RNA-binding proteins: it's better to play in a band

In this Outlook, Jeong et al. discuss a study in this issue of Genes & Development by Peyda et al. that provides insight into the RNA binding and activity of Rbfox1 and the LASR complex, describing how RNA-binding proteins (RBPs) cooperate to regulate alternative splicing. They further highlight the improved methodology used in the study and its potential to further our understanding of the complex, combinatorial mechanisms of RBP-mediated RNA processing.

Abstract

Alternative splicing (AS) is regulated by a myriad of RNA-binding proteins (RBPs) in a coordinated manner. However, most studies characterize RBPs individually. In this issue of Genes & Development, Peyda and colleagues (doi:10.1101/gad.352105.124) revealed how the LASR complex, consisting of multiple RBPs, regulates AS by recognizing multipart sequences. Their approach may be applicable to studying the combinatorial effects of other RBPs, which is critical for cracking the splicing code.

The life of an RNA is complicated. From transcription to processing, export, and eventually its degradation, each step involves precise control. Among the most critical regulators are RNA-binding proteins (RBPs), which play essential roles in orchestrating these processes. According to the latest count, the human genome encodes >1000 RBPs (Gebauer et al. 2021). These proteins bind to RNAs with a wide range of specificity and affinity, often regulating multiple aspects of RNA metabolism. Interestingly, RBPs tend to interact with other RBPs and form complexes. For example, all snRNPs contain multiple RBPs (Wahl et al. 2009). U2AF, one of the earliest splicing factors identified, contains U2AF65 and U2AF35, both well-known RBPs. Similarly, one of the earliest studies analyzing alternative splicing (AS) demonstrated that female-specific splicing of the Drosophila gene dsx requires the coordinated binding of the splicing regulators Tra and Tra2 and an SR protein on a highly conserved 13 nt sequence motif (Tian and Maniatis 1993). Many RBPs contain intrinsically disordered regions (IDRs), which can mediate weak and multivalent interactions with one another. These interactions can be further strengthened by the presence of RNA. The associations among RBPs and RNAs are so prevalent that they give rise to various large membraneless organelles, such as nuclear speckles, nucleoli, and stress granules. As such, it is critical to understand how the activities of RBPs and RBP complexes are coordinated in regulating RNA processing steps.

Although we know that RBPs frequently act in groups, our field typically investigates them one at a time. This tendency is largely driven by technical limitations. It is more straightforward to carry out loss- or gain-of-function assays for individual factors. Additionally, the gold standard method for mapping protein–RNA interactions (i.e., HITS-CLIP/CLIP-seq [UV cross-linking, immunoprecipitation and sequencing; referred to here as CLIP-seq] and its variants) characterizes one RBP at a time. As RBPs are isolated under denaturing conditions, CLIP-seq captures the ensemble of all RNA interactions mediated by the target RBP in all contexts. With current technologies, a typical investigation of an RBP involves (1) performing RNA sequencing in control and loss-of-function or, less commonly, gain-of-function conditions to determine the RBP's impact on RNA processing; (2) mapping global RBP–RNA interactions; and (3) integrating the two data sets and generating an “RNA map” that correlates RNA binding with positive or negative regulatory activities of the target RBP. Although this has been a powerful research paradigm that has yielded tremendous insights into RBP functions, it is important to look beyond individual RBPs and start to study them in their native contexts that may likely involve other RBPs. The study by Peyda et al. (2025) is an important step in this direction.

The RNA-binding Fox-1 homolog (Rbfox) proteins represent an important family of RBPs in mammals, comprising three paralogs: Rbfox1, Rbfox1, and Rbfox3 (Conboy 2017). The expression of Rbfox proteins is regulated in a tissue-specific and developmental stage-specific manner, and mutations in and aberrant expression of these genes are associated with various diseases, including cancer (Jbara et al. 2023). The functions of Rbfox proteins in the regulation of AS have been studied extensively in the past decade. However, the majority of studies have focused on the Rbfox proteins directly. Interestingly, Black and colleagues (Damianov et al. 2016) discovered several years ago that most Rbfox proteins are part of a large and stable protein complex known as the large assembly of splicing regulators (LASR), which also contains other RBPs such as hnRNP M, hnRNP H/F, and MATR3. Rbfox associates with the LASR complex via its tyrosine-rich IDR, and importantly, this association is essential for splicing regulation (Ying et al. 2017). To study how Rbfox/LASR recognizes RNAs, the investigators performed IP-seq, a technique recently developed in their laboratory (Damianov et al. 2024). In this method, the chromatin fraction—where the majority of Rbfox resides—was initially isolated, and the proteins/complexes within this fraction were liberated by nuclease treatment. Rbfox proteins were then immunoprecipitated, and the RNA fragments that they protect were isolated and sequenced. IP-seq is essentially an RNA footprinting approach, similar to ribosome profiling and NET-seq/mNET-seq for translation and transcription mapping. The investigators have successfully applied this method to studying RBPs/complexes that bind to RNAs cotranscriptionally, such as U2 snRNP (Damianov et al. 2024). Rbfox1 IP-seq yielded both overlapping and distinct signals compared with those obtained from CLIP-seq. Although CLIP-seq and its variants can provide up to nucleotide-resolution protein–RNA interaction information, not all RBPs can be efficiently cross-linked to RNA by UV irradiation. Importantly, as associated RNAs are isolated from denatured proteins, one cannot discern whether the recovered RNA is bound by the RBP individually or as part of a complex. In contrast to CLIP-seq, IP-seq isolates RNA–protein complexes under native conditions (provided that these complexes and the associated RNAs can survive nuclease treatment). This orthogonal method for mapping protein–RNA interactions is highly complementary to traditional methods, and we anticipate it will be widely applicable to the studies of other RBPs/complexes in the future.

Interestingly the RNA-binding sites identified by Rbfox1 IP-seq not only include GCAUG, the well-characterized Rbfox1-binding motif, but also feature additional elements that are known targets of the other LASR components. These sequence elements are often located close to one another in clusters, suggesting that the LASR complex binds to RNA through multivalent interactions involving multiple RBPs and their cognate sequences. Multiple lines of evidence suggest that other LASR components can influence Rbfox-mediated splicing regulation. First, deleting the cognate sequences of other RBPs in LASR, such as G-rich sequences recognized by hnRNP H/F, affects the splicing of Rbfox-regulated exons. Second, expressing a mutant Rbfox1 protein with 1500-fold reduced RNA binding affinity led to its recruitment to many of the sites bound by the wild-type protein, resulting in splicing changes similar to those observed with the wild-type protein. These results suggest that Rbfox–RNA interactions and its splicing regulatory activities are strongly influenced by other RBPs within the LASR complex. As mentioned previously, RBPs often associate with one another, and many form multisubunit complexes. Based on the results of Peyda et al. (2025) highlighting the activity of Rbfox within the LASR complex, multisubunit RBPs may enable the regulation of AS in specific exons by multiple RBPs. Furthermore, the observation that the Rbfox with defective RNA-binding activity retained much of the splicing regulatory function of the wild-type protein suggests that interactions involving multiple RBPs and RNA sequence elements can enhance the robustness of AS.

Deciphering the splicing code necessitates understanding the combinatorial activities of various RBPs within specific contexts. To understand how mutations in splicing factors contribute to cancer, for example, it is crucial to go beyond simple loss-of-function assays and examine the effects of the mutant proteins within the context of their native complexes. As seen with SNRNP200, many RBPs may bind to different RNA targets and exert different functions depending on whether they are in the nucleus or cytoplasm (Street et al. 2024). The conceptual and technological advances described by Peyda et al. (2025) are opening new avenues to investigate the complex orchestration of RBP-mediated RNA metabolism. Embracing this intricate regulatory behavior will be essential for advancing future RBP studies.