Dr Alexander Golovanov
Structural and dynamic studies of biological macromolecules (mainly proteins) and their interactions using NMR spectroscopy.
Studies of mRNA export adaptor proteins.
REF proteins are found in the spliceosome, exon junction and TREX complexes and serve as export adaptors for TAP-dependent mRNA export from the nucleus. Using NMR, we characterized the domain architecture of REF2-I which consists of RNA recognition motif (RRM) domain flanked by two long flexible arms. We also described the REF2-I NMR structure (residues 1-155) which reveals a transient helix (N-helix) within the N-terminal arm. The N-helix binds transiently to the arm itself and RRM, but complex formation with DDX39, TAP-p15 or RNA triggers a conformational change whereby the N-helix, arm and RRM present an extended binding interface for these ligands. RNA, TAP-p15 and DDX39 binding sites on REF were established and characterized by chemical shift mapping and biochemical assays. For details see (Golovanov et al, RNA 2006) below. Further studies, currently funded by the BBSRC, are aimed at revealing the mechanism by which herpes viruses hijack cellular mRNA export system to export viral mRNA.
Similar studies were conducted for another mRNA export adaptor, 9G8. The sequence-specific RNA-binding proteins SRp20 and 9G8 are the smallest members of the serine- and arginine-rich (SR) protein family, well known for their role in splicing. We established the solution structure of the free 9G8 RRM which was compared with the structure of SRp20 RRM in complex with the RNA fragment, determined by our collaborators (group of Prof. FH Allain, ETH, Switzerland). Furthermore, a short arginine-rich peptide adjacent to the SRp20 and 9G8 RRMs was identified, which does not contact RNA but is necessary and sufficient for interaction with the export factor TAP. Together, these results provide a molecular description for mRNA and TAP recognition by SRp20 and 9G8. For details see (Hargous et al., EMBO J. 2006). These studies have been conducted in collaboration with Dr Stuart Wilson, Sheffield University.
Development of NMR methods of monitoring complex multiple protein-protein-ligand(s) interactions.
Modern physico-chemical and biochemical methods allow thorough characterisation of protein-protein interactions, by measuring some kind of "reporter signal" reflecting these interactions, or by capturing a snapshot - 3D structure of complex. What if the interaction involves three or more components, which exhibit consecutive, cooperative or concurrent binding: how can we tell what happens dynamically with each existing component as new components are added? Limited number of reporter signals no longer can adequately represent such a complexity. Monitoring such dynamic and transitional events which all are elements of protein interaction networks is intrinsically difficult. My current research is aimed at development of new isotopically discriminated (IDIS) NMR approaches ebabling to monitor complex multi-component binding events in vitro (Golovanov et al., J. Amer. Chem. Soc 2007, 129(20):6528).
Probing critical protein interactions in cellular pathways suppressing cancer.
The pilot project, funded by the Royal Society Research grant, aims at applying the isotopically discriminated (IDIS) NMR to the studies of consecutive, competitive and cooperative interactions in cell signalling pathways related to cancer. The influence of biologically-active ligands and drugs on these protein interactions will be explored. This work is a collaboration with Dr Costas Demonacos, School of Pharmacy.
Solubilisation of proteins for structural studies
Increasing protein concentration in solution to the required level without causing aggregation and precipitation is often a challenging but important task, especially in the field of structural biology, including NMR. Projects involving studies of biologically interesting and important proteins sometimes fail just because no suitable experimental conditions could be found to conduct the whole wish-list of experiments. In our work we discovered that a simple addition of 50 mM L-Arginine and L-Glutamate to the sample buffer increases protein solubility up to 9 times and suppresses proteolytic degradation of protein samples in the course of long NMR experiments. Addition of these amino acids even in non-deuterated form is not causing any significant interference with the typical set of heteronuclear and triple –resonance experiments used for spectral assignment and structure calculation of isotopically-labelled proteins. The solubilisation protocol have been successfully used for a number of proteins with solubility problems, suggesting that it may be universally applicable. For details see (Golovanov et al, J.Am.Chem.Soc., 2004) below. My further studies are aimed at adoption of solubilisation protocol for optimizing the NMR Cryoprobe sensitivity.
Solution protein structure determination.
NMR is a powerful method allowing determination of atomic-resolution structures of proteins. Some proteins fail to crystallize due to the presence of flexible tails or loops, and thus are not amenable for X-ray analysis: NMR is the method of choice in these cases.
In collaboration with other members of the Department, I also have assigned the spectra and established the 3D structure of protein ParG, which is involved in active plasmid partitioning. For more details on the work see (Golovanov et al., Mol. Microbiol., 2003) below. The work has evolved further to characterisation of interactions of flexible N-terminal domain of ParG with DNA using NMR technique. In combination with such methods as two-hybrid assays, SPR, band-shift assays, etc., provided by my collaborators (group of Dr. F.Hayes and Dr. D. Barillà), NMR analysis helped to identify the presence of transient &beta-structure and to reveal the role of N-terminus (Carmelo et al., J.Biol.Chem., 2005, 280:28683).
Another NMR protein structure determined recently by me (in collaboration with the group of Dr. J.Derrick) was that of Neisseria meningitidis pilot protein PilP. The structure revealed a new, previously unknown beta-sandwich fold. The hydrophobic cavity identified within this structure suggested that the function of this protein may involve the binding of lipophilic ligands. For details of this study see (Golovanov et al., J.Mol.Biol. 2006) below. The detailed studies of PilP protein and its associations with PilQ are currently subject of studies in Dr J.Derrick’s group.
I was also involved in collaboration with the group of Dr J.Avis on NMR studies of complexes between WW domains of a Drosophila Nedd4 family protein called Suppressor of deltex (Su(dx)) and proline-rich peptides (Fedoroff et al., J.Biol. Chem. 2004, 279:34991; Jennings et al., J Biol Chem. 2007 282:29032).