Abstract
Diet-induced obesity (DIO) in laboratory rodents can serve as a model with which to study the pathophysiology of obesity, but obesogenic diets (high-sugar and/or high-fat) are often poorly characterised and simplistically aimed at inducing metabolic derangements for the purpose of testing the therapeutic capacity of natural products and other bioactive compounds. Consequently, our understanding of the divergent metabolic responses to different obesogenic diet formulations is limited. The aim of the present study was to characterise and compare differences in the metabolic responses induced by low-fat, medium-fat/high-sugar and high-fat diets in rats through multivariate statistical modelling. Young male Wistar rats were randomly assigned to CON (laboratory chow, low-fat), OB1 (high-sugar, medium-fat) or OB2 (high-fat) dietary groups (n = 24 each) for 17 weeks, after which metabolic responses were characterised. Projection-based multivariate analyses (principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA)) were used to explore the associations between measures of body composition and metabolism. Furthermore, we conducted a systematic literature survey to examine reporting trends in rat dietary intervention studies, and to determine how the metabolic responses observed in the present study compared to other recently published studies. The OB1 and OB2 dietary regimens resulted in distinct metabolic profiles, with OB1 characterised by perturbations in insulin homeostasis and adipose tissue secretory function, while OB2 was characterised by altered lipid and liver metabolism. This work therefore confirms, by means of direct comparison, that differences in dietary composition have a profound impact on metabolic and pathophysiological outcomes in rodent models of DIO. However, through our literature survey we demonstrate that dietary composition is not reported in the majority of rat dietary intervention studies, suggesting that the impact of dietary composition is often not considered during study design or data interpretation. This hampers the usefulness of such studies to provide enhanced mechanistic insights into DIO, and also limits the translatability of such studies within the context of human obesity.