Metabolic Complexity of Aggressive Primary and Metastatic Brain Tumors Beyond the Warburg Effect: What Substrates are Fueling the Citric Acid Cycle?

Background Glioblastomas and brain metastases are aggressive, highly proliferative tumors with short survival times. 13 C‐NMR analysis of resected tumors from brain tumor patients infused with 13 C‐glucose at the time of surgery established that oxidation of glucose was occurring simultaneously with high lactate production. While it was clear that the citric acid cycle was actively metabolizing glucose, there were carbons in the acetyl‐CoA pool that could not be attributed to infused 13 C‐glucose. The tumors were metabolizing another substrate(s), identified as the 'substrate gap' (Maher et al., 2012). Since astrocytes, the presumptive cell of origin of glioblastomas, use acetate as a major energy substrate, we investigated whether the tumors were capable of oxidizing acetate. Brain metastases, tumors derived from organs that do not normally oxidize acetate were studied to determine whether altered substrate choice is cell type dependent or a property of brain tumors. Methods : For these studies we used our human orthotopic tumor mouse models of glioblastoma and brain metastases from non‐small cell lung cancer, breast cancer, renal cell carcinoma, melanoma and endometrial cancer, which have remarkable fidelity to the human tumor from which they were derived. The orthotopic tumors are passaged in the brain, without adaptation to culture. We co‐infused [1,2‐ 13 C]acetate and [1,6‐ 13 C]glucose for 2 hours in tumor‐bearing mice and then analyzed the resected tumor by 13 C‐NMR. In addition, we infused [1,2‐ 13 C]acetate in patients during surgery after obtained informed consent on an IRB‐approved protocol. The 13 C‐NMR results from the resected tumors were compared to the prior human studies of infused 13 C‐glucose and to the results in the orthotopic mice. Results : The orthotopic tumors from the glioblastomas and brain metastases all have the capacity to oxidize acetate and do so simultaneously with glucose oxidation. We have validated this finding in patients with glioblastoma and brain metastases. Conclusions Together the data demonstrate a strikingly common metabolic phenotype in these diverse brain tumors that includes the ability to oxidize acetate in the citric acid cycle. It is unclear at present whether this represents an adaptation to meet the high bioenergetic demands of growth in the brain, is a general characteristic of metastatic tumors, or a basic feature of malignant transformation. Evaluation of human primary tumors in vivo will help to address this question.