A Dial, Not a Switch: How Phosphorylation Tunes the Grip of a Splicing Factor on RNA

Every human gene is transcribed into a precursor RNA that must be edited before translation, and among the proteins that orchestrate this editing, known as splicing, SRSF1 occupies a central position, helping the spliceosome decide which segments to keep and which to discard. Like many regulatory proteins, SRSF1 contains an intrinsically disordered region, the arginine/serine-rich (RS) domain, which lacks a fixed three-dimensional structure and is heavily tagged with phosphate groups by dedicated kinases, although how these modifications quantitatively reshape its interactions with RNA has long remained unclear.

To address this question, a team from the University of Zurich and ETH Zurich, in a study published in Nature Communications on October 1, compared three versions of the RS domain: phosphorylated by a kinase, chemically synthesized with phosphates at defined positions, and mutated by replacing serines with acidic residues that mimic the charge of a phosphate (phosphomimetic variants). Binding to RNA was measured by single-molecule FRET (smFRET), a fluorescence technique that, like a molecular ruler, detects the nanoscale distance between two labels allowing individual molecules to be watched as they bind or separate. The results show that the RS domain is a powerful driver of RNA binding, and that progressive multi-site phosphorylation weakens it gradually rather than abruptly, until extensive phosphorylation abolishes detectable binding: in other words, it behaves less like an on/off switch than like an electrostatic rheostat.

Importantly, binding energy varied almost linearly with the domain’s net charge, whether that charge came from genuine phosphates or from acidic substitutions, although two acidic residues were needed to match a single phosphorylation, consistent with the greater negative charge of a phosphate group. This has a direct methodological implication: phosphomimetic mutants, widely used as experimental shortcuts, may systematically underestimate the impact of real phosphorylation.

Unfortunately, the study relies on a deliberately reductionist in vitro system, built on isolated protein segments and defined RNA, which cannot capture the crowded nuclear environment or the biomolecular condensates in which SRSF1 operates. Even so, it offers a quantitative framework of clear therapeutic relevance: SRSF1 is overexpressed in several cancers, and the kinases that phosphorylate SR proteins, such as SRPK1 and the CLK family, are already being explored as drug targets, so knowing how each phosphate tunes RNA affinity could help calibrate the effects of such inhibitors. Whether this charge-based logic extends to other disordered RNA-binding proteins remains an exciting question for future research.