Colorectal cancers utilize glutamine as an anaplerotic substrate of the TCA cycle in vivo

Abstract

Cancer cells in culture rely on glutamine as an anaplerotic substrate to replenish tricarboxylic acid (TCA) cycle intermediates that have been consumed. but it is uncertain whether cancers in vivo depend on glutamine for anaplerosis. Here, following in vivo infusions of [¹³C₅]-glutamine in mice bearing subcutaneous colon cancer xenografts, we showed substantial amounts of infused [¹³C₅]-glutamine enters the TCA cycle in the tumors. Consistent with our prior observation that colorectal cancers (CRCs) with oncogenic mutations in the phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic (PIK3CA) subunit are more dependent on glutamine than CRCs with wild type PIK3CA, labeling from glutamine to most TCA cycle intermediates was higher in PIK3CA-mutant subcutaneous xenograft tumors than in wild type PIK3CA tumors. Moreover, using orthotopic mouse colon tumors estalished from human CRC cells or patient-derived xenografts, we demonstrated substantial amounts of infused [¹³C₅]-glutamine enters the TCA cycle in the tumors and tumors utilize anaplerotic glutamine to a greater extent than adjacent normal colon tissues. Similar results were seen in spontaneous colon tumors arising in genetically engineered mice. Our studies provide compelling evidence CRCs utilizes glutamine to replenish the TCA cycle in vivo, suggesting that targeting glutamine metabolism could be a therapeutic approach for CRCs, especially for PIK3CA-mutant CRCs.

Figure 1

Kinetics of [¹³C₅]-glutamine infusion in plasma. (A) Time course of labeled glutamine in mouse plasma. Mice (n = 4) were infused with [¹³C₅]-glutamine as described in detail in the methods section. Plasma was taken at the indicated times and percentage of [¹³C₅]-glutamine was measured by GC-MS. (B) Glutamine-derived lactate and glucose are negilible. Mice (n = 4) were infused in [¹³C₅]-glutamine for 4 hours. Percentages of labled glutamine, lactate and glucose in plasma are shown.

Figure 2

More [¹³C₅]-glutamine enters the TCA cycle in PIK3CA mutant tumors in subcutanous xenograft models. (A) Schematic diagram of glutamine and its metabolites in the TCA cycle. (B) Schematic diagram of mice bearing subcutanous (subcu) xenograft tumors infused with [¹³C₅]-glutamine. Isogenic HCT116 PIK3CA WT only cells, in which the mutant allele is knocked out, were injected into left flanks of nude mice, whereas HTCT116 PIK3CA mutant only cells, in which the WT allele was knocked out, were injected into the right. Two weeks post-injection, mice (n = 8) bearing similar size tumors in the two flanks were surgically catheterized for [¹³C₅]-glutamine infusion. (C) More glutamine enters the TCA cycle in HTC116 PIK3CA mutant tumors than in the isogenic WT tumors. The indicated metabolite was measured by GC-MS and the percentage of the ¹³C-labeled metabolite in the total pool was calculated. *p < 0.05, the Student’s t test. (D) A significant fraction of glutamine enters the TCA cycle in xenograft tumors. Percentages of total ¹³C-labeled glutamate, succinate, fumarate, malate and citrate are normalized to total ¹³C-labeled glutamine and plotted.

Figure 3

More [¹³C₅]-glutamine enters the TCA cycle in PIK3CA mutant tumors than adjacent normal tissue in orthotopic xenograft models. [¹³C₅]-glutamine tracing in orthotopic xenograft tumors established from WT-only (7 mice) and mutant-only (8 mice) cells. Two pieces (~1 mm³) of subcutaneous xenografts were sutured into the cecum serosa of nude mice. One day after the surgery, mice were infused with [¹³C₅]-glutamine. The M5 enrichment of glutamine and the M4 enrichments of metabolites directly derived from M5 glutamine in WT- and mutant-only tumors are shown in (A). The M5 enrichment of glutamine and the M4 enrichments of metabolites directly derived from M5 glutamine in HCT116 mutant-only tumors and adjacent cecum tissues are shown in (B). Percentages of total ¹³C-labeled metabolited normalized to total ¹³C-labeled glutamine in the mutant-only tumors are plotted in (C). *p < 0.05, **p < 0.01, ***p < 0.001; the Student’s t test.

Figure 4

More [¹³C₅]-glutamine enters the TCA cycle in PIK3CA mutant tumors than adjacent normal tissue in orthotopic patient-derived xenografts. (A,B) [¹³C₅]-glutamine tracing in orthotopic xenograft tumors established from a colon cancer patient-derived xenograft. The M5 enrichment of glutamine and the M4 enrichments of metabolites directly derived from M5 glutamine are shown in (A). Percentages of total ¹³C-labeled metabolited normalized to total ¹³C-labeled glutamine in tumors are plotted in (B). *p < 0.05, **p < 0.01; the Student’s t test.

Figure 5

More [¹³C₅]-glutamine enters the TCA cycle in PIK3CA mutant tumors than adjacent normal tissue in genetically engineered mice. CDX2P-CreER^T2 Apc^flox/+ Kras^LSL-G12D/+ Pik3ca^LSL-E545K/+ mice were treated with tamoxifen to conditionally express the Kras and Pik3ca oncogenes and to delete Apc. Two moths post-treatment, mice were infused with [¹³C₅]-glutamine. Representative image of H & E staining of a colon tumor is shown in (A). Arrow indicates a tumor invaded to the muscle layer. The M5 enrichment of glutamine and the M4 enrichments of metabolites directly derived from M5 glutamine are shown in (B). Percentages of total ¹³C-labeled metabolited normalized to total ¹³C-labeled glutamine in tumors are plotted in (C). *p < 0.05, **p < 0.01, ***p < 0.001; the Student’s t test.