Introduction

Warburg hypothesized that tumor cells have defective mitochondria and they heavily depend on glycolysis for their energy needs. It is well-known that normal brain cells metabolize ketone bodies (KBs) under hypoglycemic conditions. It has been postulated that brain tumor cells lack the ability to metabolize KBs. Based on above hypotheses, ketogenic diet (KD) has been proposed for the adjuvant therapy in the treatment of brain tumors [1]. KD, composed of high fat, low carbohydrate, and moderate protein, produces high levels of KBs in the circulation, and it is thought that defective mitochondria would not metabolize KBs and tumor cells would starve to death. Recent studies performed in rat cell lines/glioma models revealed that some tumor cells are capable of oxidizing KBs [2, 3]. These results contradict the hypothesis that brain tumors lack the ability to metabolize KBs. Cerebral ketone body metabolism differs between rodents and humans [4]. In the current study, we are investigating simultaneous oxidation of ketone body (betahydroxy-butyrate, BHB) and glucose in patient-derived glioblastoma multiforme (GBM) cell lines using ¹³C-NMR based isotopomer analysis.

Methods

GBM tumor tissues were collected from patients undergoing craniotomy at the Houston Methodist Hospital following an approved IRB protocol. Cells were extracted from the tumor tissue and were initially grown in Dulbecco’s modified Eagle’s medium (DMEM) incubated at 37°C under humidified air with 5% CO2. Final 3 hours, the cells were grown in the DMEM media containing 4 mM [2,4-¹³C]betahydroxy-butyrate ([2,4-¹³C]BHB) and 6.0 mM [U-¹³C]glucose. After 3 hours, the media was removed, cells were washed with PBS buffer and the cell pellets were snap-frozen in liquid nitrogen. Prior to NMR data collection, the cell pellets were extracted in 5% perchloric acid, extracts were lyophilized and was reconstituted in 180 µL D₂O containing 1.0 mM DSS-d₆ (pH = 7.4). The sample-solution was transferred to a 3-mm NMR tube, and ¹H and ¹H-decoupled ¹³C NMR spectra were collected on a Bruker 600 MHz spectrometer equipped with a 5-mm cryo-probe optimized for direct ¹³C detection.

Results

Figure 1 is a schematic showing metabolism of [2,4-¹³C]BHB and [U-¹³C]glucose in a tumor cell. [2,4-¹³C]BHB is converted to [2,4-¹³C]acetoacetate by the enzyme BHB dehydrogenase, and further metabolized to [2-¹³C]acetyl-CoA. On the other hand, [U-¹³C]glucose is converted to [U-¹³C]pyruvate via glycolysis which generates [1,2-¹³C]acetyl-CoA through pyruvate dehydrogenase. [2-¹³C]acetyl-CoA enters the TCA cycle, condenses with unlabeled oxaloacetate to form [4-¹³C] labeled α-ketoglutarate (α-KG) and glutamate in the first turn of the cycle. After multiple turns of the cycle, carbons 3 and 4 of α-KG and glutamate are ¹³C-labeled. [U-¹³C]glucose-derived [1,2-¹³C]acetyl-CoA, in the first turn of the TCA cycle, forms 4- and 5- ¹³C-labeled α-KG and glutamate. After multiple turns of the cycle, 3, 4, and 5 carbons of α-KG and glutamate are ¹³C-labeled. Figure 2 shows the portion of the ¹³C NMR spectrum of C4 signal of glutamate with various multiplicities. The singlet (S) and the doublet (D34) are due to [4-¹³C]glutamate and [3,4-¹³C]glutamate respectively, which are derived from the metabolism of [2,4-¹³C]BHB. Doublet (D45) and the quartet (Q) are due [4,5-¹³C]glutamate, and [3,4,5-¹³C]glutamate respectively, which are derived from [U-¹³C]glucose. Figure 3 shows a chart of normalized peak areas of S, D34, D45 and Q peaks of C4-glutamate signal. The relative contribution of BHB and glucose to C4 glutamate signal is determined as follows: In C4 glutamate carbon multiplets, the sum of S and D34 is a measure of [2,4-¹³C]BHB, whereas the sum of D45 and Q is a measure of [U-¹³C]glucose. From the ¹³C-isotomer analysis, we found that the relative contributions of [2,4-¹³C]BHB and [U-¹³C]glucose to the C4 glutamate ¹³C fractional enrichment were 42.60% and 57.40% respectively.

Discussion

KDs are under clinical trials [1] for the adjuvant therapy for the treatment of brain tumors. The rationale behind using KDs in the cancer treatment is the inability of tumor mitochondria to oxidize KBs. Ketone body metabolism in rat cell lines/glioma models [2,3] revealed that brain tumor mitochondria are capable of oxidizing KB. Results from the current study using patient-derived tumor cell lines clearly indicate that human GBMs are fully capable of oxidizing KB in the presence of glucose.

Conclusion

Our findings show that human GBM oxidize BHB even under normoglycemic conditions.

Acknowledgements

This study was supported by the Donna and Kenneth R. Peak Foundation, The Kenneth R. Peak Brain and Pituitary Tumor Treatment Center at Houston Methodist Hospital, The Taub Foundation, The John S. Dunn Foundation, The Blanche Green Estate Fund of the Pauline Sterne Wolff Memorial Foundation, The Verelan Foundation, The Houston Methodist Hospital Foundation, The American Brain Tumor Association, and by many brave patients and families who have been impacted by the devastating effects of brain cancers and central nervous system disease. We thank Sophie Lopez for helping with cell culture experiments.