Summary of case study lecture at Biophysics and Structural Biology at Synchrotrons Workshop Triumph over Adversity: structure of arylamine N-acetyltransferase from M. tuberculosis

Abstract

Mycobacterium tuberculosis (TB) remains the worldwide leading cause of death by an infectious agent. It is contagious and very widespread: in 2017, there were 10 million new cases worldwide and 1.6 million deaths from TB, and 0.3 million of these were associated with HIV infection. TB can lie latent for years in the body until the host becomes immunocompromised, and then the disease takes hold. Treatment is complicated and lengthy, involving 4 different antibiotics taken concurrently for 6 months. The bacterium has become multidrug, extensively or totally drug resistant over the last 50 years, so new treatments for TB are urgently required. To aid the search for new alternatives to classical antibiotics, which ideally would have a low ability to induce resistance and represent a large potential range ofmolecular moieties, the 3-D structures of more proteins from TB are needed. One possible drug target is arylamine N-acetyltransferase (TBNAT), a 31-kDa, 273 amino acid cytosolic enzyme. The nat gene in mycobacteria is in a gene cluster which is essential for cholesterol catabolism and intracellular survival of mycobacteria inside the macrophage. The NAT enzymes are known to catalyse the transfer of an acetyl group from acetylCoA (Ac-CoA) to an arylamine substrate. Studies suggested that the TBNAT is involved in biosynthesis of the mycolic acid component of the protective coating of the bacterium that it grows to avoid detection by the body’s immune system when within a human cell. Deletion of the gene coding for this protein inM. bovis (bovine TB) has been found to deplete the amount of mycolic acid and to render the bacterium sensitive to the antibiotic gentamicin. Starting in 1995, extensive efforts to characterise TBNAT by Professor Edith Sim and her group at the Oxford University Pharmacology Department had found the protein to be very hard to express at the purity and in the quantities required for crystallisation trials, and for 15 years, successive generations of graduate students had worked to solve the challenges involved. As a model for TBNAT, they used a homologous enzyme from M. marinum (MMNAT), a bacterium affecting fish. MMNAT has 74% sequence identity with TBNAT and had been thought to be a suitable model for drug discovery, but both high-throughput screening experiments for inhibitors and in silico modelling showed the two proteins to have different inhibitor profiles, TBNAT giving IC50 values for putative inhibitors that were between 5 and 60 times larger than those for MMNAT. The team had already determined the structure (see Fig. 1a) ofMMNATwhich readily gave crystals (see Fig. 1b), but knowledge of the TBNATstructure was clearly required to understand the different inhibitor profiles. Thus, renewed efforts were made by a new and very dedicated, skilled and determined graduate student from Jordan, Areej Abuhammad (see Fig. 1e), to obtain enough pure TBNAT protein for crystallisation trials. The first major challenge was to find a new vector and better recombinant expression system (E. coli, pVLT31, NEB express strain), and after 2 years of hard work, Areej managed to increase the expression efficiency from 2 mg/l to 8mg/l. Additionally, by optimising the purification protocol (using a cobalt gradient IMAC column for His-tagged cleaved protein and a nickel column for His-tagged protein), she pushed this up to 16 mg/l and the purity from < 90% to > 95%, good enough for crystallisation trials to commence. Screening for crystallisation conditions is still rather unscientific and is largely a process of setting up many screens at different temperatures and a range of protein concentrations with different precipitants and additives. In the case of TBNAT, over 7000 (100 nl protein+100 nl precipitant) conditions were tested, but with no crystals at all being obtained. In a final attempt to grow TBNAT crystals, we tried crossThis article is part of a Special Issue on ‘Biophysics & Structural Biology at Synchrotrons’ edited by Trevor Sewell.