oa Thymidylate Kinases as Potential Anti-Cancer and Antibiotic Drug Targets

Thymidylate kinases (TMKs) play a central role in the production of nucleotide precursors that are required for the replication of DNA. Consequently, this enzyme is a potential drug target for the discovery of anti-bacterial, anti-fungal, and anti-parasitic drugs. In addition, TMKs are also involved in the activation of prodrugs. In particular, the anti-HIV drug AZT is activated by human TMK (huTMK) and the low efficiency of huTMK towards AZT is a significant problem in the use of AZT in the treatment of HIV. Finally, nucleotide precursors are required in large amounts by cancerous cells, thus the inhibition of huTMK by chemotherapeutic agents may enhanced the arsenal of drugs that are used to treat cancer.

Although there has been some effort to develop inhibitors of TMKs, these efforts have been hampered by the difficulty in performing high throughput screening using compound libraries. In addition, the characterization of TMK-drug complexes has been limited to X-ray diffraction studies which provide static information about the enzyme-drug complex. There have been no attempts to apply high-resolution multi-nuclear NMR techniques to determine the fundamental dynamic properties of these enzymes and how the structure and dynamics of the enzyme are altered by the binding of substrates or inhibitors.

As a preliminary step in characterizing these enzymes by NMR we have over-expressed TMKs from yeast, human, and two pathogens - Plasmodium falciparum and Candida albicans. Expression of these TMKs was optimized by the design of synthetic genes for expression in bacteria. In the case of the human enzyme, we are able to routinely produce 250 mg of the enzyme/L of culture. Preliminary NMR spectra of the yeast, human, and plasmodium enzyme show that the protein is a homo-dimer in solution, as anticipated from X-ray studies. The amide and methyl spectra are well resolved, indicating that resonance assignment by traditional TROSY based methods will be feasible for both the amides and the methyl resonances. In particular the high sensitivity and dispersion of the methyl spectra will facilitate characterization of the dynamic properties of these enzymes by carbon and deuterium relaxation. Ligand induced changes in the dynamics and structure of huTMK in solution will be characterized using NMR methods. These studies will provide additional insights into the inability to huTMK to effectively activate AZT. The entropic component of the thermodynamics of substrate binding to TMK from the parasite that causes malaria will also be characterized by determining dynamic changes by NMR methods.

The development of NMR methods to study these enzymes also provides a method for high throughput screening of compound libraries by detecting chemical shift changes in the NMR spectral of the enzyme due to binding of a potential lead compound.