Hyperinsulinemia Promotes Esophageal Cancer Development in a Surgically-Induced Duodeno-Esophageal Reflux Murine Model

Abstract

Hyperinsulinemia could have a role in the growing incidence of esophageal adenocarcinoma (EAC) and its pre-cancerous lesion, Barrett's Esophagus, a possible consequence of Gastro-Esophageal Reflux Disease. Obesity is known to mediate esophageal carcinogenesis through different mechanisms including insulin-resistance leading to hyperinsulinemia, which may mediate cancer progression via the insulin/insulin-like growth factor axis. We used the hyperinsulinemic non-obese FVB/N (Friend leukemia virus B strain) MKR (muscle (M)-IGF1R-lysine (K)-arginine (R) mouse model to evaluate the exclusive role of hyperinsulinemia in the pathogenesis of EAC related to duodeno-esophageal reflux. FVB/N wild-type (WT) and MKR mice underwent jejunum-esophageal anastomosis side-to end with the exclusion of the stomach. Thirty weeks after surgery, the esophagus was processed for histological, immunological and insulin/Insulin-like growth factor 1 (IGF1) signal transduction analyses. Most of the WT mice (63.1%) developed dysplasia, whereas most of the MKR mice (74.3%) developed squamous cell and adenosquamous carcinomas, both expressing Human Epidermal growth factor receptor 2 (HER2). Hyperinsulinemia significantly increased esophageal cancer incidence in the presence of duodenal-reflux. Insulin receptor (IR) and IGF1 receptor (IGF1R) were overexpressed in the hyperinsulinemic condition. IGF1R, through ERK1/2 mitogenic pattern activation, seems to be involved in cancer onset. Hyperinsulinemia-induced IGF1R and HER2 up-regulation could also increase the possibility of forming of IGF1R/HER2 heterodimers to support cell growth/proliferation/progression in esophageal carcinogenesis.

Keywords: MKR mouse model; duodenal reflux; esophageal cancer; human epidermal growth factor receptor 2; hyperinsulinemia; insulin-like growth factor 1 receptor.

Conflict of interest statement

The Authors declare no conflict of interest.

Figures

Figure 1

Total body weight of 13-week-old wild-type (WT) and hyperinsulinemic (MKR) mice. Data were expressed as Median (Q1; Q3). F: Female mice; M: Male mice. * Mann-Whitney U-test, p = 3.8 × 10−4; ** Mann-Whitney U-test p = 2.0 × 10−6.

Figure 2

Insulin signaling in operated WT female mice esophageal tissue. Data are represented as Median (Q1; Q3). Data obtained in operated mice affected by dysplasia or cancer were compared to strain-, sex-, and age-matched, non-operated animals (43 weeks). K-W indicates the Kruskal-Wallis test result, used for multiple comparisons among the three groups of mice. The Mann-Whitney U-test was performed to compare data distribution between two groups of mice. After correction a A p-value lower than 0.016 (Bonferroni adjustment) was considered statistically significant: * p < 0.016; ** p < 1.6× 10−3. (A) Total protein expression: the quantity of each total protein considered was expressed in ng/100 mg of total protein extracted by the distal mice esophageal segment. (B) Phosphorylated protein expression: the amount of the six phospho-proteins considered was expressed as phosphorylated Units for each ng of the corresponding total protein amount. Proteins were quantified by Luminex-xMAP Technology on the basis of a standard curve. Beta-tubulin expression was used as the internal loading control.

Figure 3

Insulin signaling in operated WT male mice esophageal tissue. Data are represented as Median (Q1; Q3). Data obtained in operated mice affected by dysplasia or cancer were compared to strain-, sex-, and age-matched, non-operated animals (43 weeks). K-W indicates the Kruskal-Wallis test result, used for multiple comparisons among the three groups of mice. The Mann-Whitney U-test was performed to compare data distribution between two groups of mice. A p-value lower than 0.016 (Bonferroni adjustment) was considered statistically significant: * p < 0.016; ** p < 1.6 × 10−3. (A) Total protein expression: the quantity of each total protein considered was expressed in ng/100 mg of total protein extracted by the distal mice esophageal segment. (B) Phosphorylated protein expression: the amount of the six phospho-proteins considered was expressed as phosphorylated Units for each ng of the corresponding total protein amount. Proteins were quantified by Luminex-xMAP Technology on the basis of a standard curve. Beta-tubulin expression was used as the internal loading control.

Figure 4

Insulin signaling in operated MKR female mice esophageal tissue. Data are represented as Median (Q1; Q3). Data obtained in operated mice affected by dysplasia or cancer were compared to strain-, sex-, and age-matched, non-operated animals (43 weeks). K-W indicates the Kruskal-Wallis test result, used for multiple comparisons among the three groups of mice. Mann-Whitney U-test was performed to compare data distribution between two groups of mice. A p-value lower than 0.016 (Bonferroni adjustment) was considered statistically significant: * p < 0.016; ** p < 1.6 × 10−3. (A) Total protein expression: the quantity of each total protein considered was expressed in ng/100 mg of total protein extracted by the distal mice esophageal segment. (B) Phosphorylated protein expression: the amount of the six phospho-proteins considered was expressed as phosphorylated Units for each ng of the corresponding total protein amount. Proteins were quantified by Luminex-xMAP Technology on the basis of a standard curve. Beta-tubulin expression was used as the internal loading control.

Figure 5

Insulin signaling in operated MKR male mice esophageal tissue. Data are represented as Median (Q1; Q3). Data obtained in operated mice affected by cancer were compared to strain-, sex-, and age-matched, non-operated animals (43 weeks). The Mann-Whitney U-test was performed to compare data distribution between the two groups of mice. A p-value lower than 0.05 was considered statistically significant: * p < 0.05; ** p < 0.005. (A) Total protein expression: the quantity of each total protein considered was expressed in ng/100 mg of total protein extracted by the distal mice esophageal segment. (B) Phosphorylated protein expression: the amount of the six phospho-proteins considered was expressed as phosphorylated Units for each ng of the corresponding total protein amount. Proteins were quantified by Luminex-xMAP Technology on the basis of a standard curve. Beta-tubulin expression was used as the internal loading control.

Figure 6

Representative case of Ki-67 nuclear protein expression on healthy, dysplastic and neoplastic esophageal tissues in WT and MKR mice. Non-operated mice (healthy tissue): WT male and MKR male at 43 weeks of age. Dysplasia: squamous high-grade dysplastic lesions (HGD) in WT female and MKR female. Cancer: Moderately well differentiated ESCC in WT male and MKR male. Original magnification 40×.

Figure 7

Representative case of HER2 overexpression detected by Immunohistochemistry IHC (A) and Western blotting (B). (A) WT (wild-type mouse esophageal tissue). From left to right: 1. Non-operated (healthy tissue) WT female at 43 weeks of age (IHC negative; original magnification 100×). 2. Squamous Dysplasia (LGD/HGD) in WT male (IHC2+; original magnification 200×). 3. ESCC in WT male (IHC3+ on both dysplastic and infiltrative component; original magnification 200×). 4. EASC in WT male (IHC2+ in squamous component and IHC1+ in glandular component; original magnification 200×). MKR (MKR mouse esophageal tissue). From left to right: 1. Non-operated (healthy tissue) MKR female at 43 weeks of age (IHC negative; original magnification 100×). 2. Squamous Dysplasia (HGD) in MKR female (IHC3+; original magnification 200×). 3. ESCC in MKR female (IHC2+; original magnification 200×). 4. EASC in MKR male (IHC3+ in both squamous and glandular component; original magnification 200×). (B) Thirty micrograms of extracted total proteins were separated in SDS-PAGE and transferred to nitrocellulose membranes. Membranes were blotted with anti-HER2 (molecular weight = 185 KDa) and anti-β-actin (molecular weight = 42 KDa) antibodies. Non-operated (healthy tissue) WT and MKR male at 43 weeks of age. Dysplasia: squamous HGD dysplasia in WT and MKR female. ESCC in WT and MKR female. EASC in WT and MKR male. (C) The densitometric intensity of bands is shown in the bar graphs. Data are presented as the mean ± SEM and the relative protein expression is indicated as n-fold of protein amount detected on WT dysplastic tissue which was set equal to 1 arbitrary unit. * p < 0.05; ** p < 0.01.

Figure 8

Hypothesized hyperinsulinemia-induced cancerogenetic mechanisms. Insulin and insulin-like growth factor 1 are closely related hormones that control different aspects of growth and metabolism in many organisms. At physiological levels, insulin and IGF-1 fully activate their cognate receptors and IGF-1 can also activate IR/IGF1R hybrid receptors. At supraphysiological levels they can also bind and activate each other’s’ receptor in cell culture models, although with reduced affinity. Activated IR/IGF1R receptors phosphorylate target proteins such as Insulin Receptor Substrate (IRSs) proteins on selective tyrosine residues (Tyr) that serve as docking sites for downstream effector molecules. Phosphorylated IRS1 activates the metabolic PI3K/Akt pathway. Activated Akt regulates glucose metabolism and induces the mitogenic (ERK/MAPK) pathway involved in carcinogenic mechanism. Serine (Ser)/Threonine phosphorylation of the IRS proteins is a key negative-feedback control mechanism that uncouples the IRS proteins from their upstream and downstream effectors and terminates signal transduction in response to insulin, under physiological conditions. Analogous mechanisms are involved in the molecular pathophysiology of insulin resistance by activating serine/threonine IRS kinases (such as the p70S6K) that phosphorylate the IRS proteins and inhibit their function. Hyperinsulinemia is responsible of receptors hyperactivation inducing IR, IGF1R, and ERK over-expression. HER2 signal activation is known to have a pivotal role in esophageal cancerogenesis and disease progression. Hyperinsulinemia is able to induce also HER2 overexpression. HER2 is the preferential dimerization partner of other members of the ErbB family and other growth factor receptors. A cross-talk between HER2 and IGF1R signaling pathway has been proven in breast cancer through an heterodimerization mechanism. An analogous mechanism is proposed also in this context, in which hyperinsulinemia-induced IGF1R overexpression might enhance the possibility to form HER2/IGF1R heterodimers, promoting mitogenic pathway activation and, in turn, cancer.