RNA recognition: A family matter?

The study of proteins that bind to RNA-many of them mediators of gene expression or its regulation-has led to the identification of several protein families, each characterized by a common sequence motif that is either known or supposed to interact directly with RNA.

Abstract

lain W. Mattaj Gene Expression Programme European Molecular Biology Laboratory D-6900 Heidelberg Federal Republic of Germany The study of proteins that bind to RNA-many of them mediators of gene expression or its regulation-has led to the identification of several protein families. Each family is characterized by a common sequence motif that is either known or supposed to interact directly with RNA. Do the members of each family recognize similar RNA structures in a similar way? Current evidence suggests not, but high- lights the need for information on the detailed structure of more RNA-protein complexes. The RNA Intramolecular RNA interactions, usually base pairing, lead to characteristic higher order structures, some of whose elements are shown in Figure 1A. Helical stems are formed mainly by Watson-Crick pairs and can be inter- rupted by variable internal loopsand bulges. Bulged nucle- otides can be either stacked into the stem, leading to a kink in the helix, or looped out. Individual stems end at terminal (or hairpin) loops or at junctions between stems that vary greatly in structure. More complexity arises from tertiary interactions between unpaired bases and distant stem-loop structures or from pseudoknots that form when one strand of the stem, having reached a junction, folds back to pair with bases in a loop (Figure 1C). Nuclear magnetic resonance and X-ray crystallographic studies have revealed that loops can also be highly structured. Bases in loops often form non-Watson-Crick base pairs and/or stacking interactions with one another or with paired nucleotides at the end of a stem. Loop and bulge structures allow many opportunities for