Effective disruption of the pathway that allows cancer cells to survive under hypoxic conditions for use as anticancer therapy has long been a goal of cancer researchers. Andrew L. Kung et al. have now discovered a small molecule that disrupts the structure of a key coactivator in the hypoxia-inducible factor (HIF) pathway and inhibits tumour growth in vivo.

Activation of the HIF1 pathway is linked with resistance of tumours to therapy, increased invasion and metastasis, and poor outcome. As oxygen concentrations decrease, HIF1α accumulates and dimerizes with HIF1β, and the coactivator p300/CREB binding protein (CBP) is recruited. This complex then binds to the hypoxia-response element to trigger transcription of genes that facilitate adaptive mechanisms.

Kung and colleagues developed a high-throughput screen to search for small molecules that could inhibit the crucial interaction of HIF1 with p300. They immobilized the 41-amino-acid polypeptide p300/CBP-binding domain of HIF1α (TADC) on multiwell plates and tested a library of over 600,000 compounds for the ability to disrupt binding of the 121-amino-acid HIF1α-binding domain of p300 (CH1) to the plates. After confirmatory in vitro and cell-based assays, a single specific inhibitor — chetomin — was found. In a luciferase reporter assay, the authors showed that while chetomin did not interfere with p300-dependent transcriptional activity of factors that bind to most domains in p300, it did prevent the activity of factors that bind specifically to the CH1 domain. NMR spectroscopy revealed that CH1 becomes less structured in the presence of chetomin, indicating that this is what prevents interaction with HIF1α.

So, what happens under hypoxic conditions in vivo? When chetomin was administered to mice bearing tumour xenografts, levels of vascular endothelial growth factor, which is induced by hypoxia, were attenuated in a dose-dependent manner. Serum levels of erythropoietin, a marker of physiological HIF1 function, were also decreased. To directly determine the effect of chetomin on the HIF1 pathway, the authors placed luciferease under the control of the erythropoietin enhancer — under hypoxic conditions the reporter activity increased more than 100-fold. On injection of chetomin, but not vehicle control, into mice bearing the hypoxia-reporter cell line in the right flank and a constitutively expressed luciferase cell line in the left flank, reporter activity in the right flank only was reduced, by about 50%. These results also indicate that chetomin specifically disrupts the TADC–CH1 protein–protein interaction within tumours.

Chetomin also significantly reduced colon and prostate tumour xenograft growth and led to substantial necrosis in the tumours. Local toxicity at the injection sites was seen, but the cause of this is unknown.

Disruption of the tertiary structure of CH1 domain of p300 specifically inhibits binding to HIF1 and signalling through the HIF1 pathway, indicating a novel small-molecule approach of interfering with cancer growth.