Sustained Glucose Turnover Flux Distinguishes Cancer Cachexia from Nutrient Limitation

Abstract

Cancer cachexia is an involuntary weight loss condition characterized by systemic metabolic disorder. A comprehensive flux characterization of this condition however is lacking. Here, we systematically isotope traced eight major circulating nutrients in mice bearing cachectic C26 tumors (cxC26) and food intake-matched mice bearing non-cachectic C26 tumors (ncxC26). We found no difference in whole-body lipolysis and proteolysis, ketogenesis, or fatty acid and ketone oxidation by tissues between the two groups. In contrast, compared to ncxC26 mice ad libitum, glucose turnover flux decreased in food intake-controlled ncxC26 mice but not in cxC26 mice. Similarly, sustained glucose turnover flux was observed in two autochthonous cancer cachexia models despite reduced food intake. We identified glutamine and alanine as responsible for sustained glucose production and tissues with altered use of glucose and lactate in cxC26 mice. We provide a comprehensive view of metabolic alterations in cancer cachexia revealing those distinct from decreased nutrient intake.

Figure 1.. Whole-body lipolysis and proteolysis were not elevated in cachectic mice.

(A) Daily body weight of cxC26 and ncxC26 mice under isocaloric feeding (ICa) or ad libitum (AL). (n= 7–10) (B) Illustration of experimental design for isotope tracing. cxC26 and ncxC26 mice were fed equal amount of food from day 8 to day 12 post tumor implantation (Isocaloric feeding; ICa), based on the daily food intake of cxC26 mice ad libitum (see Figure S1b). Isotope tracer infusion was performed at pre-cachectic (day 8) and cachectic (day 12) stages. (C) Illustration of the concept of turnover flux and its calculation from tracer infusion rate (I) and isotope-labeled fraction of traced nutrient in the circulation (L). (D) Circulating palmitate pool and turnover flux in cxC26 and ncxC26 mice under isocaloric feeding at day 8 and day 12. (n= 7–10) (E) Circulating glycerol pool and turnover flux in cxC26 and ncxC26 mice under isocaloric feeding at day 8 and day 12. (n= 6) (F) Illustration of using normalized labeling of TCA intermediates such as malate by a circulating nutrient as the nutrient’s contribution to TCA cycle. (G) Normalized labeling of malate in different tissues under ¹³C-palmitate infusion at day 12. (n= 7–9) (H) Circulating valine pool and turnover flux in cxC26 and ncxC26 mice under isocaloric feeding at day 8 and day 12. (n= 5–6) (I) Normalized labeling of malate in pancreas under ¹³C-valine infusion at day 12. (n= 5–6) Data are shown as mean ± s.d. Significance of the differences: (D,E, H and I) ns: non-significance, *P < 0.05, ** P < 0.01, *** P < 0.001 between groups by two-way ANOVA or (G) no symbol: not significant, *FDR < 0.05, ** FDR < 0.01, *** FDR < 0.001 between groups by two-tailed t-test with multiple corrections.

Figure 2.. Ketogenesis was elevated in cachectic mice, but only due to reduced food intake.

(A) Circulating 3-hydroxybutyrate (3-HB) pool and turnover flux in cxC26 and ncxC26 mice under isocaloric feeding (ICa) at day 8 and day 12 post tumor implantation. (n= 6–9) (B) Normalized labeling of malate in tissues under ¹³C-3-HB infusion at day 12. (n= 6–7) Data are shown as mean ± s.d. Significance of the differences: (A) ns: non-significance, *P < 0.05, ** P < 0.01, *** P < 0.001 between groups by two-way ANOVA or (B) no symbol: not significant, *FDR < 0.05, ** FDR < 0.01, *** FDR < 0.001 between groups by two-tailed t-test with multiple corrections.

Figure 3:. Glucose turnover flux was sustained in cachectic mice while it was decreased in food intake-controlled non-cachectic mice.

(A) Circulating glucose pool and turnover flux in cxC26 and ncxC26 mice under isocaloric feeding (ICa), and ncxC26 mice ad libitum (AL), at day 8 and day 12 post tumor implantation. (n= 8–12) (B, C) Circulating glucose pool and turnover flux in cachectic (B) KP and (C) KL mice compared to their respective non-tumor-bearing control mice (NTB). (n= 8–10) (D) Heatmap showing normalized labeling of significantly changed metabolites in tissues and tumors under ¹³C-glucose infusion at day 12. Metabolites were clustered by metabolic pathways. Color indicates magnitude of normalized labeling (or contribution from glucose). A significantly changed metabolite was defined by an FDR < 0.05 in at least one tissue between cxC26 ICa and ncxC26 ICa, with a normalized labeling greater than 1% in that tissue. Filled entries for a metabolite in a tissue indicate significant change of the metabolite’s labeling across the three mouse groups in that tissue. (n= 8–11) (E) Normalized labeling of malate in tissues and tumors under ¹³C-glucose infusion at day 12. Tissues with significant differences between cxC26 ICa and ncxC26 ICa are highlighted in red. (n= 8–11) Data are shown as mean ± s.d. Significance of the differences: (A-C) ns: non-significance, *P < 0.05, ** P < 0.01, *** P < 0.001 between groups by two-way ANOVA or (D, E) no symbol: not significant, *FDR < 0.05, ** FDR < 0.01, *** FDR < 0.001 between groups by two-tailed t-test with multiple corrections.

Figure 4:. Lactate turnover flux was not altered in cachectic mice but lactate oxidation was increased in specific tissues.

(A) Circulating lactate pool and turnover flux in cxC26 and ncxC26 mice under isocaloric feeding at day 8 and day 12. Significance of the differences: ns: non-significance, *P < 0.05, ** P < 0.01, *** P < 0.001 between groups by two-way ANOVA. (B) Normalized labeling of malate in tissues and tumors under ¹³C-lactate infusion at day 12. Significance of the differences: no symbol: not significant, *FDR < 0.05, ** FDR < 0.01, *** FDR < 0.001 between groups by two-tailed t-test with multiple corrections. “P” represents raw p-value (P < 0.05) without multiple correction between groups. Data are shown as mean ± s.d. (n= 6–14)

Figure 5.. Glutamine and alanine turnover fluxes were elevated in cachectic mice.

(A) Circulating glutamine pool and turnover flux in cxC26 and ncxC26 mice at day 8 and day 12 post tumor implantation. (B) Circulating alanine pool and turnover flux in cxC26 and ncxC26 mice at day 8 and day 12. (C and D) Normalized labeling of malate under (C) ¹³C-glutamine infusion and (D) ¹³C-alanine infusion in cxC26 and ncxC26 mice at day 12. Data are shown as mean ± s.d. Significance of the differences: (A, B) ns: non-significance, *P < 0.05, ** P < 0.01, *** P < 0.001 between groups by two-way ANOVA or (C, D) no symbol: not significant, * FDR < 0.05, ** FDR < 0.01, *** FDR < 0.001 between groups by two-tailed t-test with multiple corrections. All data was obtained for mice under isocaloric feeding (ICa). (n= 5–6)

Figure 6.. Increased contribution from glutamine and alanine to glucose production and elevated glucose-alanine cycling flux in cachectic mice.

(A) Illustration of direct contribution flux (or production flux) to a circulating nutrient from other circulating nutrients. (B) Production flux to circulating glucose from each of other gluconeogenesis substrates in cxC26 and ncxC26 mice at day 12 post tumor implantation. (C) Stacked plots of production flux to each of the circulating nutrients from other circulating nutrients in cxC26 and ncxC26 mice at day 8 and day 12. (D) Volcano plot showing significantly changed interconversion fluxes between circulating nutrients in cxC26 mice compared to ncxC26 at day 12. (E) Interconversion fluxes between circulating nutrients and nutrient storages in cxC26 and ncxC26 mice at day 12. Significantly increased and decreased fluxes are highlighted with red and blue arrows, respectively. Data are shown as mean ± s.e.m. Significance of the differences: no symbol: not significant, *FDR < 0.05, ** FDR < 0.01, ** FDR < 0.001 between groups by two-tailed t-test with multiple corrections. All data was obtained for mice under isocaloric feeding (ICa). (n= 5–11).

Figure 7.. Fuel preference was altered in specific tissues in cachectic mice.

(A) Illustration of direction contribution of a circulating nutrient to tissue TCA cycle. (B) Stacked bar plot of fuel preference (direction contribution of circulating nutrients to malate) of tissues and tumors in cxC26 and ncxC26 mice under isocaloric feeding (ICa), at day 12 post tumor implantation. (C) Volcano plot of the same data as in (A). All data are shown as the mean ± sem. (n= 5–11)