We have long-standing interest in the regulation of RNA metabolism, particularly with respect to the control of alternative pre-mRNA splicing in development and disease. Our initial research contributed to the establishment of the SR (for serine-arginine rich) family of splicing regulators and demonstrated that SR proteins are responsible for committing pre-mRNA to the splicing pathway. Unlike other RNA binding splicing regulators, SR proteins are not only essential for constitutive, but also involved in alternative splicing, mainly through recognizing cis-acting exonic splicing enhancer (ESE) elements in pre-mRNAs. Along this research direction, we have also expended our research to other families (e.g. RBFox and PTB) of RNA binding splicing regulators. Importantly, in recent years, specific mutations in several SR proteins and related splicing factors have been identified as drivers for various forms of leukemia and solid tumors. By developing a new technology (R-ChIP) to map R-loop, a three-stranded RNA/DNA structure, we found that splicing factor mutations induce R-loop formation and subsequent DNA damage response to account for the cellular phenotype observed in diseased cells. We have keen interest in pursuing this research direction in future studies.
Our lab was also responsible for the discovery of SR protein-specific kinases (SRPKs), and demonstrated that these kinases regulate nuclear import of SR proteins and SR protein-mediated protein-protein interactions during splicing. We found that SRPKs are partitioned between the cytoplasm and the nucleus, serving as critical transducers of external signals to regulate splicing in the nucleus. Interestingly, SRPKs appear to have a symbiotic relationship with another family of SR protein-specific CDC2-like (or CLKs) kinases in the nucleus. SRPKs have also been implicated in human cancers, functioning either as an oncogene or a tumor suppressor in different contexts. We and others have developed specific inhibitors against SRPKs and shown their effects in attenuating tumor development in mouse models, suggesting that these splicing kinases may be further explored as targets for anti-cancer therapy.
Our research has uncovered additional functions of splicing factors and regulators in gene expression at the chromatin levels, demonstrating, for example, that the classic SR protein SC35 (a.k.a. SRSF2) has a direct role in transcriptional pause release by facilitating the recruitment of the Pol II CTD kinase pTEFb, and that RBFox2 interacts with nascent RNAs to modulate the recruitment and function of Polycomb Complex 2 (PRC2). More recently, we have pursued a large-scale analysis of RNA binding proteins on chromatin and developed a new technology to systematically map RNA-chromatin interactions (GRID-seq). Our results suggest that literally all regulatory chromatin regions may involve specific RNAs and RNA binding proteins. These observations point to a new research dimension in understanding the contribution of RNAs, including nascent protein-coding RNAs and various non-coding RNAs, and specific RNA binding proteins to regulated gene expression at both transcriptional and co-transcriptional levels, particularly in the framework of phase separation in mediating enhancer-promoter interactions in 3D genome.
During our basic research of the PTB family of RNA binding proteins in regulated RNA metabolism, we unexpectedly found that depletion of PTB is able to potently convert various non-neuronal cells to neurons. Pursuing this intriguing phenomenon, we revealed that PTB suppresses a key neuronal induction pathway whereas its neuronal paralog nPTB provides a critical checkpoint for neuronal maturation, and as a result, sequential inactivation of PTB and nPTB is necessary and sufficient to convert both mouse and human non-neuronal cells to functional neurons. We have recently applied this basic scientific discovery in the mouse brain, showing that this new cellular reprogramming strategy is able to potently convert midbrain astrocytes into nigra neurons in situ, leading to effective correction of the phenotype associated with Parkinson’s Disease in a chemically induced mouse model. These findings showcase the power of basic research in fostering new approaches against human diseases.