Development of a Cell-Based Assay for the Detection of pre-miRNA-Protein Interactions

Abstract

MicroRNA (miRNA), a class of small, non-coding RNA, are the product of a series of precise processing steps and responsible for regulating the translation of >60% of human protein-coding transcripts. Consequently, the dysregulation of miRNA levels has been linked to many human diseases, including cancers and neurodegenerative and cardiovascular diseases. In turn, RNA-binding proteins (RBPs) have been studied as regulators of miRNA biogenesis. Advancements in large-scale technologies have enabled the identification of proteins that bind specific sequences of precursor (pre-) miRNA and the discovery of disease-relevant miRNAs has inspired efforts to identify small molecule inhibitors of such interactions. To aid the study and identification of inhibitors of RPIs, in vitro and in cellular RPI detection systems have been developed. However, requirements of biochemical and cellular methods limit their utility, particularly for use with small, highly processed RNAs. RiPCA, or RNA interaction with Protein-mediated Complementation Assay, an assay for the direct detection of RPIs in live cells, was developed to enable the validation and manipulation of pre-miRNA-protein interactions. In RiPCA, cells stably expressing the small subunit (SmBiT) of a split nanoluciferase (NanoLuc) fused to HaloTag are transiently co-transfected with a functionalized pre-miRNA probe and a plasmid encoding the RBP-of-interest fused to the large subunit (LgBiT) of NanoLuc. The pre-miRNA probe becomes covalently conjugated to SmBiT via HaloTag and subsequent interaction between the pre-miRNA and RBP drives the reconstitution of functional NanoLuc. Initially optimized using the RPI between the let-7 family of miRNA and the Lin28 RBPs, RiPCA was shown to detect the let-7/Lin28 interaction in both the cytoplasm and the nucleus. Furthermore, RiPCA was capable of discerning sequence-specific binding preference of Lin28 as well as indicating the relative binding affinity of Lin28 and its individual RNA-binding domains. These results encouraged the expansion of RiPCA for the detection of other functional pre-miRNA-protein interactions involving the RBPs hnRNP A1, Msi1, and Msi2. RiPCA was similarly capable of detecting the relative binding affinities of these RBPs for several pre-miRNA sequences, including let-7 family members and pre-miR-18a. In addition, the ability of RiPCA to detect site-specific binding was probed using a small library of pre-miRNA probes. While data reflected site-specific binding for Lin28, it was not shown for hnRNP A1, Msi1, and Msi2. Nevertheless, RiPCA demonstrated broad applicability of detecting pre-miRNA-RBP interactions. Finally, to enable high throughput screening (HTS) of inhibitors of pre-let-7d/Lin28, RiPCA was miniaturized, and a partially automated workflow was optimized. A screen of ~18,000 small molecules derived from a curated library resulted in the identification of seven potential let-7/Lin28 inhibitors. Further characterization of the top hits is required to fully elucidate their mechanism of action and activity against let-7/Lin28 in cells. Future efforts should focus on further engineering RiPCA to enable more precise detection as well as detection of RPIs involving other classes of RNAs, including mRNAs, lncRNAs, and expanded repeats. Overall, the technology reported herein promises to advance the characterization of RPIs and provide a platform for the discovery of RPI inhibitors.