Targeting hypoxia for cancer therapy
Tumors contain localized areas of low oxygen (hypoxia). Cells adapt to survive in these areas via multiple compensatory mechanisms including stimulation of angiogenesis and metabolic switching to glucose as an energy source, so reducing their oxygen dependency. We have developed a physiologically relevant human cell-based assay to screen small molecules for their ability to inhibit such adaptive responses to hypoxia. Specifically, we exposed cancer cells and macrophages to hypoxic conditions and measured VEGF secretion as a marker of angiogenic pathway activation plus lactate secretion and glucose uptake to identify changes in glucose metabolism. Cell number was also measured to highlight compounds with toxic effects.
A specialised hypoxia chamber (BioSpherix Inc) was used to treat A172 glioblastoma cells to 1% oxygen to mimic the level reported at the hypoxic core of tumours. Cells were pre-treated with 10µM compound prior to exposure to hypoxia. Temsirolimus, a robust inhibitor of hypoxia driven VEGF and lactate secretion and also glucose uptake was used as a positive control. Cell density, VEGF and lactate secretion plus glucose remaining in culture were measured 48 hours after treatment using the following commercially available assays respectively; CyQUANT (Thermofisher), VEGF AlphaLISA (PerkinElmer), L-Lactate assay (Cayman) and a glucose detection assay (Promega).
U937 monocytic cells were differentiated into a macrophage phenotype using phorbol 12-myristate 13- Acetate (PMA). Differentiated macrophages were exposed to hypoxia or high levels of lactate to mimic tumour micro-environment conditions. Control macrophages were differentiated with IL-4 and IL-13 to generate an M2 phenotype, known to share characteristics with tumour associated macrophages. Differentiated macrophages were characterised by flow cytometry and following hypoxia treatment, levels of VEGF in the supernatant measured as described above.
Figure 1. A and B – A172 glioblastoma cells respond robustly to hypoxia treatment, increasing VEGF secretion and glucose utilisation from media, demonstrating an adaptive phenotype enabling survival of tumour cells in areas of low oxygen. This hypoxia driven response is inhibited by Temsirolimus. Data = mean of triplicate cultures +/- SD.
Fig 1.A – Inhibition of VEGF secretion and glucose uptake with a single concentration of Temsirolimus (10µM)
Fig 1.B – Inhibition of VEGF secretion and glucose uptake with a titration of Temsirolimus
Figure 2. A, B and C – A panel of small molecules were screened at 10µM in A172 glioblastoma cells for their ability to inhibit hypoxia driven responses. Of the compounds tested, 149/823 (18%) inhibited VEGF secretion (Fig 2.A), 38/823 (5%) decreased glucose uptake (Fig 2.B) and 41/823 (5%) inhibited both VEGF secretion and glucose uptake (Fig 2.C).
Fig 2.A – Small molecule activity on hypoxia driven VEGF secretion
Cell density and VEGF secretion in the presence of test compound were calculated as a percent of that measured in DMSO control treated cells. Data = mean of triplicate wells. Compounds in the box are hits, defined as those that inhibited VEGF secretion greater than 40% without decreasing cell density by more than 30%.
Fig 2.B – Small molecule activity on hypoxia driven increased glucose uptake
The concentration of glucose remaining in the media in the presence of test compound was calculated as a percent of that measured in the supernatant from Temsirolimus control treated cells. Cell density was calculated as a percent of DMSO control cells. Data = mean of triplicate wells. Compounds in the box are those that demonstrated a greater than 40% inhibition of glucose uptake without decreasing cell density by more than 30%.
Fig 2.C – Small molecule activity on both hypoxia driven VEGF secretion and glucose uptake
VEGF versus glucose measurements in the media were plotted against each other to identify small molecules that were hits in both assays. Compounds in the box are those that demonstrated a greater than 40% inhibition of VEGF secretion and a greater than 40% inhibition of glucose uptake.
Figure 3 – In addition to increased glucose uptake, A172 glioblastoma cells secrete higher levels of lactate into the media in response to hypoxia, a by-product of elevated glucose metabolism. Data = mean of triplicate cultures +/- SD.
Figure 4.A, B – U937 monocytic cells were differentiated into a macrophage phenotype with PMA treatment as determined by increased adherence, granularity and up-regulated expression of surface CD11b. Exposure of macrophages to hypoxia or treatment with lactate stimulated VEGF secretion, suggesting macrophages contribute to tumour progression by enhancing angiogenesis and increasing local oxygen.
Fig 4.A – Evidence of Macrophage differentiation – phase contrast images and flow cytometry data
Fig 4.B – VEGF secretion from macrophages cultured under hypoxic conditions or treated with exogenous lactate. Data = mean of triplicate cultures +/- SD
For more details,see the poster entitled “Macrophage responses under hypoxic conditions- a drug target for the discovery of anti-cancer agents?” which was presented at ELRIG Drug Discovery (Liverpool, 2016)
Would you like more information?