In the work described here, we have performed a chemical genetic screen designed to find compounds that can preferentially inhibit the growth of yeast cells deleted for MTAP. Using this screen we have identified a series of compounds containing a 1,3,4-thiadiazine ring that have enhanced growth inhibition in both yeast and human cells lacking MTAP activity. The technology used to perform the screening was simple and did not require large amounts of expensive robotic equipment. To differentially screen 30,080 compounds in duplicate required approximately 200 plates and could be performed by a single person in a few weeks. However, there are some limitations to the approach. First, there was a high rate of false positives in the primary screen, and it was essential to include secondary and tertiary screens to confirm potential hits. We suspect that the high rate of false positives may be related to the variability of drug transferred to each well by the disposable pin transfer units. A second limitation of the approach is that only one concentration of the starting compound was tested. Since most of the compounds tested have relatively steep dose response curves, compounds that may be differentially effective may be missed either due to too high or too low drug concentrations. A third limitation concerns the ease of re-synthesis of “hit” compounds. It is sometimes difficult to re-synthesize the compound in larger quantities, as was the case in our initial hit compound, 1487–1117. A final limitation of this approach is that the actual molecular targets of the drug molecules identified are unknown and determination of the target(s) would not be trivial.
The compounds identified as having increased growth inhibition in meu1
Δ yeast contained a 1,3,4-thiadiazine ring and a phenol ring. Compounds containing these moieties have been shown to have a variety of biological effects including anti-microbial activity, anti-inflammatory activity, act as phosphodiesterase inhibitors, and metalloprotease inhibitors 23,25–27
In yeast, the halogenation of the phenol ring appeared to be important, as the one compound lacking halogenation, 93A did not show differential growth inhibition. Interestingly, halogenation of the same position of the phenol was shown to increase the effectiveness of related compounds in both an anti-inflammatory assay and as an inhibitor of matrix metalloproteases.23,27
It should also be noted that the parent compound 1487-117 was the most potent with regards to growth inhibition in yeast in both MEU1 and meu1Δ cells. This compound is distinguishable from the other compounds in the series by two modifications: 1) the addition of a brominated phenol group attached to the amino group hanging off C2 of the thiadizaine ring and 2) alkylation of the N4 position of the same ring. Further studies will be required to understand the importance of these modifications with regards to the increased potency of this compound.
The reason for the increased sensitivity of MTAP−
cells to these compounds is unknown. MTAP is highly conserved in evolution, but at the cellular level seems to be entirely dispensable for growth. MTAP's known enzymatic activity is to provide the cell with adenine and methionine from MTA. However, in the yeast assay described here, adenine and methionine are supplemented in the media, so interference with the de novo
production of these metabolites is not likely to result in a growth phenotype. However, it may be that MTAP has other as yet undescribed functions. Interestingly, the yeast MTAP orthologue, MEU1
nhancer of U
), was initially identified in a genetic screen designed to identify transcription factors regulating the Adh2
Also, microarray experiments show there are over 300 transcripts that are differentially regulated between MTAP+
HT1080 cells (WDK, in preparation). Taken together, these observations suggest that loss of MTAP may have more wide ranging effects on cellular physiology than previously thought. It is possible that the compounds described here have increased sensitivity in MTAP−
cells because it affects one of these other MTAP regulated pathways.
Even though the screening system described here was successful in identifying compounds that had some increased potency in inhibiting the growth of MTAP− yeast and mammalian cells, the differential IC50 of these compounds was relatively modest. Furthermore, in HT1080 cells none of the compounds had IC50 values less than 10 μM. However it is possible that further optimization of the compounds may lead to increased potency and potential usefulness. Alternatively, screening a much larger compound library using isogenic MTAP+ and MTAP− human cell lines may reveal new compounds with greater selectivity and potency. The feasibility of this approach is strengthened by the findings presented here.