Transgenic tomatoes expressing the 6F peptide and ezetimibe prevent diet-induced increases of IFN-β and cholesterol 25-hydroxylase in jejunum

Abstract

Feeding LDL receptor (LDLR)-null mice a Western diet (WD) increased the expression of IFN-β in jejunum as determined by quantitative RT-PCR (RT-qPCR), immunohistochemistry (IHC), and ELISA (all P < 0.0001). WD also increased the expression of cholesterol 25-hydroxylase (CH25H) as measured by RT-qPCR (P < 0.0001), IHC (P = 0.0019), and ELISA (P < 0.0001), resulting in increased levels of 25-hydroxycholesterol (25-OHC) in jejunum as determined by LC-MS/MS (P < 0.0001). Adding ezetimibe at 10 mg/kg/day or adding a concentrate of transgenic tomatoes expressing the 6F peptide (Tg6F) at 0.06% by weight of diet substantially ameliorated these changes. Adding either ezetimibe or Tg6F to WD also ameliorated WD-induced changes in plasma lipids, serum amyloid A, and HDL cholesterol. Adding the same doses of ezetimibe and Tg6F together to WD (combined formulation) was generally more efficacious compared with adding either agent alone. Surprisingly, adding ezetimibe during the preparation of Tg6F, but before addition to WD, was more effective than the combined formulation for all parameters measured in jejunum (P = 0.0329 to P < 0.0001). We conclude the following: i) WD induces IFN-β, CH25H, and 25-OHC in jejunum; and ii) Tg6F and ezetimibe partially ameliorate WD-induced inflammation by preventing WD-induced increases in IFN-β, CH25H, and 25-OHC.

Keywords: apolipoprotein A-I mimetic peptides; ather­osclerosis; hyperlipidemia; oxysterols.

Fig. 1.

Addition of either ezetimibe or Tg6F to WD ameliorated dyslipidemia in LDLR-null mice and addition of both to WD (combined formulation) were significantly better than addition of either agent alone. Female LDLR-null mice age 3–6 months (n = 20–28 per group) were fed standard mouse chow or WD or the mice were fed WD plus 0.06% by weight of Tg6F concentrate, which was prepared as described in Materials and Methods; or the mice were fed WD with ezetimibe added to give a daily dose of 10 mg per kg of body weight (WD + ezetimibe); or the mice were fed WD with Tg6F added at 0.06% by weight plus ezetimibe (each added separately to the diet) to give a daily dose of ezetimibe of 10 mg per kg of body weight as described in Materials and Methods for the combined formulation. After feeding the diets for 2 weeks, the mice were bled, and plasma lipid levels were determined as described in Materials and Methods. A: The decrease in plasma cholesterol compared with WD alone for each treatment. B: The decrease in plasma triglycerides compared with WD alone for each treatment. C: The increase in plasma HDL cholesterol compared with WD alone for each treatment. The data shown are mean ± SEM. NS, not significant.

Fig. 2.

Adding Tg6F and ezetimibe to WD using the novel method was more effective than adding the same doses of Tg6F and ezetimibe to WD using the combined formulation. Female LDLR-null mice age 9–12 months (n = 25 per group) were fed standard mouse chow; or WD; or WD with Tg6F added at 0.06% by weight of diet (WD + Tg6F); or WD with ezetimibe added to give a daily dose of 10 mg per kg of body weight (WD + ezetimibe); or Tg6F added at 0.06% by weight of diet plus ezetimibe (each separately mixed into the diet) to give a daily dose of ezetimibe of 10 mg per kg of body weight (WD + combined formulation); or, by using the novel method as described in Materials and Methods, WD containing Tg6F at 0.06% by weight of diet and ezetimibe sufficient to provide the mice with 10 mg/kg/day (WD + novel method). After feeding the diets for 2 weeks, the mice were bled, the jejunums were harvested as described in Materials and Methods, and plasma lipid levels and plasma levels of SAA were determined. A: The decrease in plasma cholesterol compared with WD alone for each treatment. B: The decrease in plasma triglycerides compared with WD alone for each treatment. C: The Increase in plasma HDL cholesterol compared with WD alone for each treatment. D: The decrease in plasma SAA levels compared with WD alone for each treatment. NS, not significant.

Fig. 3.

WD induced increased levels of IFN-β in the enterocytes of the jejunum of the mice described in Fig. 2, which was better ameliorated by the novel method for adding Tg6F and ezetimibe compared with adding the same doses of Tg6F and ezetimibe to WD by the combined formulation. The induction of IFN-β by WD in the enterocytes of the jejunum was determined by IHC and ELISA as described in Materials and Methods. A: An example of IHC showing induction of IFN-β expression by WD and amelioration by the novel method. B: Quantification of IHC was performed as described in Materials and Methods. C: The results of IHC were confirmed by ELISA as described in Materials and Methods. The results shown are mean ± SEM. Abbreviations are the same as in the Fig. 2 legend.

Fig. 4.

WD increased levels of CH25H in the enterocytes of the jejunum of the mice described in Fig. 2, which was better ameliorated by the novel method for adding Tg6F and ezetimibe compared with adding the same doses of Tg6F and ezetimibe to WD by the combined formulation. The induction of CH25H by WD in the enterocytes of the jejunum was determined by IHC and ELISA as described in Materials and Methods. A: An example of IHC showing induction of CH25H expression by WD and amelioration by the novel method. B: Quantification of IHC was performed as described in Materials and Methods. C: The results of IHC were confirmed by ELISA as described in Materials and Methods. The results shown are mean ± SEM. Abbreviations are the same as in the Fig. 2 legend.

Fig. 5.

WD induced increased levels of 25-OHC in the jejunum of the mice described in Fig. 2, which was better ameliorated by the novel method described in Materials and Methods for adding Tg6F and ezetimibe compared with adding the same doses of Tg6F and ezetimibe to WD by the combined formulation. The 25-OHC was determined by LC-MS/MS as described in Materials and Methods. The results shown are mean ± SEM. Abbreviations are the same as in the Fig. 2 legend.

Fig. 6.

Feeding LDLR-null mice WD resulted in increased mRNA levels for IFN-α, IFN-β, and CH25H in the duodenum and jejunum of the mice described in Fig. 2, and adding Tg6F and ezetimibe to WD prevented the increase. IFN-α, IFN-β, and CH25H mRNA levels in the duodenum and jejunum were determined by RT-qPCR as described in Materials and Methods. A: Results for IFN-α are shown. B: Results for IFN-β are shown. C: Results for CH25H are shown. In each case, the left panel shows the results for the duodenum, and the right panel shows the results for the jejunum. The results are mean ± SEM. Abbreviations are the same as in the Fig. 2 legend.

Fig. 7.

Addition of Tg6F or ezetimibe as single agents to chow or adding both via the novel method significantly increases FNS excretion. Female LDLR-null mice were housed four mice per cage, and three to five cages per group were used in each of three separate experiments. The mice were ages 3–4, 9–12, and 14–15 months, respectively, for the three experiments. In each experiment, the mice were switched from chow to WD. After 2 weeks, the mice were switched back to chow containing a nonabsorbable color marker as described in Materials and Methods. Forty-eight hours after the marker appeared in the feces, the mice were continued on chow alone (chow) or Tg6F was added to the chow at 0.06% by weight of diet, or ezetimibe was added to the chow at 10 mg/kg/day, or ezetimibe and Tg6F prepared by the novel method were added to the chow to give Tg6F at 0.06% by weight of diet and ezetimibe 10 mg/kg/day. After 1 week, while still on these diets, the feces were collected from each cage, and FNS was determined as described in Materials and Methods. Shown are mean ± SEM of the pooled data from the three separate experiments.

Fig. 8.

A hypothesis to explain how WD induces inflammation in the jejunum that Tg6F and ezetimibe ameliorate. WD provides cholesterol, triglycerides, and phospholipids [phosphatidylcholine (PC)]. In the lumen of the small intestine, phospholipase A₂ group 1B (PLA2G1B) removes the fatty acid moiety from the sn-2 position of PC, yielding LysoPC. The triglycerides are acted upon by pancreatic lipase to yield two fatty acids and a monoglyceride (data not shown). Also not shown is the action of bile in forming micelles. The enterocytes of the small intestine take up the dietary lipids together with small amounts of ligands for TLRs. This induces IFN-β in the enterocytes, which in turn induces CH25H, which contributes to increased levels of 25-OHC in the enterocytes. Subsequently, 25-OHC enters the lamina propria of the villi of the small intestine, where it acts on immune cells. This leads to increased levels of inflammatory cytokines [e.g., 25-OHC is known to induce increased levels of IL-6, IL-8 (mouse functional homologs MIP2 and KC), and M-CSF], which induce inflammation in the villus of the small intestine. Additionally, IFN-β can induce the expression of autotaxin that increases the conversion of unsaturated LysoPC to unsaturated LPA, which further stimulates inflammation and leads to dyslipidemia as discussed in refs. , , and . Intestinal inflammation leads to systemic inflammation as measured by increased levels of plasma SAA (data not shown). Ezetimibe and Tg6F act to decrease levels of IFN-β, CH25H, and 25-OHC, thus partially ameliorating inflammation and dyslipidemia in LDLR-null mice fed WD.