U1A SPLICEOSOMAL PROTEIN BINDS TO TWO DIFFERENT RNA TARGETS USING THE SAME STRATEGY. Nagai, K., Oubridge, O., Price, S., Jovine, L., Avis, J. & Evans, P. R., MRC Laboratory of Molecular Biology, Hills Road, Cambridge, UK

U1A spliceosomal protein binds to hairpin II of U1 small nuclear RNA and, together with other proteins, forms the U1 snRNP essential in pre-mRNA splicing. It also binds to the 3' untranslated region (3' UTR) of its pre-mRNA and controls its own production through inhibition of polyadenylation at the 3' end of mRNA. The crystal structure of U1A protein bound to a 21 nucleotide RNA hairpin, representing its binding site in U1 snRNP, was solved at 1.92Å resolution using a single isomorphous derivative. The AUUGCAC sequence within the ten nucleotide loop fits tightly into the groove on the surface of the protein and the bases of the heptanucleotide are splayed out. The bases of the heptanucleotide and the C:G base pair, which closes the loop, stack onto an aromatic protein side chain, an adjacent base or both. These bases also form an extensive hydrogen bond network with protein side chains, main chain amide and carbonyl groups. Bases of the last three nucleotide in the loop have no apparent interaction with the protein.

The 3' UTR of U1A protein mRNA folds into an evolutionarily conserved secondary structure with two internal loops containing the sequence AUUGCAC, found in hairpin II of U snRNA, and its variant AUUGUAC. The C:G base pair is also found at the equivalent positions in the 3'UTR binding site. The conservation of the heptanucleotide and the C:G base pair strongly suggests that they form identical protein contacts in the 3'UTR complex. Based on this assumption and the 2-fold symmetry in the 3'UTR binding site we have been able to model of the 3'UTR complex in which two molecules of U1A protein form a dimer.

U2B" protein, a component of U2 snRNP, shares a high sequence homology with U1A protein and binds to hairpin IV of U2 snRNA only when it is complexed with U2A' protein which contains the Leu-rich sequence repeats. We have been able to grow crystals of a ternary complex between U2B"/U2A' proteins and the cognate binding site and its crystallographic analysis is underway. This structure will reveal how U1A and U2B" proteins bind their cognate binding sites.

(1) Oubridge, C. et al. Nature 372, 432-438 (1994)

(2) Jovine, L. et al. Structure in press