Molecular mechanisms of picornaviral proteases 2A and Lb

The family of picornaviruses contains a number of important human and animal pathogens such as poliovirus (PV), human rhinovirus (HRV), coxsackievirus (CV) and foot-and-mouth disease virus (FMDV). Vaccines have been used successfully to control infections by PV and FMDV. Vaccines are however not available for HRVs and CVs. In addition, the FMDV vaccines can only be used when certain conditions are met and the PV vaccines may not be sufficiently effective to ensure completion of the PV eradication program without support and backup of anti-viral agents. Given this background, it is therefore important to concentrate on the development of suitable anti-viral agents against these picornaviruses. Such attempts have over the years concentrated on using two picornaviral proteins, the proteinase 3Cpro and the polymerase 3Dpol; these enzymes are common to all picornaviruses and play roles in numerous steps of the virus replication cycle. Recently, however, a report has been published (Crowder and Kirkegaard, (2005) Nature Genetics, 37, 701-709) suggesting that the proteinase 2Apro of PV, CV and HRV would be an excellent target for anti-viral agents. This enzyme only performs a single, intramolecular cleavage on the viral polyprotein; however, inhibition of this cleavage will lead to elongated capsid proteins which will prevent the correct assembly of new progeny. Even if the enzyme is only partially inhibited, the effect of a few elongated capsid proteins on the overall viral infectivity should be significant. Similar effects with leader proteinase (Lb pro ) inhibitors of FMDV can be expected. It is the aim of the project to provide the necessary structural, biochemical and virological information to allow 2Apro and Lbpro to be developed as drug targets. Both proteinases carry out self-processing intramolecularly; structural data which reveal the mechanism of this reaction is however lacking. We will attempt to determine the X-ray crystal structure of 2A pro molecules possessing N-terminal extensions of different length from the preceding protein VP1. We will purify recombinant VP1-2Apro proteins from PV, CV and three different genetic groups of HRV in order to enhance the changes of success and increase the amount of information generated. We will also use NMR (nuclear magnetic resonance) to investigate FMDV Lbpro self-processing. We will further study the relationship of 2Apro by examining their sensitivity to a specific 2Apro inhibitor and certain derivatives which we will develop during the project and by investigating the role of C-terminal extensions on substrate specificity and sensitivity to inhibitors. 2Apro and Lbpro both cleave a cellular protein, eukaryotic initiation factor 4G, when this molecule is complexed to a second protein eIF4E. We will use NMR to study the interaction of a fragment of eIF4G with eIF4E and Lbpro to generate information on how Lbpro recognises this substrate. In summary, the combination of structural, biochemical and virological data should provide the information necessary for the design of anti-viral substances in the future.

The family of picornaviruses contains important human and animal pathogens such as poliovirus (PV), human rhinovirus (HRV), coxsackievirus (CV) and foot-and-mouth disease virus (FMDV). Vaccines have been used successfully to control infections by PV and FMDV. Vaccines are however not available for HRVs and CVs. In addition, the FMDV vaccines can only be used when certain conditions are met and the PV vaccines may not be sufficiently effective to ensure completion of the PV eradication program without support and backup of anti-viral agents. In this project, we examined two potential anti-viral candidates, the proteinase 2A pro from PV, HRV and CV and the leader proteinase (L pro ) from FMDV. One characteristic of the 2Apro is their ability to free themselves from a growing peptide chain by self-cleavage at their own N-terminus. We set out to determine the crystal structure of an inactive 2A pro containing an extension at its N-terminus. We were able to produce large amounts of this protein from the HRV serotype 2 as well as from PV and CV. Unfortunately, none of these proteins formed crystals, leaving us unsuccessful in this endeavour. In a second series of experiments with 2A pro , we could show that the cleavage site for the HRV14 enzyme is not accepted by the HRV2 enzyme, indicating the enzymes have, despite a similar structure, different specificities. Most importantly, we could show that the HRV2 enzyme recognises the shape and size of the amino acids at its cleavage site rather than the specific amino acids themselves. This has important implications for the development of anti-viral compounds. With the Lpro , we worked with a Brazilian group to determine the exact specificity of the enzyme using oligopeptide substrates. We could then use this information to generate a potent inhibitor of the enzyme and examine its structure bound to the enzyme. This showed us for the first time how the enzyme recognises its substrate on both sides of the cleaved bond, again providing information for the further design of anti-viral compounds. We also examined the self-cleavage of the Lpro and were able to document for the first time that this is an intramolecular one. In the final part of this project, we also examined the structure of a cellular protein termed "eukaryotic initiation factor 4G (eIF4G)". This protein is specifically cleaved by both the 2A pro and Lpro . However, the mechanism of this cleavage is poorly understood. We prepared an appropriate fragment of the eIF4G protein that was suitably labelled for examination by NMR and could show that the fragment is partially structured and partially random coil. Signals for about 50 of the 200 amino acids were assigned. However, the complexity of the molecule precluded the assignment of the remainder. It is known that eIF4G cleavage is enhanced by another cellular protein, eIF4E. In pilot experiments, we could detect changes in the eIF4G signals on addition of eIF4E. These results show that this system should provide insight into the mechanism of cleavage of Lpro and 2Apro and explain why cleavage is enhanced by the addition of eIF4E.