Epidermal growth factor receptor is required for colonic tumor promotion by dietary fat in the azoxymethane/dextran sulfate sodium model: roles of transforming growth factor-{alpha} and PTGS2

Abstract

Purpose: Colon cancer is a major cause of cancer deaths. Dietary factors contribute substantially to the risk of this malignancy. Western-style diets promote development of azoxymethane-induced colon cancer. Although we showed that epidermal growth factor receptors (EGFR) controlled azoxymethane tumorigenesis in standard fat conditions, the role of EGFR in tumor promotion by high dietary fat has not been examined.

Experimental design: A/J x C57BL6/J mice with wild-type Egfr (Egfr(wt)) or loss-of-function waved-2 Egfr (Egfr(wa2)) received azoxymethane followed by standard (5% fat) or western-style (20% fat) diet. As F(1) mice were resistant to azoxymethane, we treated mice with azoxymethane followed by one cycle of inflammation-inducing dextran sulfate sodium to induce tumorigenesis. Mice were sacrificed 12 weeks after dextran sulfate sodium. Tumors were graded for histology and assessed for EGFR ligands and proto-oncogenes by immunostaining, Western blotting, and real-time PCR.

Results: Egfr(wt) mice gained significantly more weight and had exaggerated insulin resistance compared with Egfr(wa2) mice on high-fat diet. Dietary fat promoted tumor incidence (71.2% versus 36.7%; P < 0.05) and cancer incidence (43.9% versus 16.7%; P < 0.05) only in Egfr(wt) mice. The lipid-rich diet also significantly increased tumor and cancer multiplicity only in Egfr(wt) mice. In tumors, dietary fat and Egfr(wt) upregulated transforming growth factor-alpha, amphiregulin, CTNNB1, MYC, and CCND1, whereas PTGS2 was only increased in Egfr(wt) mice and further upregulated by dietary fat. Notably, dietary fat increased transforming growth factor-alpha in normal colon.

Conclusions: EGFR is required for dietary fat-induced weight gain and tumor promotion. EGFR-dependent increases in receptor ligands and PTGS2 likely drive diet-related tumor promotion.

Fig. 1. Egfr^wa2 mutation prevents dietary fat-induced weight gain

Following AOM treatment mice were started on standard (Std, 5% fat) or high fat diet (20% fat) and weighed weekly. Shown are monthly average weights for the indicated genotype and diet normalized to the first month weight. Within 4 months of diet initiation weights were stable. Saline-treated control groups, matched for genotype and diet, gained slightly more weight than AOM treated animals, but then closely paralleled AOM treated groups for the remainder of the study with no significant differences. *p<0.05 compared to age-matched Egfr^wt animals on standard fat diet.

Fig. 2. CTNNB1 and MYC expression levels in colonic tumors are controlled by Egfr genotype and diet

Colonic tumors were immunostained and Western blotted as described in “Materials and Methods”. Shown are representative tumors from each group. Upper panel: CTNNB1 IHC. A. Egfr^wt, std fat; B. Egfr^wt, high fat; C. Egfr^wa2, std fat; D. Egfr^wa2, high fat. Images are 20× and insets 100×. CTNNB1 Western blot. Proteins from colonic tumors (T) and control colons (N) from animals on standard fat (Std fat) or high fat diets were Western blotted for CTNNB1. Densitometry units were expressed as fold-control matched for Egfr genotype and diet. In Egfr^wt animals, CTNNB1 levels were significantly higher in tumors compared to control under both Std fat (1.4±0.1-fold, p<0.05) and high fat conditions (1.6±-0.2 fold, p<0.05). In animals with Egfr^wa2 on a high fat diet, CTNNB1 was 3.8±0.8-fold higher in tumors compared to control (p<0.05). Lower panel: MYC IHC. A. Egfr^wt, std fat; B. Egfr^wt, high fat; C. Egfr^wa2, std fat; D. Egfr^wa2, high fat. Images are 20× and insets 100×. MYC Western blot. In Egfr^wt animals under std fat and high fat conditions, MYC in tumors was 8.7±1.9-fold (p<0.05) and 2.5±0.5 fold of control (p<0.05), respectively. In tumors from Egfr^wa2 animals, MYC was significantly increased 3.8±0.8-fold control (p<0.05) in high fat conditions. Note that CTNNB1 and MYC levels were controlled by Egfr genotype and dietary fat. Under high fat conditions CTNNB1 and MYC were expressed predominantly in malignant epithelial cells with both cytoplasmic and nuclear distributions. Fold-increases in CTNNB1 and MYC were higher in tumors from Egfr^wa2 mice on high fat compared to Egfr^wt mice since the normalizing control mucosal levels were lower. Expression levels of these proto-oncogenes, however, were higher in tumors from Egfr^wt mice compared to Egfr^wa2 mice.

Fig. 3. CCND1 and PTGS2 expression levels in colonic tumors are controlled by Egfr genotype and diet

Colonic tumors were immunostained and Western blotted as described in “Materials and Methods”. Shown are representative tumors from each group. Upper panel: CCND1 IHC. A. Egfr^wt, std fat; B. Egfr^wt, high fat; C. Egfr^wa2, std fat; D. Egfr^wa2, high fat. Images are 20× and insets 100×. CCND1 Western blot. Proteins from colonic tumors (T) and control colons (N) from animals on standard fat (Std fat) or high fat diets were Western blotted for CCND1. Densitometry units were expressed as fold-control matched for Egfr genotype and diet. CCND1 levels were significantly higher in tumors compared to controls under high fat conditions in both Egfr^wt animals (5.5±-1.4 fold control, p<0.05) and Egfr^wa2 animals (10.0±0.5.4-fold control, p<0.05). Note that while fold-increase in tumor CCND1 was higher in Egfr^wa2 mice on high fat compared to Egfr^wt mice since the normalizing control mucosal levels was lower, CCND1 expression levels were much higher in tumors from Egfr^wt mice. Lower panel: PTGS2 IHC. A. Egfr^wt, std fat; B. Egfr^wt, high fat; C. Egfr^wa2, std fat; D. Egfr^wa2, high fat. Images are 20× and insets 100×. PTGS2 was increased in Egfr^wt animals on a high fat diet and was predominantly expressed in stromal cells (in lower panel, compare B to A). PTGS2 Western blot. In Egfr^wt animals, PTGS2 levels were significantly higher in tumors compared to control under high fat conditions (21.8±-2.8 fold, p<0.005).