Featured Article - January 2009
Short description: Homodimerization of type II poly(A)-binding proteins prevents binding in the absence of RNA.Proc Natl Acad Sci. USA 105, 15317-15322 (2008)
The RNA-recognition motif (RRM) is one of the must abundant protein domains in multicellular organisms. It recognizes RNA and DNA, and also interacts with proteins. The overall structure of the domain is well characterized, but slight structural variations between domains allow RRMs to alter their specificity and carry out diverse biological roles.
Poly(A)-binding proteins (PABPs) contain this motif, and these proteins are present in all eukaryotic cells. They bind to 3′ poly(A) tails of pre-mRNAs through their RRM domain. There are two types of PABPs: type I has four RRM domains (but only uses two for efficient binding) and stimulates translation through interactions with transcription factors; type II contains a single RRM domain.
The best understood type II PABP is BtnPABP2 (Bos taurus nuclear PABP2), which binds to short poly(A) tails, tethers poly(A) polymerase and stimulates the polymerase's enzymatic activity. A similar protein, XlePABP2 (Xenopus laevis embryonic PAPB2), found in developing frog embryos was classified as a type II PABP2 on the basis of its sequence homology with BtnPABP2. But despite several structures of RRM domains, the PABP2 structure was not known.
Song et al., from the PSI CESG and the University of Wisconsin School of Medicine and Public Health, have now determined the high-resolution NMR structure of a fully functional protease-resistant RRM domain of XlePABP2 (XlePABP2-TRP) in isolation and bound to poly(A) RNA. Their NMR data indicate that in the absence of RNA, the RRM domain is a dimer, and gel-filtration studies support this observation.
The authors found that each subunit of XlePABP2-TRP folds into an α/β sandwich structure that is typical of RRM domains. Two helices (α2 and α3) pack against one face of the β-sheet, and on the other, a polyproline motif from the N terminus of XlePABP2 arches over the β-plane, blocking the RNA-binding site in the absence of RNA.
The NMR structure of XlePABP2-TRP in the presence of poly(A) was monomeric, with the polyproline motif displaced, unfolded and exposed to solvent. Polyproline is replaced by the adenosine in poly(A) in the RNA-bound structure. NMR data also showed that the C-terminal region of XlePABP2-TRP is unstructured in the absence of poly(A) and becomes ordered upon RNA contact.
Song et al. confirmed this transition from dimer to monomer using chemical crosslinking in the presence and absence of poly(A). Without RNA, the protein was predominantly dimeric; upon addition of poly(A), a monomer was observed.
From this information, they generated a model of the XlePABP2-TRP:Poly(A) complex. According to this, RNA adopts an extended conformation in the β-sheet surface, and protein–RNA recognition is mediated by extensive stacking interactions and van der Waals contacts.
Many regions of the PABP2 sequence are conserved across species, and BtnBABP2 and XlePABP2 share 70% sequence identity. It is likely that all PABP2 proteins have the same homodimeric structure in the absence of RNA and the same mechanism for RNA binding.
More than 600 different human genes encode proteins with RRM domains, and about one-third of these encode single-domain proteins. The majority of these have not been structurally analyzed. The homodimer–monomer transition upon RNA binding seen in XleBAP2 might help resolve the mechanisms of other domains.
Song Jikui, McGivern Jered V., Nichols Karl W., Markley John L. and Sheets Michael D. Structural basis for RNA recognition by a type II poly(A)-binding protein.
Proc Natl Acad Sci. USA 105, 15317-15322 (2008).