| 摘要: | T細胞受體組庫（TCR repertoire）之定義為生物體內所有T細胞受體（T-cell receptor, TCR）的集合，由T細胞群中所有T細胞受體的種類與數量所決定。在T細胞發育的過程中，每顆T細胞會經歷T細胞受體基因的V(D)J重組（V(D)J recombination），進而產生不同的T細胞受體基因序列，使得每顆T細胞具有不同的T細胞受體克隆型（clonotype）並產生T細胞的高度多樣性。據估計，這種基因重組機制能產生高達1015種的T細胞受體克隆型，使得T細胞群具有和各式各樣的非自體抗原進行高度專一性結合的能力，T細胞免疫因而得以辨認並消滅外來的病原體和體內的非正常細胞－例如：被病毒感染的正常細胞或癌細胞。因此T細胞組庫的組成和狀態與免疫系統的活化和功能有著高度的相關。
過去由於技術／方法上的侷限，大幅度限制了T細胞受體組庫在分析時能取得的通量和資訊上，因而造成分析大量的T細胞受體成為一件困難的工作。有鑒於這十年間次世代定序（next-generation sequencing, NGS）技術的發展以及在T細胞受體組庫分析上的應用，今日已能透過T細胞受體定序（TCR sequencing, TCR-seq）方法，使得完整並同時解析數百萬個不同T細胞受體的V(D)J序列成為可能。雖然目前市面上已有針對T細胞受體定序的文庫製備平台／方法，但其大部分都有價格昂貴或表現不佳的缺點。因此本論文之第一部分藉由參考和測試先前文獻在T細胞受體文庫製備（library preparation）所需的寡核苷酸（oligonucleotides）和試劑組，讓T細胞受體定序更為經濟。從本研究的結果中可發現，基於5’端快速擴增互補DNA末端（rapid amplification of cDNA ends, RACE）技術的修改顯著地增加了T細胞受體文庫製備所需的cDNA模板產量，也大幅度地減少文庫製備所需的花費。這些初步成果顯示測試中的T細胞受體定序平台為T細胞受體定序分析提供了更低的成本以及具可比性的建庫表現。
此外，本論文之目標包含建立探討T細胞受體組庫與人類疾病關聯性的分析流程，因此本論文於第二部分透過T細胞受體定序解析類風濕性關節炎（rheumatoid arthritis, RA）患者於不同生物製劑療程－包含阿達木單抗（adalimumab：腫瘤壞死因子-α（tumor necrosis factor alpha, TNF-α）抑制劑）、利妥昔單抗（rituximab：CD20抑制劑）和塔西單抗（tocilizumab：白血球介素-6受體（interleukin-6 receptor, IL-6R）抑制劑）治療後的T細胞受體組庫特徵。我們的結果顯示類風濕性關節炎患者達到疾病緩解後，其T細胞受體多樣性、V/J基因使用程度和CDR3長度之分佈在不同生物製劑療程的治療下沒有差異；然而，我們進一步發現這些類風濕性關節患者在經由生物製劑治療後達到疾病緩解或低疾病活性的情況下，其T細胞受體組庫多樣性的高低和疾病活性的嚴重度呈現負相關。本研究的結果顯示T細胞受體多樣性與疾病活性之間在生物製劑治療後到達疾病穩定期的類風濕性關節患者上具有關聯性。
接著，本論文透過T細胞受體定序平台研究感染性疾病患者的T細胞受體組庫特性，因此本論文於第三部分探討新冠肺炎疾病2019（coronavirus disease 2019, COVID-19）患者於急性期為輕度疾病（mild disease）或肺炎（pneumonia）時其恢復期的T細胞受體組庫特徵。我們的結果顯示肺炎患者相較於輕度疾病患者具有較低的T細胞受體多樣性以及不同的CDR3長度分佈。進一步透過T細胞受體集群（clustering）和註解（annotation）之整合分析，我們發現與嚴重急性呼吸道症候群冠狀病毒2型（severe acute respiratory syndrome coronavirus 2, SARS-CoV-2）相關且為輕度疾病特異的T細胞受體集群具有較高的T細胞受體生成機率、較高的共享性以及較高的CDR3抗原結合區域位點多樣性。本研究結果顯示T細胞受體組庫與SARS-CoV-2相關T細胞受體集群之特性在輕度疾病和肺炎COVID-19患者之間具有差異。
最後，本論文呈現如何透過T細胞受體定序平台探討原發性高血壓（essential hypertension, EH）和醛固酮分泌腺瘤（aldosterone-producing adenoma, APA）誘發之次發性高血壓患者在T細胞受體組庫之間的差異。我們初步的結果顯示醛固酮分泌腺瘤患者相較於原發性高血壓患者具有較低的T細胞受體多樣性；此外我們進一步觀察到T細胞受體生成機率與頻率之間的關聯性強度在醛固酮分泌腺瘤患者有較高的現象。這些發現顯示高血壓患者的T細胞受體組庫狀態會受到醛固酮分泌腺瘤影響。
總結來說，本論文之重要性為：以高通量定序技術鑑別T細胞受體組庫與人類疾病之間的關聯性。
TCR repertoire is defined as the sum of all T cell receptors (TCRs) by the T cells in an organism. It is determined by the total types and numbers of TCRs of the T cell population. During the T cell development, each T cells experience the V(D)J recombination on TCR gene locus and subsequently generate different TCR gene sequences as distinct TCR clonotypes, leading to a high T-cell diversity. It is estimated that 1015 possible TCR clonotypes can be generated via such DNA recombination mechanism, which enables T cell population to bind to vast variety of non-self-antigens with high specificities. The T cell immunity can recognize and eliminate external pathogens and endogenous abnormal cells, such as virus-infected cells or tumor cells, accordingly. The composition and status of TCR repertoire is thus highly connected with the activation and function of immune system.
In the past, it was difficult to analyze a large number of TCRs due to the limitation of technologies/approaches which largely restricted the throughput and information of TCR repertoire analysis. Given that next-generation sequencing (NGS) technology has been developed and employed to analyze TCR repertoire in recent ten years, comprehensively profiling of V(D)J sequences of millions of different TCRs in parallel is practicable by TCR sequencing (TCR-seq) methods nowadays. However, most marketed platforms/approaches for TCR-seq library preparation are expensive or have unsatisfying performances. Herein, the first part of the dissertation was to test oligonucleotides and reagents used for TCR library preparation according to previous references in order to achieve economic TCR-seq analysis. The data demonstrated that the modification of 5’ RACE-based platform not only increased the template yield but also reduced the expense for TCR library preparation. Preliminary results of tested TCR-seq platform provided a reasonable cost and a comparable performance for TCR-seq analysis.
In addition, the dissertation aimed to establish analytic pipelines to study the connection between TCR repertoire and human diseases. The second part of the dissertation was to profile TCR repertoire in rheumatoid arthritis (RA) patients with different treatment courses using biologic disease-modifying antirheumatic drugs (bDMARDs), including adalimumab (anti-TNF), rituximab (anti-CD20) and tocilizumab (anti-IL6R). The results showed that RA patients in clinical remission after different treatments of bDMARDs had no difference in TCR diversity, V/J gene usage and CDR3 length distribution. Nevertheless, the diversity of TCR repertoire was found to be negatively correlated with disease activities of bDMARD-treated RA patients in remission or low disease activity. The study suggested a correlation between TCR diversity and disease activity in RA patients after biologic treatment even at the stable stage.
Furthermore, our TCR-seq platform was applied to investigate characteristics of TCR repertoire in individuals with infectious diseases. The third part of the dissertation was to characterize TCR repertoire of convalescent coronavirus disease 2019 (COVID-19) patients diagnosed with either mild disease or pneumonia at acute phase. Our results revealed that compared to the patients with mild disease the pneumonia patients had slightly decreased TCR diversity and skewed CDR3 length distribution. An integrated analysis based on TCR clustering and annotation further showed that SARS-CoV-2-associated mild disease-specific TCR clusters had higher cluster generation probabilities (Pgen), higher sharing level, and higher positional diversity within CDR3 antigen binding region. The study indicated distinct characteristics of TCR repertoire and SARS-CoV-2-associated TCR clusters between mild disease and pneumonia COVID-19 patients.
Finally, the prospective section of the dissertation presented the works regarding the use of our TCR-seq platform to study the difference in TCR repertoire of hypertension patients between essential hypertension (EH) and aldosterone-producing adenoma (APA). Our preliminary data identified that the TCR diversity was lower in APA patients compared to EH individuals. A higher strength of correlation between TCR Pgen and frequencies was observed in APA relative to EH patients. These findings suggested that the TCR repertoire status of hypertension patients could be potentially affected by APA.
In overall, the significance of the dissertation is to apply high-throughput sequencing approach to identify the connection between TCR repertoire and human diseases. |
