Extrusion process to retain resistant starch in a pet food for the purpose of altering colonic fermentation end products that benefit dog health

Abstract

Starch that escapes enzymatic digestion in the small intestine is considered resistant (RS). This RS can be fermented by saccharolytic bacteria in the colon and may benefit gut health. Raw crystalline starch granules contain RS type II. Mild extrusion of pet foods may retain some RS II or develop RS type III (retrograded) that may indirectly benefit animal health. The objectives of the present work were to determine the amount of RS present in typical commercial pet foods, develop a response surface model that predicts RS to maximize RS in corn based extruded kibble, produce dog kibbles with three levels of RS based on model criteria, and determine the effects of these diets on dog colonic health, markers of satiety, metabolomics of serum and feces, and fecal microbiome. In commercial dog and cat foods which were either grain-free or grain based (20 total diets) from a sampling frame of 654 pet foods had low RS (average < 0.6% of the kibble) and did not differ among groups. For the second part of the experiment, a single nutritionally complete diet for adult dogs at maintenance (AAFCO, 2019) was formulated with corn as the sole starch ingredient (65%). Experimental diets were produced on a small-scale intermeshing co-rotating twin-screw extruder following a central composite design with 6 central points (replicates) and 20 total samples per variable. There were three factors at three levels: corn particle size (PS), extruder shaft speed (SS) and in-barrel moisture (IBM). Starch transformations were determined as RS, starch cook, and by rapid-visco analysis (RVA). Results indicated that an increased IBM and decreased SS during extrusion process favored RS retention in the kibble, with a final model of RS= 0.018*PS + 0.00161*SS + 0.0601*IBM – 0.000013*SS*PS – 3.78. Experimental foods to test in vivo with dogs were produced on a single screw extruder in a completely randomized design with 3 replicates per treatment. Starch transformations were measured as starch cook (glucoamylase procedure), RS, rapidly (RDS), slowly (SDS) and total digestible starch (TDS), and RVA. Resistant starch, starch cook and raw:cooked starch RVA AUC increased linearly from high (HS) to medium (MS) and low shear (LS) foods, while SDS was greater in the LS treatment. These foods were fed to 24 dogs in a 3x3 William’s Latin square design with blood and fecal samples collected at the end of each period (28 d). Microbiome was determined by high throughput targeted sequencing of 16S bacterial rRNA. Short-Chain fatty acids, serum metabolomics, and fecal microbiome were determined at a commercial laboratory (Metabolon, Morrisville, NC). Satiety hormones were measured on plasma by ELISA. Immunological markers were measured on feces and serum by a commercial laboratory (MD Biosciences; Oakdale, MN). Data were analyzed as a mixed model with diet, period, treatment sequence and carryover as fixed effects. Fecal quality was determined on a subjective consistency scale. Microbiome alpha diversity was unchanged (P > 0.05) by dietary treatments, but dogs fed the MS and LS foods had more indication of saccharolysis, promoted by a higher fecal glucose and oligosaccharides concentration (P < 0.05, q < 0.1). Fecal butyrate concentration increased in dogs fed the LS diet relative to HS, and MS was similar to both extremes (P < 0.05). Diets had no effect on satiety hormones or local immunity, which might suggest the need for a longer feeding period and(or) a diet produced with lower thermomechanical energy for a dietary effect. In conclusion, we suggest a low to mild shear extrusion process to produce corn-based kibbles that may promote gut health.