Acetohydroxyacid synthase is a multisubunit enzyme that catalyses the first committed step in the biosynthesis of the branched chain amino acids viz., valine, leucine and isoleucine. In order to understand the structural basis for the observed allosteric feedback inhibition in AHAS, the regulatory subunit of AHAS isozymes I from E. coli was cloned, expressed, purified and the conditions were optimized for solution NMR spectroscopy. IlvN was found to exist as a dimer both in the presence and absence of the feedback inhibitor. Using high-resolution multidimensional, multinuclear NMR experiments, the structure of the dimeric valine-bound 22 kDa IlvN was determined. The ensemble of twenty low energy structures shows a backbone root mean square deviation of 0.73 ± 0.13 Å and a root mean square deviation of 1.16 ± 0.13 Å for all heavy atoms. Furthermore, greater than 98% of the backbone φ, ψ dihedral angles occupy the allowed and additionally allowed regions of the Ramachandran map. Each protomer exhibits a βαββαβα topology that is a characteristic feature of the ACT domain fold that is observed in regulatory domains of metabolic enzymes. In the free form, IlvN exists as a mixture of conformational states that are in intermediate exchange on the NMR timescale. Important structural properties of the unliganded state were probed by H-D exchange studies by NMR, alkylation studies by mass spectrometry and other biophysical methods. It was observed that the dynamic unliganded IlvN underwent a coil-to-helix transition upon binding the effector molecule and this inherent conformational flexibility was important for activation and valine-binding. A mechanism for allosteric regulation in the AHAS holoenzyme was proposed. Study of the structural and conformational properties of IlvN enabled a better understanding of the mechanism of regulation of branched chain amino acid biosynthesis. Solution structural studies of 32 kDa Protease-VPg (PVPg) from Sesbania mosaic virus (SeMV) Polyprotein processing is a commonly found mechanism in animal and plant viruses, by which more than one functional protein is produced from the same polypeptide chain. In Sesbania Mosaic Virus (SeMV), two polyproteins are expressed that are catalytically cleaved by a serine protease. The VPg protein that is expressed as a part of the polyprotein is an intrinsically disordered protein (by recombinant expression) that binds to various partners to perform several vital functions. The viral protease (Pro), though possessing the necessary catalytic residues and the substrate binding pocket is unable to catalyse the cleavage reactions without the VPg domain fused at the C-terminus. In order to determine the structural basis for the aforementioned activation of protease by VPg I undertook the structural studies of the 32 kDa PVPg domains of SeMV by solution NMR spectroscopy. NMR studies on this protein were a challenge due to the large size and spectral overlap. Using a combination of methods such as deuteration, TROSY-enhanced NMR experiments and selective ‘reverse-labelling’, the sequence specific assignments were completed for ~80% of the backbone and 13C nuclei. NMR studies on mutants such as the C-terminal deletion mutant, I/L/V to A mutants in VPg domain were conducted in order to identify the residues important for aliphatic-aromatic interactions observed in PVPg. Attempts were made to obtain NOE restraints between Pro and VPg domains through ILV labelled samples; however these proved unsuccessful. It was observed that ‘natively unfolded’ VPg possessed both secondary and tertiary structure in PVPg. However, 30 residues at the C-terminus were found to be flexible. Even though atomic-resolution structure could not be determined, the region of interaction between the domains was determined by comparing NMR spectra of Pro and PVPg. The conditions for reconstitution of the Protease-VPg complex by recombinantly expressed Pro and VPg proteins were standardised. These studies lay the foundation for future structural investigations into the Protease-VPg complex.

Branched chain amino acids are classified as essential amino acids since their biosynthetic routes or pathways are restricted only to micro-organisms, fungi and plants. Given their unique distribution, the enzymes of the branched chain amino acid biosynthetic pathway are ideal targets for the development of herbicides, anti-bacterials and potentially antifungal agents. Acetohydroxy acid synthase (AHAS) catalyses the firs step in the biosynthesis of branched chain amino acids. AHAS activity had been first identified in extracts of E. coli as early as in 1958 by Brown and Umbarger . Ever since its discovery, AHAS have been found to exist in all eubacteria, archaebacteria, algae, yeast and plants. The enzymatic properties of prokaryotic and eukaryotic AHASs have been thoroughly investigated. A single isoform of the enzyme is known to exist in all organisms except in enterobacteria which have three isoforms of the enzyme. Activity of the three isoforms of E. coli AHAS (I, II and III) have been studied using various biochemical and biophysical methods. AHAS enzyme expressed in bacteria and yeast are heterotetrameric composed two large catalytic and two small regulatory subunits. While much has been learnt from the structure of the catalytic subunits (yeast and Arabidopsis thaliana) and the regulatory subunits (regulatory subunit of E. coli AHAS III) in isolation, the structural properties of the holoenzyme remain unexplored. AHAS is unique from the point of view that it exhibits a striking domain organization in the catalytic subunit and also in the regulatory subunits. Thus understanding the nature of protein – protein interactions both as domain – domain interactions within the subunit as well as protein – protein interactions across subunits is crucial to understanding the structural basis for the activity and regulation of this important enzyme. Of these, understanding the structural basis for the interaction between the regulatory and the catalytic subunits within the holoenzyme is paramount. The poor solubility and the intrinsic instability of the proteins have hampered the efforts to structurally characterize any of the AHAS holoenzymes. An active AHAS I construct has been created by Vyazmensky et. al., where the catalytic and the regulatory subunit have been expressed together as a single chain separated by a flexible linker. While this single chain construct is catalytically active, there have been no reports of successful crystallization of this single chain AHAS I enzyme. The crystallographically determined structure of the catalytic subunit of yeast and A. thaliana AHAS has shown that the protein is composed of three independently folded domains, α, β and γ. More importantly the polypeptide sequence of the catalytic subunits of AHAS across all species is largely conserved. This indicates that the overall tertiary folds of the catalytic subunit would be alike. The unique domain architecture of the AHAS catalytic subunit and the relatively small size of the regulatory subunit forms the basis for implementation of a novel strategy, in which structural interactions between the domains (catalytic site as well as the non catalytic site interactions) as well as structural interactions between the domains of the catalytic and the regulatory subunit of E. coli AHAS I can be explored in an incremental manner. Initiation of structural characterization of the individual domains of the catalytic subunit of E. coli AHAS I and understanding the structural basis of the interaction between the domains of the catalytic and the regulatory subunits of the protein, using solution NMR methods, forms the theme for this study. The domains of the catalytic subunit (ilvB) of E.coli AHAS I were identified based on the similarity in the sequence of this subunit with the yeast protein and the structural information of the yeast protein. The individual domains of the ilvB protein (ilvBα, ilvBβ and ilvBγ) and ilvN, the regulatory subunit of AHAS I, were cloned, expressed and purified for structural studies. The problem of poor expression and solubility profiles of the AHAS proteins was circumvented with the help of a novel cytb5 fusion system developed in our laboratory during the course of this study. The high expression levels of the fusion protein in minimal medium enabled the preparation of isotopically (15N, 13C/15N, 2H/13C/15N) enriched samples of the proteins in a cost effective manner. The cytb5 fusion system has provided very uniform and reliable expression of these proteins without accumulation of any protein in the insoluble fraction. From the structure of the catalytic subunit of yeast AHAS it is known that the α and γ domains of the protein interact to form the active site. The two domains provide group specific interaction sites for anchoring the co-factor TPP in an appropriate conformation for catalysis. The β domain on the other hand does not directly participate in the ormation of the active site but anchors the co-factor FAD which in turn plays a structural role in enzyme catalysis. In the present study we employed biochemical and biophysical methods to establish the structural integrity of the individually expressed domains of the catalytic subunit (ilvB) and the regulatory subunit of AHAS I. Reactions catalyzed by enzymes formed by assembling different domain and subunits indicate that the proteins when reconstituted in vitro form a catalytically competent complex. Formation of S-acetolactate, the product of the reaction catalyzed by the AHAS I holoenzyme, has been confirmed using colorimetric as well as spectroscopic methods such as CD and NMR. Multinuclear, multidimensional NMR methods have been utilized to obtain sequence specific assignments of apo - ilvBβ (non FAD bound form). Preparation of an NMR amenable sample of ilvBβ proved to be the rate limiting step due to the predisposition of the protein to undergo aggregation at concentrations required for solution NMR studies. However, careful screening of large number of buffer conditions enabled us to establish an optimum sample condition where the protein was soluble, stable and free of aggregation and hence suitable for NMR studies. Uniformly enriched 15N, 13C/15N, and 2H/13C/15N samples of ilvBβ were prepared to obtain sequence specific assignments and secondary structural information. From the secondary chemical shifts of backbone 13Cα atoms and short and medium range NOEs the secondary structure of the non FAD bound (apo) form of ilvBβ has been determined. Using chemical shift mapping methods, the residues of the ilvBβ domain that are involved in FAD binding have been identified. The distribution of the secondary structural elements and the residues that are involved in binding the co-factor FAD were found to be conserved for the E. coli and yeast proteins. This suggests that the tertiary Fold of the FAD binding β domain of the catalytic subunit of E. coli AHAS is identical to that in the yeast protein. The interaction between the individual domains of ilvB and ilvN (the regulatory subunit) has been investigated using spectroscopic methods. Changes in CD spectra indicate that ilvN interacts with ilvBα and ilvBβ domains of the catalytic subunit and not with the ilvBγ domain. NMR chemical shift mapping methods has shown that ilvN binds close to the FAD binding site in ilvBβ and proximal to the intra-subunit ilvBα/ilvBβ domain interface. The implication of this interaction and the role of the regulatory subunit on the activity of the holoenzyme are discussed.