Exercise suppresses tumor growth independent of high fat food intake and associated immune dysfunction

Abstract

Epidemiological data suggest that exercise training protects from cancer independent of BMI. Here, we aimed to elucidate mechanisms involved in voluntary wheel running-dependent control of tumor growth across chow and high-fat diets. Access to running wheels decreased tumor growth in B16F10 tumor-bearing on chow (- 50%) or high-fat diets (- 75%, p < 0.001), however, tumor growth was augmented in high-fat fed mice (+ 53%, p < 0.001). Tumor growth correlated with serum glucose (p < 0.01), leptin (p < 0.01), and ghrelin levels (p < 0.01), but not with serum insulin levels. Voluntary wheel running increased immune recognition of tumors as determined by microarray analysis and gene expression analysis of markers of macrophages, NK and T cells, but the induction of markers of macrophages and NK cells was attenuated with high-fat feeding. Moreover, we found that the regulator of innate immunity, ZBP1, was induced by wheel running, attenuated by high-fat feeding and associated with innate immune recognition in the B16F10 tumors. We observed no effects of ZBP1 on cell cycle arrest, or exercise-regulated necrosis in the tumors of running mice. Taken together, our data support epidemiological findings showing that exercise suppresses tumor growth independent of BMI, however, our data suggest that high-fat feeding attenuates exercise-mediated immune recognition of tumors.

Figure 1

Illustrations of the two used study designs. (A) In study design 1, C57bl/6 mice were randomized to cages containing running wheels for 4 weeks prior to tumor inoculation and during tumor challenge (EX) or no running wheel (CON), and further randomized to chow or high-fat feeding (HF) (n = 12). (B) In a parallel study (study design 2), mice were randomized to cages containing running wheels for 4 weeks prior to tumor inoculation and during tumor challenge (4 + 2), 4 weeks just prior to tumor inoculation (4 + 0), 2 weeks during tumor challenge (0 + 2), or no running wheel (0 + 0) (n = 12), while either receiving a HF or standard chow diet.

Figure 2

Effect of wheel running on B16 melanoma growth in mice fed chow or high-fat diet. C57bl/6 mice were randomized to cages containing running wheels for 4 weeks prior to tumor inoculation and during tumor challenge (EX) or no running wheel (CON), and further randomized to chow or high-fat feeding (HF) (n = 12, study design 1). In these mice, (A) tumor volume, (B) body weight, (C) running distance, and (D) weight of the heart were determined at termination, while (E) the average daily food intake was monitored both prior and after tumor induction. In a parallel study (study design 2), mice were randomized to cages containing running wheels for 4 weeks prior to tumor inoculation and during tumor challenge (4 + 2), 4 weeks just prior to tumor inoculation (4 + 0), 2 weeks during tumor challenge (0 + 2), or no running wheel (0 + 0) (n = 12). In these mice, (F) tumor volume, (G) change in body weight, (H) association between tumor volume and change in body weight, (I) the average daily food intake (including the period both prior and after tumor induction) and (J) percentage of necrotic area in B16 tumors from (2F) were determined. Statistical significance was determined by 2-way ANOVA with Tukey’s post hoc test (A, B, D, E, F, G, I, J) or linear regression analysis (H). P values for diet and EX describe the results of the 2-way ANOVA, while $ indicates statistical significance difference between Chow (0 + 0) and HF (0 + 0) at p < 0.05 and individual stars indicate difference from control (CON/0 + 0) group within each feeding group in the post hoc tests. **p < 0.01, ***p < 0.001.

Figure 3

Tumor growth is correlated with systemic regulation of glucose, leptin and ghrelin. From mice in study design 1, serum concentrations and association to tumor volume of (A) insulin, (B) glucose, (C) leptin and (D) ghrelin were determined. Statistical significance was determined by 2-way ANOVA with Tukey’s post hoc test and linear regression analyses. P values for diet and EX describe the results of the 2-way ANOVA, while individual stars indicate difference from control (CON) group within each diet group. **p < 0.01, ***p < 0.001.

Figure 4

High-fat feeding suppresses exercise-mediated innate immune recognition. Expression of selected markers for different immune subsets by microarray analysis in C57bl/6 mice that were randomized to cages containing running wheels for 4 weeks prior to tumor inoculation and during tumor challenge (EX) or no running wheel (CON), and further randomized to chow or high-fat feeding (HF) (n = 5, study design 1) (A). Gene expression by PCR analysis of (B) markers of myeloid cells/macrophages and NK cells, and (C) markers and cytolytic enzymes of T cells (n = 9–12). Statistical significance was determined by 2-way ANOVA with Tukey’s post hoc test for each gene. Individual stars indicate difference from the control group. *p < 0.05, **p < 0.01.

Figure 5

ZBP1 is induced by wheel running and associated with innate immunity. Expression of (A) ZBP1 and (B) ZBP1-related genes in B16 tumors from mice randomized to control (CON) or 6 weeks of wheel running (EX), and further randomized to chow or high-fat feeding (0) (n = 9–12, study design 1). (C) Association between intratumoral ZBP1 expression and markers of innate immune cells, i.e. CD68 and NKG2D. (D) Verification of ZBP1 knock-down (siZBP1) after siRNA transfection in B16 cells. (E) High-content imaging of cell cycle distribution in ZBP1 KO B16 cells obtained with quantitative image-based cytometry (QIBC). The intensity of the red color depicts the levels of γH2AX expression. (F) Quantification of the γH2AX impaired expression upon ZBP1 silencing. Statistical significance was determined by 2-way ANOVA with Tukey’s post hoc test, multiple t-testing, linear regression analyses, and Mann–Whitney tests. Individual stars indicate difference from the CON chow group. *p < 0.05, **p < 0.01, ****p < 0.0001.