Our laboratory has published the following review on protein-RNA complexes:
Serganov, S. & Patel, D. J. (2008). Towards deciphering the principles underlying a mRNA recognition code. Curr. Opin. Struct. Biol. 18, 120-129. [PubMed Abstract]
Autoimmune Disease Syndromes
Diverse aspects of RNA metabolism are dictated by the La autoantigen, an abundant RNA-binding phosphoprotein found in the nucleus of all eukaryotes, and originally identified as an autoantigen in patients with systemic lupus erythematosus and Sjörgen's syndrome. La specifically targets and protects the UUUOH 3'-terminii of nascent RNA polymerase III transcripts, including pre-tRNAs, 5S rRNAs and snRNAs from exonuclease digestion, while discriminating against 3'-phosphate-containing internal oligo U tracts and degraded RNA. La plays a role in 5'- and 3'-end processing of pre-tRNA precursors and exhibits RNA chaperone-like activity, thereby playing a key role in facilitating correct transcript folding, downstream processing and maturation, and ribonucleoprotein particle assembly. In addition La binds viral RNAs by site-specifically targeting their internal ribosome entry sites and stimulating translational inhibition.
We have recently solved the crystal structure of the N-terminal domain (NTD) of human La, consisting of La and RRM1 motifs, bound to a 9-mer ending in UUUOH and also the NMR solution structure of the La NTD complexed to a 3-mer UUUOH complex. The UUUOH 3'-end, in a splayed apart orientation, is sequestered in a basic and aromatic amino acid-lined cleft between the La and RRM1 motifs. The specificity-determining central U base bridges both motifs, in part through unprecedented targeting of the β-sheet edge, rather than the anticipated β-sheet face, of the RRM1 motif. Both hydroxyls of the sugar ring of the last U are hydrogen-bonded, with neither phosphate nor bulky modifications tolerated at this site. Our structural and mutation results establish how the La NTD protects the UUUOH 3'-ends of nascent RNA transcripts during downstream processing and maturation events. Current efforts in the laboratory are focused on structural characterization of complexes of the La C-terminal domain (CTD) with its RNA targets.
Teplova, M., Yuan, Y. R., Phan, A. T., Malinina, L., Ilin, S., Teplov, A. & Patel, D. J. (2006). Structural basis for recognition and sequestration of UUUOH 3'-terminii of nascent mRNA polymerase III transcripts by La autoantigen. Mol. Cell 21, 75-85. [PubMed Abstract]
The generation of functionally diverse proteins required for cell growth and differentiation in metazoan organisms is critically dependent on alternate splicing of pre-mRNAs. Alternative splicing regulators control the expression of tissue-specific or developmental stage-specific protein isoforms through binding either directly to splice sites or to other sequences in pre-mRNA, thereby enhancing or repressing inclusion of alternative exons. The proteins of the muscleblind-like family, MBNL, have been identified as important tissue-specific alternative splicing regulators that play a key role in terminal muscle differentiation. Normal splicing pattern is altered specifically in the neuromuscular disease myotonic dystrophy (DM), in part, due to inactivation of MBNL.
Recognition of pre-mRNA GC Elements by Tandem Zinc Finger Domains of Muscleblind-Like Protein MBNL1
MBNL proteins harbor tandem CCCH Zn finger (ZnF) domains that target pre-mRNAs containing YGCU(U/G)Y sequence elements. In myotonic dystrophy, reduced levels of MBNL proteins leads to aberrant alternative splicing of a subset of pre-mRNAs. Our crystal structure of MBNL1 ZnF3/4 bound to r(CGCUGU) establishes that both ZnF3 and ZnF4 specifically target GC steps. The guanine and cytosine bases of the GC step insert into adjoining pockets, where they stack on conserved arginine and aromatic residues and form a network of hydrogen bonds primarily with main-chain groups of the protein, while the 2'-OH groups are hydrogen bonded to conserved side-chains. The relative alignment of ZnF3 and ZnF4 domains is dictated by the topology of the interdomain linker, with the resulting anti-parallel orientation of bound GC elements, supportive of a chain-reversal loop trajectory for MBNL1-bound pre-mRNA targets, thereby impacting on alternative splicing regulation.
Teplova, M. & Patel, D. J. (2008). Structural insights into RNA recognition by the alternate splicing regulator muscleblind-like MBNL1. Nat. Struct. Mol. Biol. 15, 1343-1351. [PubMed Abstract]
Neurodegenerative Disease Syndromes
Nova proteins, expressed in central nervous system neurons, are target antigens of the autoimmune disorder POMA syndrome, a neurodegenerative disease that originates when systemic malignant tumors express proteins normally sequestered in the central nervous system. The immune system targets these antigens to be non-self, and the ensuing response results in neurodegeneration. Our structural studies of the POMA syndrome are directed toward providing a structural understanding of how full-length Nova, which contains three K-homology (KH) domains, targets and regulates alternate splicing events within the 2 glycine receptor subunit pre-mRNA.
The fragile X syndrome, the most common form of inherited mental retardation in humans, results from the expansion and hypermethylation of trinucleotide CGG repeats located within the 5'-untranslated region of the FMR1 gene. Fragile X mental retardation protein (FMRP) contains RNA-binding KH and RGG domains and is known to regulate mRNA localization and/or translation. The onset of the fragile X mental retardation (FXMR) syndrome is associated with loss of FMRP function, which is essential for higher cognitive function. Recently, mRNAs encoding proteins that contain FMRP-binding elements have been identified, and it appears that their dysregulation may underlie human mental retardation. We are interested in the characterization of RNA complexes formed with RNA-binding elements within the FMRP protein, in our attempts to understand factors that contribute to the FXMR syndrome.
The Nova and fragile X projects involve a collaboretive effort with the Robert Darnell laboratory at The Rockefeller University.
Neurodegenerative POMA Syndrome
We have determined the crystal structure of the first two KH domains (KH1/KH2) of Nova-1 bound to a RNA hairpin, whose loop contains a pair of tandem UCAN high-affinity binding sites. The structure establishes that KH1 is involved in RNA recognition while KH2 unexpectedly participates in protein dimerization. Though both UCAN repeats have been shown to be involved in KH-RNA recognition by mutation studies in our collaborator Robert Darnell's laboratory, sequence-specific contacts are primarily associated between KH1 and the second UCAN repeat, with the RNA hairpin scaffold buttressed by interactions between repeats. Strikingly, the same KH face is involved in adenine-mediated KH1-RNA recognition and tryptophan-mediated KH2-KH2 dimerization. These results establish the functional diversity of Nova KH domains implicated in the pathogenesis of POMA syndrome and highlight how protein-protein interactions can impact on RNA target recognition. We are currently mutating residues at the unanticipated KH2 dimeric interface to evaluate its functional role, as well as attempting to provide a molecular explanation for protein engineering efforts aimed at extending the KH-RNA interface.
Malinina, L., Teplova, M., Musunuru, K., Teplov, A., Darnell, J. C., Burley, S. K., Darnell, R. B. & Patel, D. J. (2009). Structural insights into RNA recognition versus dimerization by KH domains of the neuronal splicing factor Nova-1.
Our long-term goal is to structurally characterize a complex of Nova KH1/KH2/KH3 bound to an UCAU-U-UCAU-C-UCAU-containing RNA sequence element. The challenge will be to identify which KH domains target which UCAU elements, and how partitioning occurs between RNA recognition and KH dimerization.
Neurodegenerative FXMR Syndrome
Our initial structural efforts on the FXMR syndrome have focused on the complex between an RGG peptide and a quadruplex-duplex neuronal RNA scaffold identified in our collaborator Robert Darnell's laboratory for which we have obtained exceptional NMR spectra of samples site-specifically and uniformly 13C,15N-labeled in the peptide and the RNA components. To date, the peptide and RNA resonances have been identified, the intermolecular NOEs assigned, and structure calculations are currently under way. These structural efforts are being coupled with mutation experiments to identify energetic contributions involving key residues associated with molecular recognition. A molecular perspective of the recognition code between FXMR and its neuronal mRNA quadruplex-containing targets could lead to structure-based design approaches to mitigate the severity of the disease.
A patient severely affected with the FXMR syndrome has been shown to contain a missense I304N mutation in the KH2 domain of FMRP, suggestive of this KH2 domain playing a key role in translational regulation. The Robert Darnell laboratory has identified a sequence-specific element, consisting of a kissing RNA hairpin, as the RNA target for the KH2 domain. Further, this group demonstrated that association of FMRP with brain polyribosomes is abrogated by competition with the FMRP kissing complex RNA. We have initiated an effort aimed at the structure determination of a complex between the FXMR KH2 domain and the kissing-hairpin RNA target. The combined collaborative structure-function studies with the Darnell laboratory could highlight the potential role for RNAs harboring the kissing complex motif as targets for FMRP translational repression.