Electron transport via DHODH activity will
Electron transport via DHODH activity will allow the generation of mitochondrial ROS without the consumption of α-ketoglutarate or NADH. Glutamine-dependent flux through the TCA cycle and oxidative phosphorylation can also enable generation of mitochondrial ROS (Weinberg et al., 2010). The latter process requires partitioning of αKG derived from glutamine into the TCA cycle to generate reducing equivalents in the form of NADH, leading ultimately to reduction of molecular oxygen by Complex III to generate ROS (Weinberg et al., 2010). However, since production of NADPH via GOT1/MDH1/ME1 (Son et al., 2013, Yun et al., 2015) also requires consumption of αKG and NADH, these metabolites must partition between the two pro-survival pathways. DHODH activity may thus be required to enable high flux through both of these pathways simultaneously. KRAS mutant cells, unlike WT cells, also exhibit a large brequinar-mediated increase in fructose-1,6-bisphosphate levels (Figure S7). However, there does not appear to be a clear mechanism for DHODH inhibition to mediate this effect directly. One possibility is that developing arrest of the cells in S phase leads to upregulation of glycolytic flux (Kaplon et al., 2015) and an increase in fructose-1,6-bisphosphate (which is downstream of the rate-limiting enzymes in this pathway). In that case, the upregulation of glycolytic flux would appear to be mostly through the anaerobic (lactic acid) branch of the pathway; brequinar treatment is associated with a decrease in the steady-state level of citrate without a concomitant decrease of isotopic exchange into this metabolite (Figure S7), suggesting a decreased flux of metabolites into the citric (-)-Bicuculline methobromide molecular cycle. The relative functional importance of DHODH\'s pyrimidine biosynthetic, energy-producing, and redox-modulating functions is likely to depend on cellular context. DHODH has previously been characterized as a synthetic lethal target in several different oncogenic cell contexts, including the BRAF V600E mutation (White et al., 2011), PTEN deficiency (Mathur et al., 2017), and triple-negative breast cancer cell lines (Brown et al., 2017). In the case of the BRAF V600E mutation, oncogenic mutations of both BRAF and KRAS are gain-of-function with respect to the Ras-Raf-MEK-ERK pathway, which may account for the shared phenotype with respect to DHODH inhibition. However, KRAS signaling affects multiple signaling pathways in addition to Raf-MEK-ERK, so, in general, interventions that affect KRAS mutant cells may or may not affect BRAF mutant cells. Recently, de novo pyrimidine metabolism through carbamoyl phosphate generated in the urea cycle has been shown to be important for growth/survival in the KRAS/LKB1 double-mutant context (Kim et al., 2017), suggesting that DHODH inhibition might be useful specifically in this cellular context as well. The in vitro and in vivo results presented here suggest that DHODH inhibition may be clinically useful in the treatment of KRAS mutant tumors. However, in previous clinical trials, brequinar failed to demonstrate efficacy in multiple tumor types, including pancreatic and colorectal, a substantial fraction of which almost certainly harbored KRAS mutations (Moore et al., 1993, Peters et al., 1992). Understanding the basis for the discrepancy between in vitro/in vivo models and clinical results might enable the design of new and more effective clinical treatment strategies. One possible explanation for the lack of observed clinical effectiveness for brequinar is that target engagement in the clinical studies was insufficient to induce tumor cell death. In vivo studies suggest that prolonged exposure to brequinar is required to achieve the metabolite depletion effects required for maximum efficacy (Dexter et al., 1985, Schwartsmann et al., 1988). However, the clinical trials of brequinar were designed following a traditional chemotherapeutic regimen, in which the maximum tolerated dose was delivered, followed by an intervening recovery period. Pharmacokinetic data from these clinical trials (Arteaga et al., 1989) reveals that between doses, the plasma drug concentrations dropped below the efficacy threshold seen in our preclinical studies, and investigators in the clinical studies speculated that maintaining effective drug concentrations over a longer period of time might be required to achieve clinical efficacy (Braakhuis et al., 1990).