Abstract
BACKGROUND: Low-carbohydrate diet (LCD) is effective for weight loss and glycaemic control in humans. Here, the study aimed to explore the effects of LCD/high-fat diet (HFD) in both humans and mice. METHODS: Twenty-two overweight or obese participants received LCD for 3 weeks. Based on carbohydrate intake > 10% or ≤ 10% of calories, the participants were divided into moderate LCD (MLCD) and very LCD (VLCD) groups. The participants completed a 10-question food preference survey. Meanwhile, C57BL/6J mice were assigned to five groups: chow diet (CD, 10% fat), HFD with 60%, 70%, and 75% fat from cocoa butter (HFD-C), and HFD with 60% fat from lard (HFD-L) and fed for 24 weeks. Eight mice were acclimatised for the food-choice test. RESULTS: LCD decreased the total energy intake in humans. The VLCD group showed greater weight loss and better glycaemic control than the MLCD group. A food preference survey showed that 65% of participants tended to choose high-carbohydrate foods. In mice, HFD resulted in energy overconsumption, obesity, and metabolic disorders. When CD and HFD-L were administered simultaneously, mice rarely consumed CD. In the HFD-C groups, the energy intake and body weight increased with increasing dietary fat content. Compared with the HFD-C group, the HFD-L group consumed more energy and had poorer metabolism. CONCLUSIONS: Lower carbohydrate intake contributed to lower energy intake and improved metabolism in humans. In mice, diets with a higher proportion of fat become more attractive and obesogenic by fixing the fat sources. Since the mice preferred lard to cocoa butter, lard induced excess energy intake and poorer metabolism. Different food preferences may be the underlying mechanism behind the opposite effects of the LCD/HFD in humans and mice. TRIAL REGISTRATION: The clinical trial was registered with the Chinese Clinical Trial Registry ( www.chictr.org.cn ). The registration number is ChiCTR1800016786. All participants provided written informed consent prior to enrolment.