The gut microbiota has been implicated in host nutrient absorption and energy homeostasis. We studied the influence of different diets on body composition in germ-free (GF) and conventional (CV) mice. GF and CV male adult C3H mice were fedad libituma semi-synthetic low-fat diet (LFD; carbohydrate–protein–fat ratio: 41:42:17; 19·8 kJ/g), a high-fat diet (HFD; 41:16:43; 21·4 kJ/g) or a commercial Western diet (WD; 41:19:41; 21·5 kJ/g). There was no difference in body weight gain between GF and CV mice on the LFD. On the HFD, GF mice gained more body weight and body fat than CV mice, and had lower energy expenditure. GF mice on the WD gained significantly less body fat than GF mice on the HFD. GF mice on both HFD and WD showed increased intestinal mRNA expression of fasting-induced adipose factor/angiopoietin-like protein 4 (Fiaf/Angptl4), but they showed no major changes in circulating Fiaf/Angptl4 compared with CV mice. The faecal microbiota composition of the CV mice differed between diets: the proportion of Firmicutes increased on both HFD and WD at the expense of the Bacteroidetes. This increase in the Firmicutes was mainly due to the proliferation of one family within this phylum: the Erysipelotrichaceae. We conclude that the absence of gut microbiota does not provide a general protection from diet-induced obesity, that intestinal production of Fiaf/Angptl4 does not play a causal role in gut microbiota-mediated effects on fat storage and that diet composition affects gut microbial composition to larger extent than previously thought.

Background: Reference standards for resting energy expenditure (REE) are widely used. Current standards are based on measurements made in the first part of the past century in various races and locations. Objective: The aim of the present study was to investigate the application of the World Health Organization (WHO) equations from 1985 in healthy subjects living in a modern, affluent society in Germany and to generate a new formula for predicting REE. Design: The study was a cross-sectional and retrospective analysis of data on REE and body composition obtained from 2528 subjects aged 5–91 y in 7 different centers between 1985 and 2002. Results: Mean REE varied between 5.63 and 8.07 MJ/d in males and between 5.35 and 6.46 MJ/d in females. WHO prediction equations systematically overestimated REE at low REE values but underestimated REE at high REE values. There were significant and independent effects of sex, age, body mass or fat-free mass, and fat mass on REE. Multivariate regression analysis explained up to 75% of the variance in REE. Two prediction formulas including weight, sex, and age or fat-free mass, fat mass, sex, and age, respectively, were generated in a subpopulation and cross-validated in another subpopulation. Significant deviations were still observed for underweight and normal-weight subjects. REE prediction formulas for specific body mass index groups reduced the deviations. The normative data for REE from the Institute of Medicine underestimated our data by 0.3 MJ/d. Conclusions: REE prediction by WHO formulas systematically over- and underestimates REE. REE prediction from a weight group–specific formula is recommended in underweight subjects.

The intestines of obese humans and mice are enriched withErysipelotrichi, a class within theFirmicutes.Clostridium ramosum, a member of theErysipelotrichi, is associated with symptoms of the metabolic syndrome in humans. To clarify the possible obesogenic potential of this bacterial species and to unravel the underlying mechanism, we investigated the role ofC. ramosumin obesity development in gnotobiotic mice. Mice were associated with a simplified human intestinal (SIHUMI) microbiota of eight bacterial species, includingC. ramosum, with the SIHUMI microbiota exceptC. ramosum(SIHUMIw/oCra), or withC. ramosumonly (Cra) and fed a high-fat diet (HFD) or a low-fat diet (LFD). Parameters related to the development of obesity and metabolic diseases were compared. After 4 weeks of HFD feeding, the mouse groups did not differ in energy intake, diet digestibility, gut permeability, and parameters of low-grade inflammation. However, SIHUMI and Cra mice fed the HFD gained significantly more body weight and body fat and displayed higher food efficiency than SIHUMIw/oCra mice fed the HFD. Gene expression of glucose transporter 2 (Glut2) in jejunal mucosa and of fatty acid translocase (CD36) in ileal mucosa was significantly increased in the obese SIHUMI and Cra mice compared with the less obese SIHUMIw/oCra mice. The data demonstrate that the presence ofC. ramosumin SIHUMI and Cra mice enhanced diet-induced obesity. Upregulation of small intestinal glucose and fat transporters in these animals may contribute to their increased body fat deposition.IMPORTANCEObesity is a growing health problem worldwide. Changes in the proportions ofBacteroidetesandFirmicutes, the two dominant phyla in the human and the murine intestinal tract, link the intestinal microbiota to obesity.Erysipelotrichi, a class within theFirmicutes, increase in response to high-fat feeding in mice.Clostridium ramosum, a member of theErysipelotrichi, has been linked to symptoms of the metabolic syndrome. We hypothesized thatC. ramosumpromotes obesity development and related pathologies. Our experiments in gnotobiotic mice show thatC. ramosumpromoted diet-induced obesity, probably by enhancing nutrient absorption. Identification of obesogenic bacteria and understanding their mode of action enable the development of novel strategies for the treatment of this epidemic disease. Pharmaceuticals that target obesogenic bacteria or their metabolism could help to prevent and treat obesity and related disorders in the future.