DNA Binding of Human and Viral Transcription Factors is Associated with Rheumatoid Arthritis Risk Loci

本Merit Review申请基于我们的假设，即来自 类风湿性关节炎（RA）患者感染Epstein -巴尔病毒（EBV）后会有独特的基因 表达特性改变宿主对EBV的免疫应答，并且这些差异影响EBV感染的风险。 发展RA。这些假设基于以下观察： 我们评估了6例风湿性关节炎（RA）患者和6例骨关节炎（OA）患者 对照组中含有一些感染EBV的细胞。我们发现FLS可以与EBV重叠感染。我们 观察到RA（而不是OA）遗传风险基因座富含DNA变异，这些变异与免疫沉淀有关。 EB病毒核抗原-2（EBNA 2）（相对风险（RR）=4.5，Bonferroni校正概率 (Pc)=2.4x10-14）。初步数据显示，RA和OA在基因表达方面存在差异， 由RA风险基因座中的元件控制。这些观察和Pubmed上的792篇文章，通常 相互矛盾的是，EBV和RA之间存在许多有趣的关系。我们认为，这些意见是 与EBV参与RA的发病机制一致。 我们计划三个倡议与这个DVA优异奖项目。首先，我们将描述病毒感染， FLS（目的1）：评估EBV FLS感染的频率，对FLS中的EBV DNA进行测序， 比较RA、OA和正常（NL）FLS中的EBV基因表达。我们将检验FLS感染 在EBV感染人群中很常见。其次，我们将探讨FLS对EBV的反应 通过用RNA-seq确定宿主基因表达来评估EBV感染（目的2）， 影响来自RA、OA和NL受试者的FLS，并通过探索RA、OA和NL中RA位点的等位基因特异性 FLS。一旦确定实验条件，单细胞RNA-seq将表征细胞群 感染EBV的细胞和对EBV感染有反应的细胞。这些实验将评估假设 RA FLS将具有与来自OA或NL受试者的FLS不同的基因表达谱， 表达数量性状基因座（eQTL）将RA与OA和NL FLS区分开。第三，我们将评估 对EBV感染的FLS的免疫应答（目的3）。我们的假设是，宿主对EBV感染的反应， FLS将RA与OA和NL受试者区分开。具体来说，我们详细的工作假设和目前的 模型是：1. OA和NL FLS通常是EBV感染的流产，2。OA & NL EBV感染 FLS以最低水平存在，可能仅表达少数潜伏性EBV基因，如在潜伏感染的B中所见 细胞，3。EB病毒感染的OA和NL FLS通常对免疫识别是静止的。相反，在RA中， 有证据表明，EBV裂解程序的早期步骤是由早期- EB病毒直接抗原。这些变化与其他因素（例如，抗环瓜氨酸 肽自身抗体，可能的微生物组变化等）唤醒静止的免疫反应， 导致针对EBV感染的RA FLS的压倒性炎症反应， 不发生在OA或NL滑膜中。正如我们在以前提交的文件中所得到的经验一样， 可能会与这些想法中的任何一个相矛盾，从而导致这项工作进入一个新的概念和战略方向。对 另一方面，这项工作有可能提供目前未知方面的机制细节， RA的发病机制，这将成为新的治疗方法和预防策略的基础。 无论EB病毒是否是RA发病机制的一个组成部分，所提出的研究将提供深入了解， 以前未知的组织特异性位点和常见慢性病毒感染的可能储库， 感染了绝大多数人类

This Merit Review application is based on our hypotheses that Fibroblast-Like Synoviocytes (FLS) from Rheumatoid Arthritis (RA) patients infected with Epstein - Barr virus (EBV) will have unique gene expression properties that alter the host immune response to EBV, and that these difference influences the risk for developing RA. These hypotheses is based on the following observations: All 12 of the FLS cell lines that we have evaluated, six from Rheumatoid Arthritis (RA) patients and six from Osteoarthritis (OA) disease controls, contained some cells infected with EBV. We discovered that FLS can be superinfected with EBV. We observed that RA (but not OA) genetic risk loci are enriched for DNA variants that are immunoprecipitated with Epstein-Barr Nuclear Antigen-2 (EBNA2) (Relative Risk (RR)=4.5, Bonferroni corrected probability (Pc)=2.4x10-14). Preliminary data showed differences between RA and OA in the expression of the genes controlled by elements in RA risk loci. These observations and the 792 publications in Pubmed, often contradictory, present many intriguing relationships between EBV and RA. In our view these observations are consistent with EBV being involved in the pathogenesis of RA. We plan three initiatives with this DVA Merit Award project. First, we will characterize the virus infection in FLS (Aim 1) by: assessing the frequency of EBV FLS infection, sequencing the EBV DNA in FLS, and comparing EBV gene expression in RA, OA, & normal (NL) FLS. We will test the hypothesis that FLS infection is common among the EBV infected human population. Second, we will explore the FLS response to EBV infection (Aim 2) by determining host gene expression with RNA-seq to assess EBV infected and secondarily influenced FLS from RA, OA, & NL subjects and by exploring allele specificity at RA loci in RA, OA, and NL FLS. Once experimental conditions are determined, single cell RNA-seq will characterize the cell population infected with EBV and the cells responding to EBV infection. These experiments will evaluate the hypotheses that RA FLS will have a different gene expression profile than the FLS from OA or NL subjects and that the expression quantitative trait loci (eQTL) will distinguish RA from OA and NL FLS. Third, we will assess the immune response to EBV infected FLS (Aim 3). Our hypothesis is that the host response to EBV infection of the FLS will differentiate RA from OA and NL subjects. In specific, our detailed working hypothesis and current model is: 1. That usually the OA & NL FLS are abortively infected with EBV, 2. That OA & NL EBV infected FLS exist at a minimal level, perhaps only expressing a few EBV latent genes, as seen in latently infected B cells, and 3. That EBV infected OA & NL FLS are usually quiescent to immune recognition. In contrast, in RA there is evidence that the early steps of the EBV lytic program are initiated with the expression of Early- Immediate EBV antigens. These changes in combination with other factors (e.g., anti-cyclic citrullinated peptide autoantibodies, possible microbiome changes, etc.) awaken the quiescent immune response and result in an overwhelming inflammatory response directed against the EBV infected RA FLS in a way that does not happen in the OA or NL synovium. As has been our experience in the previous submissions, the results obtained may contradict any of these ideas leading this work into a new conceptual and strategic direction. On the other hand, this work has the potential to provide mechanistic details of the currently unknown aspects of RA pathogenesis, which would become the basis for new therapeutic approaches and preventive strategies. Whether or not EBV is a component of RA pathogenesis, the studies proposed will provide insight into a previously unknown tissue specific site of and possible reservoir for a common chronic viral infection that infects the vast majority of human beings.