 |
| The speakers at the symposium: (left to right), back row, Dr Griffin,
Professor Denny, Dr Johnson and Professor Potter; front row, Professor Hecht and Professor Hurley |
The determination of protein crystal structures complexed with small molecule
inhibitors was an invaluable aid to guiding the design of new anticancer drugs,
suggested Dr Roger Griffin (anticancer drug discovery initiative, University
of Newcastle).
In the first of two examples of crystallography-led rational inhibitor design,
Dr Griffin described the development of inhibitors of cyclin-dependent kinases
(CDKs). The CDK enzymes and their “partner” cyclins were important
in the control of the cell cycle, particularly at checkpoints, he said. Inhibition
of these enzymes might arrest cell cycle progression and block the growth of
tumours. CDKs were serine-threonine protein kinases and so had an adenosine
triphosphate (ATP) binding site that could be targeted for drug design, he added.
[In a similar way to tyrosine kinases described on p489, CDKs catalyse transfer
of phosphoryl groups from ATP to serine and threonine groups of the substrate
protein].
In collaboration with researchers at the University of Oxford, purine- and pyrimidine-based
inhibitors of CDK-1 and CDK-2 had been identified, said Dr Griffin. The compounds
demonstrated a competitive activity that was comparable with the benchmark CDK
inhibitor, olomoucine. Studying the crystal structure of a substituted purine,
O6-cyclohexylmethylguanine (NU2058) in complex with CDK-2, identified a triplet
of hydrogen bonding (Glu81 to NH-9, Leu83 to 2-NH2 and N-3). This hydrogen-bonding
was also displayed by the pyrimidine compound, 2,6-diamino-4-cyclohexylmethyloxy-5-nitrosopyrimidine
(NU6027). Olomoucine, which had a similar structure to NU2058, bound in a different
orientation. Such differences in binding were reflected in a pattern of cellular
sensitivities for NU2058 and NU6027 that were distinct from olomoucine.
The second example of structure-based design concerned an enzyme involved in
DNA repair, the poly(ADP-ribose) polymerase, PARP, said Dr Griffin. This was
a ubiquitous nuclear enzyme with an absolute requirement for DNA damage for
activation and, as such, was implicated in resistance to several DNA-damaging
anticancer agents and radiation therapy.
Agents that inhibited PARP and enhanced such treatments were now being designed.
These bound to the nicotinamide adenine dinucleotide (NAD+) cofactor binding
site located in one of the three domains of the PARP enzyme.
Among several promising classes of inhibitors, the benzimidazole-4-carboxamides
showed particular affinity for the NAD+ binding site (IC50 2.5nM), said Dr Griffin.
Examination of the crystal structures of these compounds complexed with the
catalytic domain of PARP identified extensive interactions between the agents
and the a-carbon backbone of the protein. In assays of cell growth inhibition,
sub-micromolar concentrations of novel benzimidazoles significantly enhanced
the activity of both the topoisomerase I inhibitor, topotecan, and the DNA-methylating
agent, temozolomide, Dr Griffin concluded.