These numbers check details underline the potential of solution-state techniques to study flexible assemblies and at the same time suggest that this

potential has not yet been fully exploited for RNP complexes. NMR-based structural studies of RNPs have so far addressed small to medium-size complexes (briefly reviewed in the next session). However, most molecular machines involved in RNA metabolism and in regulatory RNA pathways are multi-component assemblies of more than 50 kDa. Due to their modular architecture, the divide-and-conquer approach is useful to decipher the atomic details of RNA–protein interfaces. On the other hand, since only the full complex retains functionality, the architecture of high-molecular-weight RNPs in solution is relevant to understand structure–function relationships. This perspective article discusses recent advances in NMR methodologies to investigate large proteins and nuclei acids and proposes ways to exploit these developments, possibly in combination with complementary techniques in structural biology, to study

high-molecular-weight RNP complexes in their functional forms. The structure of RNA–protein complexes with molecular weight (MW) < 50 kDa can be solved by standard NMR techniques, taking advantage of 13C/15N labeling of either the Trichostatin A solubility dmso protein or the RNA component of the complex. 13C/15N edited, 12C/14N filtered NOESY experiments [9] and [10] are instrumental for the detection of intermolecular NOEs. Structural studies in

solution are particularly relevant for proteins in complex with single-stranded short RNA sequences, which maintain some extent of disorder in the complex. Many RNA-binding domains are quite tolerant in terms Pyruvate dehydrogenase lipoamide kinase isozyme 1 of the RNA sequences they bind to; therefore, prior to the structural investigation, it is important to find the RNA sequence with the highest affinity for the cognate protein, which is likely to yield the best quality intermolecular NOEs. To this end, an NMR based method has been developed that uses the magnitude of the protein chemical shifts deviations upon titration of RNA to derive the sequence specificity of an RNA-binding domain [11]. The nucleotide type is varied systematically at each position within the RNA target, where the nucleotides at positions other than the one under analysis have a random identity. Analysis of the deviations of the chemical shifts of the target protein allows identifying patterns of sequence specific recognition at each nucleotide position, in a manner that is independent of the RNA structural and sequence context. The method works for target RNAs as long as 6–8 nucleotides, which in most cases covers the length of the RNAs recognized by the widespread RRM (RNA recognition motif), KH (K-homology), PAZ (Piwi/Argonaute/Zwille) and Zn-finger domains.