Mitochondrial complex I promotes kidney cancer metastasis. Kidney cancers are metabolically dysfunctional, but how this dysfunction affects cancer progression is unclear. Researchers infused 13C-labeled nutrients in over 80 patients with kidney cancer during surgery. Labeling from [U-13C] glucose varied across subtypes, indicating that the kidney environment alone cannot account for all tumor metabolic reprogramming. Clear cell renal cell carcinomas (ccRCCs) showed suppressed labeling of TCA cycle intermediates, indicating that this suppression is tissue intrinsic. [1,2-13C] acetate and [U-13C] glutamine infusions, combined with measurements of respiration in isolated human kidney and tumor mitochondria, revealed lower electron transport chain activity in ccRCCs, contributing to decreased oxidative and enhanced reductive TCA cycle labeling. However, ccRCC metastases had enhanced TCA cycle labeling compared to primary ccRCCs, indicating a divergent metabolic program during metastasis. In mice, stimulating respiration or NADH recycling in kidney cancer cells promoted metastasis, whereas inhibiting the electron transport chain complex I decreased metastasis. These findings suggest that metabolic properties and liabilities evolve during kidney cancer progression, and that mitochondrial function is limiting for metastasis but not growth at the original site. Mitochondrial alterations are common in renal cell carcinomas (RCCs), but the mechanisms vary among subtypes. In ccRCC, approximately 90% of tumors have biallelic inactivation of the gene encoding the tumor suppressor von Hippel–Lindau (VHL). Loss of VHL leads to pseudohypoxic stabilization of HIFα subunits and chronic activation of HIF target genes, many of which promote glycolysis and suppress glucose oxidation. A subset of chromophobe RCCs contain mutations in complex I of the electron transport chain, and oncocytomas accumulate defective mitochondria through somatic mutations in complex I and impaired mitochondrial elimination programs. Pathogenic defects in the metabolic enzymes fumarate hydratase (FH) and succinate dehydrogenase (SDH) are also initiating events in some renal cancers. These data imply that many RCCs select for reduced mitochondrial metabolism during their initiation and growth in the kidney. Despite genetic evidence for mitochondrial dysfunction, how these mutations affect nutrient metabolism in human RCCs in vivo is unknown. Intraoperative infusion of 13C-labeled nutrients and analysis of 13C labeling in metabolites extracted from surgically resected samples can reveal metabolic differences between tumors and adjacent tissue or among tumors from different patients. We previously reported suppressed contribution of glucose carbon to TCA cycle intermediates in five human ccRCCs, implying reduced glucose oxidation in these tumors. Here we studied why this phenotype occurs, whether it generally characterizes primary kidney tumors and how metabolic properties evolve during ccRCC progression to metastatic diseaseMitochondrial complex I promotes kidney cancer metastasis. Kidney cancers are metabolically dysfunctional, but how this dysfunction affects cancer progression is unclear. Researchers infused 13C-labeled nutrients in over 80 patients with kidney cancer during surgery. Labeling from [U-13C] glucose varied across subtypes, indicating that the kidney environment alone cannot account for all tumor metabolic reprogramming. Clear cell renal cell carcinomas (ccRCCs) showed suppressed labeling of TCA cycle intermediates, indicating that this suppression is tissue intrinsic. [1,2-13C] acetate and [U-13C] glutamine infusions, combined with measurements of respiration in isolated human kidney and tumor mitochondria, revealed lower electron transport chain activity in ccRCCs, contributing to decreased oxidative and enhanced reductive TCA cycle labeling. However, ccRCC metastases had enhanced TCA cycle labeling compared to primary ccRCCs, indicating a divergent metabolic program during metastasis. In mice, stimulating respiration or NADH recycling in kidney cancer cells promoted metastasis, whereas inhibiting the electron transport chain complex I decreased metastasis. These findings suggest that metabolic properties and liabilities evolve during kidney cancer progression, and that mitochondrial function is limiting for metastasis but not growth at the original site. Mitochondrial alterations are common in renal cell carcinomas (RCCs), but the mechanisms vary among subtypes. In ccRCC, approximately 90% of tumors have biallelic inactivation of the gene encoding the tumor suppressor von Hippel–Lindau (VHL). Loss of VHL leads to pseudohypoxic stabilization of HIFα subunits and chronic activation of HIF target genes, many of which promote glycolysis and suppress glucose oxidation. A subset of chromophobe RCCs contain mutations in complex I of the electron transport chain, and oncocytomas accumulate defective mitochondria through somatic mutations in complex I and impaired mitochondrial elimination programs. Pathogenic defects in the metabolic enzymes fumarate hydratase (FH) and succinate dehydrogenase (SDH) are also initiating events in some renal cancers. These data imply that many RCCs select for reduced mitochondrial metabolism during their initiation and growth in the kidney. Despite genetic evidence for mitochondrial dysfunction, how these mutations affect nutrient metabolism in human RCCs in vivo is unknown. Intraoperative infusion of 13C-labeled nutrients and analysis of 13C labeling in metabolites extracted from surgically resected samples can reveal metabolic differences between tumors and adjacent tissue or among tumors from different patients. We previously reported suppressed contribution of glucose carbon to TCA cycle intermediates in five human ccRCCs, implying reduced glucose oxidation in these tumors. Here we studied why this phenotype occurs, whether it generally characterizes primary kidney tumors and how metabolic properties evolve during ccRCC progression to metastatic disease