，如：種或者株，研究細菌的功能在憂鬱特徵的致病或者是嚴重度機轉中扮演的角色。
並列摘要
Major depressive disorder (MDD) and bipolar disorder (BPD) are two major categories in affective disorder and obtain substantial disease burden. The high disease burden originated from the current situation that patients are not efficiently responding to the treatment regimen and the characteristics of repeated recurrence during the disease course. At present, we have limited knowledge regarding the etiology of affective disorder. In order to lower the disease burden, it is thus crucial to discover the disease mechanisms to prevent and treat the disease. Growing evidence suggests the regulation of microbiota-gut-brain axis in depressive mood. Previous studies revealed several microbiota targets related to depression by comparing dozens of MDD patients and healthy controls. However, these studies lack of reasonable control of the confounding factors. The targets revealed previously may be confounded. BPD also exhibits the depressive feature, where accumulating evidence showed the similarity between MDD and BPD in genetic correlation or other biology functions. Currently, several studies demonstrated the microbiota targets in depressive BPD patients. However, the comparisons between the microbiota targets in MDD and BPD are still limited. Additionally, with growing studies discovered the disease-related targets, another question emerges if the depressive severity and other mood-related traits, for instance, anxiety and perceived stress level, also associated with microbiota. There are five aims in the current research: (1) Explore microbiota targets while consider the confounding factors in patients with affective disorders; (2) Discover if MDD and BPD share common microbiota targets; (3) Discuss if the microbiota targets correlate with depressive severity via two study designs; (4) Explore if the microbiota targets correlate with mood-related traits; (5) Apply putative functional annotative software to analyze the microbiota functions in MDD and BPD. The current research has been separated into three parts. In part I, we apply a case-control design in 36 MDD and 37 healthy controls. After the DNA extraction, 16S ribosomal RNA gene sequencing was applied to obtain microbiota composition in the subjects. QIIME served as a vital tool to conduct the taxonomy classification. ANCOM was applied to discover the microbiota targets while adjusting for diet and sequencing information. The correlations between microbiota targets and symptom severities was conducted via Spearman's rank correlation. Finally, PICRUST was used to inference the microbial functions. In part II, a case-control design was also applied in 33 BPD patients, 47 MDD patients and 53 healthy controls and we also conducted the analysis mentioned above. We further compared if MDD and BPD share common targets via comparing microbiota composition between patients with affective disorder and healthy controls. In part III, a follow-up designed was applied. We collected the information from 11 patients with affective disorders during the acute phase and remission phase. The depressive severity was assessed via HAMD-17. ANCOM and Spearman's rank correlation was applied to discover depressive severity-related targets and the correlations with depressive symptoms. In the current study, 12 genera were discovered comparing MDD and controls, while 11 genera were explored comparing BPD and controls. Four genera, including Megamonas, Prevotella, Blautia, and Sutterella consistently showed up. After comparing microbiota composition between affective disorders and control, the 4 targets remain significant while the other 4 genera (Enterococcus, Dialister, Oscillospira, and an unclassified genus in Desulfovibrionaceae) emerge. For the correlation analysis, genus Sutterella consistently revealed negative correlations with depressive and perceived stress levels. While in part III, two genera (Faecalibacterium and Streptococcus) exhibited differential abundance in the acute and remission phase. Genus Faecalibacterium further exhibited a strong positive correlation with HAMD factor 1- insomnia. For the functional pathway analysis in PICRUST, the results revealed that pathways related to metabolism or biosynthesis process were different between depressive BPD, MDD and control groups. The current research suggested the involvement of microbiota in the depression mechanism and the depressive severity mechanism. Both depression and microbiota composition contain very high inter-individual variations; therefore, the future study requires a larger sample size to discover robust microbiota targets related to depression. Additionally, if we combine whole-genome shotgun sequencing, the metabolomics, and animal models might further shed light on the mechanism of microbiota targets related to depressive diagnosis and depressive severity.
參考文獻
7 References
1 Aagaard, K., Ma, J., Antony, K.M., Ganu, R., Petrosino, J., Versalovic, J., 2014. The placenta harbors a unique microbiome. Sci Transl Med 6(237), 237ra265-237ra265.
2 Alcock, J., Maley, C.C., Aktipis, C., 2014. Is eating behavior manipulated by the gastrointestinal microbiota? Evolutionary pressures and potential mechanisms. Bioessays 36(10), 940-949.
3 American Psychiatric Association, 2013. Diagnostic and statistical manual of mental disorders (DSM-5®). American Psychiatric Pub.
4 Amir, A., McDonald, D., Navas-Molina, J.A., Kopylova, E., Morton, J.T., Zech Xu, Z., et al., 2017. Deblur Rapidly Resolves Single-Nucleotide Community Sequence Patterns. mSystems 2(2).