Supplementary Materials Supporting Information supp_293_11_3913__index. metabolic state induces MICA expression. Specifically, we show that glucose transport into the cell and active glycolytic metabolism are necessary to up-regulate MICA expression. Active purine synthesis is necessary to support this effect of glucose, and increases in purine nucleotide levels are sufficient to induce MICA expression. Metabolic induction of MICA expression directly influences NKG2D-dependent cytotoxicity by immune cells. These findings support a model of MICA regulation whereby the purine metabolic activity of individual cells is reflected by cell-surface MICA expression and is the subject of surveillance by NKG2D receptor-expressing immune cells. has many potential clinical applications: up-regulation of MICA could promote cancer immunity, and down-regulation could be beneficial in autoimmune disease or transplantation. Multiple factors have been associated with changes in MICA expression, including activation of the DNA damage response pathway (19), Toll-like receptor (TLR) stimulation (10), histone deacetylation (20), heat shock transformation (21), ionizing radiation (22), growth factor pathway activation (23), cell-surface shedding (24), and microRNA expression (25). In addition, a number of gene-regulatory elements and transcription factors are known to play a role in MICA induction (11, 26). However, an integrated understanding of the mechanisms determining MICA expression remains elusive. MICA expression in human primary cells or tissue samples is found in settings independently associated with high metabolic activity (increased glucose uptake, glycolysis, high lactate output, and proportionate reduction in TCA cycle metabolism, or Warburg metabolism (27,C31)). This state of activated metabolism can be considered as a biosynthetic state, where enhanced glycolytic flux generates intermediate substrates for biomolecule synthesis (32). High-energy purine nucleotides, such as ATP, are among the downstream products. Here, we show that glucose metabolism leading to the generation of high-energy purine nucleotides, a process at the core of the Warburg effect, induces cell-surface expression N-Shc of MICA. We demonstrate that MICA induction by high-energy purine nucleotides is associated with increased NKG2D-dependent cellular immunogenicity and susceptibility to NK cell cytotoxicity, supporting our hypothesis that NKG2D provides a mechanism for immune oversight of metabolically activated cells. Results Glucose induces MICA expression We hypothesized that the transition from quiescent to activated or Warburg metabolism plays an important role in NKG2D ligand induction. To test this hypothesis, we used glucose restriction to model quiescent activated metabolism and observed BMS512148 novel inhibtior a direct correlation between the glucose concentration of culture medium and cell-surface expression of MICA in human embryonic kidney (HEK)-293T cells, cervical cancer cells (HeLa), fibrosarcoma cells (HT1080), and breast cancer cells (MCF7) (Fig. 1). Open in a separate window Figure 1. Glucose induces MICA expression. 293T (human embryonic kidney), HeLa (cervical cancer), HT1080 (fibrosarcoma), and MCF7 (breast cancer) cells were cultured for 48 h in medium containing 5 mm glucose that was then replaced with fresh medium containing either 0, 2.5, 5, 12.5, or 25 mm glucose. The cells BMS512148 novel inhibtior were cultured for a further 48 h in these conditions before cell-surface MICA expression was measured by flow cytometry. MICA expression rose with the glucose concentration. The represents the isotype control sample, and the glucose concentration is indicated by the and 0.05) (Fig. 2 0.005) in cells cultured in 25 mm glucose (Fig. 2and and and and and 0.0001). 0.0001). 0.05). 0.005). 0.0001), but eGFP itself is not induced by high glucose. and and nucleotide synthesis (Fig. 4). We hypothesized that nucleotide synthesis might mediate GIME. Because the synthesis of the purine nucleobase is directly dependent on the supply of proximal glycolytic metabolites, we first tested this hypothesis by treating cells cultured in high glucose (25 mm) with two inhibitors of purine synthesis, 6-diazo-oxo-norleucine (DON) and azaserine. Both compounds prevented GIME (Fig. 5, and purine BMS512148 novel inhibtior synthesis was tested using hypoxanthine, aminopterin, and thymidine (HAT)-selected cells. Whereas cells grown in standard culture medium depend BMS512148 novel inhibtior on purine synthesis, HAT-selected cells use the salvage pathway exclusively for new purine nucleotide synthesis. Azaserine inhibited GIME only in cells grown in standard culture medium and had no effect on HAT-selected cells (Fig. 5, and purine synthesis. DON probably has additional off-target inhibitory effects. The addition of a purine salvage pathway substrate to azaserine-treated cells in high glucose caused dose-dependent MICA expression (Fig. 5purine synthesis. The inhibitors DON and azaserine inhibit enzymes in the purine synthesis pathway proximal to the.