Role of microRNAs in Regulating Gene Expression

干扰素-g 3UTR的片段在脊椎动物中是保守的，这种保守的非编码序列(CNS)被认为具有调节潜力。我们对干扰素-g基因序列的分析表明，在3个非编码区的ARE元件下游有一个很有希望的靶点是miRNAs 29a和b。该区域存在于从许多物种中克隆的干扰素-g基因中，因此暗示该miRNA结合位点对干扰素-g基因的表达具有调节作用。通常，ARE元件和miRNA靶序列都参与基因表达的转录后调控。而在干扰素-g 3的非编码区中，ARE位于5端，miRNA-29BS位于3端，尚无证据支持miRNAs与干扰素-g的ARE区结合的模型。为了验证这一假设，我们通过突变ARE和miRNA-BS产生了几个3个干扰素-g-UTR报告结构，我们的实验揭示了ARE和miRNA之间的竞争效应，影响了mRNA的稳定性。根据这些结果，我们假设Are介导的衰变(AMD)和RNA诱导的沉默复合体(RISC)途径竞争性地相互作用，从而调节干扰素-g的转录后调控。我们假设RISC复合体可能通过与ARE结合蛋白相互作用来抑制AMD过程。我们的体外结果使我们得到了RNA结构专家Bruce Shapiro(NCI)博士的帮助。Shapiros博士的实验室分析了3个非编码区结构，并评估了miRNA的结合是否改变了3个RNA结构。他的实验室提供了计算证据，证明mir-29在1090-1109区域的结合导致了mRNA二级结构的重新排列。与相邻区域的miRNA结合相比，这种结构变化似乎更大，性质也不同(见图6)。特别是，AU富集区和miRNA结合区之间的长距离相互作用似乎受到miRNA结合的调节。与最初预期的相反，调制似乎是在相反的方向上：预测在miRNA结合时存在长距离相互作用，并以未结合的形式减弱或不存在。然而，计算分析的几点需要注意的是：1.不基于比对的RNA二级结构预测方法不能指望完全准确。2.通过计算分析估计了结合的miRNA对mRNA二级结构的影响，通过对mRNA序列进行二级结构预测(RNAold)，约束条件是miRNA结合位点的核苷酸不参与mRNA折叠(保持单链)。3.不对超出RNA碱基配对的结构效应进行建模。为了确认所预测的miRNA结合时确实发生了RNA结构变化，我们与斯图尔特·勒格莱斯博士利用SHAPE技术(通过引物延伸分析的选择性29-羟基酰化反应)展开了合作。这项技术使我们能够更好地了解miRNA结合如何影响干扰素-g RNA结构，并确定预测的RNA结构的理论变化是否真的发生在与miRNAs相互作用之后。如果这个模型得到验证，我们将测试已经被报道与干扰素-g ARE相互作用的蛋白质TTP的结合是否在miRNA存在的情况下发生改变。我们假设miRNA与mRNA的结合会干扰TTP与RNA的结合，从而稳定RNA。作为整个项目的一部分，我们还在寻找可能通过改变干扰素-g基因稳定性来影响基因表达的干扰素-g基因的多态。

Segments of the IFN-g 3UTR are conserved across vertebrates and such conserved non-coding sequence (CNS) are hypothesized to have regulatory potential. Our analysis of the IFN-g gene sequence revealed that there is a promising target for miRNAs 29a and b binding downstream of the ARE element in the 3 UTR. This region is present in IFN-g genes cloned from many species, thus implicating this miRNA binding site as having a role in regulating IFN-g gene expression. Typically, both the ARE element and miRNA target sequences are known to be involved in post-transcriptional regulation of gene expression. While in the IFN-g 3 UTR, the ARE is located in the 5 end and the miRNA-29BS at the 3 end, there is no reported evidence supporting the model that miRNAs bind to the ARE region of IFN-g. To test this hypothesis, we generated several 3 IFN-g-UTR reporter constructs by mutating the ARE and miRNA-BS and our experiments revealed competing effects between the ARE and miRNA, affecting the stability of the mRNA. From these results, we hypothesize that the ARE mediated decay (AMD) and RNA induced silencing complex (RISC) pathways competitively interact and thereby regulate post-transcriptional control of IFN-g. We hypothesize that the RISC complex may inhibit the AMD process, by potentially interacting with ARE binding proteins. Our in vitro results led us to enlist the help of Dr. Bruce Shapiro (NCI), an expert in RNA structure. Dr. Shapiros laboratory analyzed the 3 UTR structure and evaluated whether binding of the miRNA altered 3 RNA structure. His laboratory provided computational evidence that binding of mir-29 in the region 1090-1109 leads to a rearrangement of the secondary structure of the mRNA. This structural change appears to be larger and different in nature compared to binding of miRNA in neighboring regions (see Figure 6). In particular, a long-distance interaction between the AU-rich region and the miRNA binding region seems to be modulated by the miRNA binding. The modulation appears to be in the reverse direction as initially expected: the long-distance interaction is predicted to be present upon binding of the miRNA and weakened or not present in the unbound form of the mRNA. However, several caveats of the computational analysis should be mentioned: 1. RNA secondary structure prediction methods that are not based on alignments cannot be expected to be completely accurate. 2. The influence of the bound miRNA on the mRNA secondary structure was estimated with a computational analysis, by subjecting the mRNA sequence to a secondary structure prediction (RNAfold) with the constraint that the nucleotides of the miRNA binding site did not participate in the mRNA folding (remained single stranded) and 3. Structural effects that go beyond RNA base pairing are not modeled. To confirm that the predicted RNA structural changes upon binding of the miRNA actually occur, we initiated a collaboration with Dr. Stuart LeGrice, (NCI) utilizing SHAPE technology (selective 29-hydroxyl acylation analyzed by primer extension). This technology permits us to gain a better understanding of how miRNA binding may affect IFN-g RNA structure and determine if the predicted theoretical changes in RNA structure actually occur following interaction with the miRNAs. If this model is validated, we will then test whether the binding of the protein TTP, already reported to interact with the IFN-g ARE, is altered in the presence of the miRNA. We hypothesize that miRNA binding to the mRNA will interfere with the binding of TTP to the RNA, thus stabilizing the RNA. As part of this overall project, we are also searching for polymorphisms in the IFN-g gene that may effect gene expression by altering IFN-g mRNA stability.