Institute of Molecular Biology, University of Oregon, Eugene, OR 97403, USA.
Nucleic Acids Res. 2018 Mar 16;46(5):2613-2623. doi: 10.1093/nar/gkx1288.
Pentatricopeptide repeat (PPR) proteins are helical-repeat proteins that offer a promising scaffold for the engineering of proteins to bind specified RNAs. PPR tracts bind RNA in a modular 1-repeat, 1-nucleotide fashion. An amino acid code specifying the bound nucleotide has been elucidated. However, this code does not fully explain the sequence specificity of native PPR proteins. Furthermore, it does not address nuances such as the contribution toward binding affinity of various repeat-nucleotide pairs or the impact of mismatches between a repeat and aligning nucleotide. We used an in vitro bind-n-seq approach to describe the population of sequences bound by four artificial PPR proteins built from consensus scaffolds. The specificity of these proteins can be accounted for by canonical code-based nucleotide recognition. The results show, however, that interactions near the 3'-end of binding sites make less contribution to binding affinity than do those near the 5'-end, that proteins with 11 and 14 repeats exhibit similar affinity for their intended targets but 14-repeats are more permissive for mismatches, and that purine-binding repeats are less tolerant of transversion mismatches than are pyrimidine-binding motifs. These findings have implications for mechanisms that establish PPR-RNA interactions and for optimizing PPR design to minimize off-target interactions.
五肽重复（PPR）蛋白是螺旋重复蛋白，为工程设计靶向特定 RNA 的蛋白提供了一个有前途的支架。PPR 结构域以模块化的 1 个重复、1 个核苷酸的方式结合 RNA。已经阐明了指定结合核苷酸的氨基酸密码。然而，该密码并不能完全解释天然 PPR 蛋白的序列特异性。此外，它也没有解决各种重复核苷酸对之间结合亲和力的贡献或重复与对齐核苷酸之间不匹配的影响等细微差别。我们使用体外结合-测序方法描述了由共识支架构建的四个人工 PPR 蛋白结合的序列群体。这些蛋白的特异性可以通过基于规范密码的核苷酸识别来解释。然而，结果表明，与 5' 端相比，结合位点 3' 端的相互作用对结合亲和力的贡献较小，具有 11 个和 14 个重复的蛋白对其预期靶标表现出相似的亲和力，但 14 个重复对不匹配的容忍度更高，嘌呤结合重复对转换不匹配的容忍度低于嘧啶结合基序。这些发现对建立 PPR-RNA 相互作用的机制以及优化 PPR 设计以最小化非靶向相互作用具有重要意义。
