Influence of a ketogenic diet, fish-oil, and calorie restriction on plasma metabolites and lipids in C57BL/6J mice

Abstract

Background: Diet therapies including calorie restriction, ketogenic diets, and fish-oil supplementation have been used to improve health and to treat a variety of neurological and non-neurological diseases.

Methods: We investigated the effects of three diets on circulating plasma metabolites (glucose and β-hydroxybutyrate), hormones (insulin and adiponectin), and lipids over a 32-day period in C57BL/6J mice. The diets evaluated included a standard rodent diet (SD), a ketogenic diet (KD), and a standard rodent diet supplemented with fish-oil (FO). Each diet was administered in either unrestricted (UR) or restricted (R) amounts to reduce body weight by 20%.

Results: The KD-UR increased body weight and glucose levels and promoted a hyperlipidemic profile, whereas the FO-UR decreased body weight and glucose levels and promoted a normolipidemic profile, compared to the SD-UR. When administered in restricted amounts, all three diets produced a similar plasma metabolite profile, which included decreased glucose levels and a normolipidemic profile. Linear regression analysis showed that circulating glucose most strongly predicted body weight and triglyceride levels, whereas calorie intake moderately predicted glucose levels and strongly predicted ketone body levels.

Conclusions: These results suggest that biomarkers of health can be improved when diets are consumed in restricted amounts, regardless of macronutrient composition.

Keywords: Calorie restriction; Glucose; Ketogenic diet; Ketones; Omega-3 fatty acids.

Figure 1

Influence of dietary regimen on body weight. Weights are expressed as means ± SEM. Arrow represents initiation of respective diets after a 16 hour fast. SD-UR, standard diet unrestricted; SD-R, standard diet restricted; KD-UR, ketogenic diet unrestricted; KD-R, ketogenic diet restricted; FO-UR, fish-oil supplemented diet unrestricted; FO-R, fish-oil supplemented diet restricted.

Figure 2

HPTLC plate of plasma lipids in mice under Standard Diet (SD), Ketogenic Diet, and Fish-Oil Diet (FO) under both unrestricted (UR) and restricted (R) feeding conditions.FFA, free fatty acids; CE, cholesteryl ester; TG, triglycerides; IS, internal standard (oleyl alcohol); Chol, cholesterol; Cer, ceramide; CB, cerebrosides; PE, phosphatidylethanolamine; PC, phosphatidylcholine; SM, sphingomyelin; LPC, lyso-phosphatidylcholine. Lipids were run on two separate plates (neutral lipids and free fatty acids), as described in the methods. All lipid values are quantified in Table 3, except for ceramides and cerebrosides, since there are only trace amounts present.

Figure 3

Linear regression analyses of body weight, glucose, β-hydroxybutyrate, and triglyceride levels in mice. The strongest predictor (highest coefficient of determination) for each dependent variable is plotted. Irrespective of diet, glucose levels most strongly predict body weight and triglyceride levels, calories most strongly predict glucose and ketone levels, and dietary fat intake most strongly predicts cholesterol levels. SD-UR, standard diet unrestricted; SD-R, standard diet restricted; KD-UR, ketogenic diet unrestricted; KD-R, ketogenic diet restricted; FO-UR, fish-oil supplemented diet unrestricted; FO-R, fish-oil supplemented diet restricted.