Investigation of chloroform-methanol soluble wheat proteins and sphingolipids as potential dietary triggers of diabetes in BBdp rats

Abstract

The present study is a follow-up to the study by Coleman (1990), which found that the incidence of diabetes reduced in non-obese diabetic (NOD) mice when chloroform-methanol soluble substances were removed from a cereal-based diet. We used a 2:1 chloroform–methanol (C–M) mixture to extract C–M soluble sphingolipids and proteins from wheat gluten. The C–M extract was subjected to silica gel chromatography and saponification to remove triglycerides, glycolipids and phospholipids. This sphingolipid-enriched fraction from wheat gluten was subsequently incorporated into a BioBreeding diabetes-prone (BBdp) rat diet. Five diets were fed over a 120-day period to BBdp rats: a negative control hydrolysed casein (HC) diet, a positive control National Toxicology Program 2000 (NTP 2000) diet, a wheat gluten (WG)-based diet, a WGGSLF diet containing the WG residue remaining after chloroform-methanol extraction, and a hydrolysed casein plus sphingolipid-enriched fraction (HCGSL) diet. There was a significant increase in diabetes incidence in rats fed the HC + wheat sphingolipid diet versus the HC diet at days 70 and 80 of the feeding study, but not at the end of the study (120 days). There were no significant differences in diabetes incidence in rats fed the WG diet and WG diet with sphingolipids removed. There were no significant differences in pancreatic insulitis scores or in the numbers of jejunal CD4+ and γδTCR + T cells in BBdp rats due to the dietary addition or absence of CM soluble substances from wheat gluten. Sera IFN-γ levels were significantly higher in the sphingolipid-removed WG diet group compared to the BBdp rats in the other four diet groups. Overall, the results of this study provide some evidence for a slight promotional effect of C–M soluble sphingolipids and proteins in wheat gluten on the incidence of diabetes and up-regulation of immune responses in the BBdp rat.