We have been involved in solving the crystal structures of catalytic fragments of various human ARTD enzymes and this has made it possible to understand the structural basis of inhibitor selectivity. The catalytic fragment is an α/β domain and it includes the conserved ARTD signature motif (Figure 2). ARTD has two binding sites: donor and acceptor sites. The donor site where NAD+ binds is similar in all ARTDs, but the surroundings of the acceptor site are not well conserved. Most inhibitors target the donor NAD+ binding site and structural studies have elucidated the binding modes of several of them. While the nicotinamide binding cleft is conserved, interactions with surrounding residues can be utilized to achieve selectivity.
Figure 2. Crystal structure of the catalytic domain of human tankyrase 1, with NAD+ (donor) and target protein (acceptor) binding sites labelled. The ARTD signature motif is shown in red.
As commercial assays are expensive or nonexistent, we have made an effort to develop and optimize activity-based screening assays for ARTDs. The current assay we use is based on conversion of a substrate to a fluorescent analogue on a 96-well plate and we then measure a decrease in fluorescence after the enzymatic reaction. We have been able to optimize the assay for several ARTDs (Narwal et al. 2012a, Venkannagari et al., 2013) and we have demonstrated that it can be used to determine selectivity of the inhibitors. This makes it possible for us to efficiently screen inhibitor libraries to find new inhibitor scaffolds and to assess inhibitor selectivity.
Our inhibitor discovery efforts have recently been focused on tankyrases (ARTD5 and ARTD6). In 2012 we described the first inhibitor of the family that binds to a new site within the catalytic domain (Narwal et al. 2012b). Binding of tankyrase and ARTD1 selective inhibitors to tankyrase 2 caused conformational changes in the substrate-binding loop and revealed interactions explaining isoenzyme selectivity (Figure 3). The structures enabled us to create new models and gave us ideas to be used in structure-based drug design. During assay development work we identified flavones as being potent and selective tankyrase inhibitors (Narwal et al. 2012a). Further evaluation of tankyrase inhibition by flavonoids has now been studied using biochemical assays and by crystallography (Narwal et al., submitted). This revealed that flavones can be modified to be even more potent and selective tankyrase inhibitors despite the fact that they bind to the conserved nicotinamide binding site of the enzyme (Figure 3). These studies will hopefully facilitate the development of tankyrase inhibitors as drugs against cancer.
Figure 3. Superposed crystal structures of tankyrase 2 in complex with flavone (blue) and IWR-1 (magenta) bound to different parts of the donor site. The conformational change induced by binding of IWR-1 is highlighted.
ARTDs in parasites
In contrast to humans, Trypanosomal parasites contain only one identified ARTD enzyme. In order to evaluate the biological functions of the DNA damage-activated ARTD of Trypanosoma cruzi, we studied inhibition of the recombinantly expressed protein by known ARTD inhibitors. Together with our collaborators, we identified the best inhibitors and verified that the compounds also inhibited ADP-ribose polymer formation in the parasites. Notably, the best compound, Olaparib, was able to significantly slow down the growth of the proliferative form of the parasite in culture even at nanomolar concentrations. The inhibitor also decreased the number of infected human cells in culture, indicating that ARTD inhibition could be a potential new way of interfering with parasitic infection (Vilchéz Larrea et al. 2012).
Last updated: 28.10.2016