Glycated bovine serum albumin for use in feeding trials with animal models - In vitro methodology and characterization of a glycated substrate for modifying feed pellets.

This study investigated different methods to produce Nε-carboxymethyl-lysine (CML)-enriched bovine serum albumin (BSA) as alternatives to the classical approach using glyoxylic acid (GA) and sodium cyanoborohydride (NaBH3CN) which results in toxic hydrogen cyanide (HCN). The reaction of GA (6 mmol/L) and NaBH3CN (21 mmol/L) to produce CML remained the most effective with CML yields of 24-35%, followed by 13-24% using 300 mmol/L glyoxal (GO). GA promoted specific modification of lysine to CML, and fewer structural modifications of the BSA molecule compared with GO, as evidenced by fluorescence and proteomic analyses. GO promoted greater arginine modification compared with GA (76 vs 23%). Despite structural changes to BSA with GO, murine fecal clearance of CML was similar to literature values. Hence, BSA glycation with 300 mmol/L glyoxal is a suitable alternative to GA and NaBH3CN for generating CML-enriched protein free of HCN, but a CML-only fortification model remains to be described.

Carbamylation is a competitor of glycation for protein modification in vivo.

AIM: Chronic kidney disease (CKD) and diabetes mellitus are two diseases that accelerate protein molecular ageing through carbamylation and glycation reactions, characterized by the binding of urea-derived isocyanic acid and of sugars on proteins, respectively. These two reactions target the same protein amino groups and, thus, compete with each other. Such competition may arise especially in diabetic patients with nephropathy. This study aimed to evaluate their potential competitive effects in vitro and under conditions reproducing CKD and/or diabetes in vivo. METHODS: Albumin was incubated in vitro with glucose, urea or cyanate. Carbamylation in vivo was enhanced in normal and diabetic (db/db) mice by either subtotal nephrectomy or cyanate consumption. Homocitrulline, carbamylated haemoglobin and furosine were measured by LC-MS/MS, fructosamine by colorimetric assay and HbA1c by immunological assay. RESULTS: Reciprocal inhibition between carbamylation and glycation was observed during albumin incubations in vitro. Besides, 5 weeks after induction of CKD in vivo, plasma homocitrulline concentrations were similar in both diabetic and non-diabetic mice, whereas fructosamine and HbA1c were decreased (-23% and -42%, respectively) in diabetic mice with CKD compared with only diabetic ones. Fructosamine and HbA1c were also decreased in cyanate-spiked water-drinking mice compared with plain water-drinking diabetic mice. CONCLUSION: Carbamylation competes with glycation in vivo, especially under conditions of high glycation. Thus, the classic markers of glycaemic control should be interpreted with caution in diabetic patients with CKD because of this competitive effect.

Fecal excretion of Maillard reaction products and the gut microbiota composition of rats fed with bread crust or bread crumb.

A comparison between the impacts of advanced (Nε-carboxymethyllysine - CML) and terminal (melanoidins) Maillard reaction products from bread on gut microbiota was carried out in this study. Gut microbiota composition as well as fecal excretion of CML from both bread crust and bread crumb, and of melanoidins from bread crust were assessed on a rodent model. Rats were fed with pellets supplemented or not with 13% of bread crust, bread crumb, a fiber-free bread crust model (glucose, starch and gluten heated together) or a fiber-free-melanoidin-free bread model (glucose-starch and gluten heated separately) for four weeks. These model systems were developed to limit the presence of wheat-native dietary fibers such as cellulose, hemicelluloses and lignin. CML and melanoidins in pellets and feces were evaluated by LC/MS-MS and HPLC/fluorescence respectively, and gut microbiota composition was determined by cultivation and molecular approaches. Diets supplemented with crumb or the fiber-free-melanoidin-free model contained respectively 17% and 64% less melanoidins than their respective controls. A higher excretion of melanoidins was observed for rats fed with crust or bread crust model compared to their controls, confirming that melanoidins are in contact with gut microbiota. No impact of diets was observed on Firmicutes, Bacteroidetes and lactic flora. A decrease of enterobacteria was only observed for rats fed with the diet supplemented with the fiber-free bread crust model. Moreover, a significant increase of bifidobacteria numbers in the presence of crust, crumb and both bread models was observed, showing that this bifidogenic effect of bread is not due to the presence of melanoidins or wheat-native dietary fibers.