Arginine Forks Are a Widespread Motif to Recognize Phosphate Backbones and Guanine Nucleobases in the RNA Major Groove

Abstract

RNA recognition by proteins is central to biology. Here we demonstrate the existence of a recurrent structural motif, the "arginine fork", that codifies arginine readout of cognate backbone and guanine nucleobase interactions in a variety of protein− RNA complexes derived from viruses, metabolic enzymes, and ribosomes. Nearly 30 years ago, a theoretical arginine fork model was posited to account for the specificity between the HIV-1 Tat protein and TAR RNA. This model predicted that a single arginine should form four complementary contacts with nearby phosphates, yielding a two-pronged backbone readout. Recent highresolution structures of TAR−protein complexes have unveiled new details, including (i) arginine interactions with the phosphate backbone and the major-groove edge of guanine and (ii) simultaneous cation−π contacts between the guanidinium group and flanking nucleobases. These findings prompted us to search for arginine forks within experimental protein−RNA structures retrieved from the Protein Data Bank. The results revealed four distinct classes of arginine forks that we have defined using a rigorous but flexible nomenclature. Examples are presented in the context of ribosomal and nonribosomal interfaces with analysis of arginine dihedral angles and structural (suite) classification of RNA targets. When arginine fork chemical recognition principles were applied to existing structures with unusual arginine−guanine recognition, we found that the arginine fork geometry was more consistent with the experimental data, suggesting the utility of fork classifications to improve structural models. Software to analyze arginine−RNA interactions has been made available to the community.

Arginine multivalency stabilizes protein/RNA condensates

“…MD trajectories indicate that both arginine and lysine side chains form a variety of distinct intermolecular contacts with polyU, which encompass H-bonding with the oxygens of the negatively-charged phosphates, the hydroxyl oxygen of the ribose and the carbonyl oxygens of the base, as well as guanidinium-uracil stacking in the case of arginine. All these dynamical interactions in peptides/nucleotides mixtures closely resemble those observed in X-ray structures of stable protein/RNA complexes 29,30 .…”

Arginine multivalency stabilizes protein/RNA condensates

“…Altogether, the versatility of arginine guanidinium groups may eventually lead to complex binding patterns where a single arginine side chain coordinates several RNA groups, and possibly multiple oligonucleotide chains. These multivalent interactions are reminiscent of arginine forks, which are a widespread structural motif in protein/RNA complexes according to the extensive analysis of protein structural databases 29 . In conclusion, our results suggest that the peculiar structural features of arginine side chain play a key role in the formation of multivalent interactions with model unstructured RNAs, such as polyU oligonucleotides, thus rationalizing the important role of this amino acid in protein/RNA phase separation.…”

“…MD trajectories indicate that both arginine and lysine side chains form a variety of distinct intermolecular contacts with polyU, which encompass H‐bonding with the oxygens of the negatively‐charged phosphates, the hydroxyl oxygen of the ribose and the carbonyl oxygens of the base, as well as guanidinium‐uracil stacking in the case of arginine. All these dynamical interactions in peptides/nucleotides mixtures closely resemble those observed in X‐ray structures of stable protein/RNA complexes 29,30 …”

Affinity and Structural Analysis of the U1A RNA Recognition Motif with Engineered Methionines to Improve Experimental Phasing

“…A take-home message is that organic acid salts do not promote a specific helix C conformation and that the C-terminal region is malleable for RNA recognition, as demonstrated by mainchain displacements as great as 10 Å. Interestingly, the dmU1A-derived TBP6.9 protein does not use its C-terminus to recognize HIV TAR RNA (Figure 1b). In fact, this region of the polypeptide chain was disordered in all TBP-TAR complexes [51,59]. The average all-atom displacement of TBP6.9 superimposed on the dmU1A(F37M/F77M) variant was 1.33 Å for residues 5-91.…”

“…(b) (Left) Lab-evolved TBP6.9 recognizes TAR RNA at the internal bulged loop, which includes a U23 A27-U40 base triple and bulged A35 within helical stem S1b [51,52]. An arginine fork is a key determinant of recognition at the G26 major-groove edge and the U23 backbone [59]. (Right) Cartoon drawing of the TBP6.9-TAR co-crystal structure [52], emphasizing protein recognition of the TAR internal bulged loop (PDB entry 5xh0); the apical loop and S1a (white) are not involved in TAR binding and are hypothesized herein to allow replacement within a duplex target sequence to promote RNA crystallization and phasing.…”