Physical Activity Delays Obesity-Associated Pancreatic Ductal Adenocarcinoma in Mice and Decreases Inflammation

Abstract

Background & aims: Obesity is a risk factor for pancreatic ductal adenocarcinoma (PDAC), a deadly disease with limited preventive strategies. Lifestyle interventions to decrease obesity might prevent obesity-associated PDAC. Here, we examined whether decreasing obesity by increased physical activity (PA) and/or dietary changes would decrease inflammation in humans and prevent PDAC in mice.

Methods: Circulating inflammatory-associated cytokines of overweight and obese subjects before and after a PA intervention were compared. PDAC pre-clinical models were exposed to PA and/or dietary interventions after obesity-associated cancer initiation. Body composition, tumor progression, growth, fibrosis, inflammation, and transcriptomic changes in the adipose tissue were evaluated.

Results: PA decreased the levels of systemic inflammatory cytokines in overweight and obese subjects. PDAC mice on a diet-induced obesity (DIO) and PA intervention, had delayed weight gain, decreased systemic inflammation, lower grade pancreatic intraepithelial neoplasia lesions, reduced PDAC incidence, and increased anti-inflammatory signals in the adipose tissue compared to controls. PA had additional cancer prevention benefits when combined with a non-obesogenic diet after DIO. However, weight loss through PA alone or combined with a dietary intervention did not prevent tumor growth in an orthotopic PDAC model. Adipose-specific targeting of interleukin (IL)-15, an anti-inflammatory cytokine induced by PA in the adipose tissue, slowed PDAC growth.

Conclusions: PA alone or combined with diet-induced weight loss delayed the progression of PDAC and reduced systemic and adipose inflammatory signals. Therefore, obesity management via dietary interventions and/or PA, or modulating weight loss related pathways could prevent obesity-associated PDAC in high-risk obese individuals.

Keywords: Obesity; diet intervention; high-risk PDAC; voluntary wheel running; weight loss.

Figure 1.. PA decreases the expression of pro-inflammatory cytokines in the serum of overweight/obese subjects.

Mean fold change of serum cytokines after the program to before the program. Analyzed with a paired Wilcoxon test.

Figure 2.. PA delays weight gain, inflammation, fibrosis, PanIN progression, and PDAC incidence in an obesity-associated GEMM of PDAC.

(A) 40-day old Kras^G12D and control mice induced with tamoxifen on either a DIO intervention with HFD or a DIO intervention with HFD + PA for 31 days (n=7 control + HFD, n=5 Kras^G12D + HFD, n=6 control + HFD + PA, n=5 Kras^G12D + HFD+PA). (B) Average distance ran (km/day), analyzed via two-way ANOVA. (C) Average BW change over the 31 days, analyzed via two-way ANOVA. Significance displayed for the last day. (D) Average BW change and (E) pancreas weight after the intervention, analyzed with two-way ANOVA. (F) Representative H&E-stained pancreas sections of Kras^G12D mice (original magnification: x20, scale bar =50μm). Pathology scores for (G) inflammation, (H) fibrosis and (I) total pathology scores, based on the average of all H&E slides, analyzed with Mann-Whitney test. (J) Correlation between total pathological score and kilometers for the PA group and Pearson correlation coefficient. (K) Average number of PanIN lesions analyzed with two-way ANOVA (L) Percent of mice that developed PDAC. (M) Fold change serum cytokines of Kras^G12D mice at week 4 of intervention compared to baseline, analyzed with a paired Wilcoxon test. (N) Comparison of fold change serum cytokines at 4 weeks between HFD and HFD+PA groups, only displaying the cytokines with a trend or statistical significance in Figure 2M or had a P ≤ 0.999 at week 4, analyzed via two-way ANOVA.

Figure 3.. Diet-induced weight modulation and/or increased PA intervention reduces body weight, pancreas weight, non-fasting glucose levels and delays PDAC development in an obesity-induced GEMM of PDAC.

(A) 40-day old Kras^G12D mice were induced with tamoxifen for Kras activation and placed on a DIO intervention with a HFD for 33–54 days followed by 50 days with either no intervention (HFD) (control n = 6; Kras^G12D n = 5), CD (control n = 12; Kras^G12D n = 9), or CD+PA (control n = 7; Kras^G12D n = 5). (B) Distance ran by the PA mice, analyzed by a repeated measures two-way ANOVA. (C) BW change over time, analyzed via two-way ANOVA compared to HFD control. Significance displayed for the last day. (D) BW change from day 0 and (E) average pancreas weight, analyzed via two-way ANOVA. (F) Average non-fasting glucose at the end of the intervention, analyzed via two-way ANOVA after log transformation. Pathology score for (G) fibrosis and (H) inflammation and (I) Total pathology score, analyzed via Kruskal-Wallis test. (J) PanIN lesions analyzed via two-way ANOVA compared to the respective PanIN on the HFD control group. (K) Percent of Kras^G12D mice that developed PDAC at the end of each intervention.

Figure 4.. A diet and PA interventions reduce body weight but do not delay tumor growth in an obese orthotopic mouse model of PDAC

(A) 42-day old male C57BL/6J mice (n=10 per group) were placed on a DIO intervention with HFD for nine weeks before starting a 59-day CD and/or PA intervention. Mice were implanted with tumors and resumed their diet intervention. (B) Distance ran in kilometers analyzed via two-way ANOVA after log transformation. (C) BW change over time, analyzed with mixed-effects analysis comparing to HFD control. (D) Average change in BW from day 56 to the end of the intervention analyzed with two-way ANOVA. (E) Average tumor weight, analyzed via two-way ANOVA after log-transformation. (F) Average photon radiance signal from the tumor/week, analyzed after log-transformation via two-way ANOVA.

Figure 5.. PA modulation of the anti-inflammatory pathways in the adipose tissue significantly delays PDAC tumor growth

(A) Heat map showing hierarchical clustering of genes from the adipose tissue of control and Kras^G12D mice after a DIO intervention of a HFD or HFD+PA from Figure 2A. Blue = low expression, red = high expression. (B) Gene ontology displaying top significant pathways modulated based on adipose tissue gene expression due to PA from Figure 2A cohort (C) Fold change adipose tissue gene expression of IL-15ra in control and Kras^G12D mice from Figure 2A, analyzed via Mann-Whitney test. (D) 9-week old Female C57BL/6J mice (n=10/group, 5/cage) received an empty/control or IL-15 adipocyte-targeting rAAV vector and were implanted with tumors 3 weeks later. (E) Fold change adipose tissue IL-15 gene expression at the end of the study, analyzed via Mann-Whitney test. (F) Representative IVIS images and (G) quantified fold change average photon radiance signal of mice with and without IL-15 adipose targeting, analyzed via two-way ANOVA after log transformation.