TABLE 2.

Summary of results from intervention studies assessing energy metabolism and the gut microbiome.

Reference	Population and time	Intervention	Assessments	Main outcomes
Parallel arms randomized controlled trials
Bendtsen et al., 2018 [31]	-People with overweight or obesity (n = 52) -24 wk	-Energy restriction with low dairy diet (<600 mg of calcium/d) or high dairy diet (∼1500 mg of calcium/d, ∼1200 mg from dairy products)	-REE and RER (metabolic cart) -Microbiome (16S rRNA)	-Reduced weight and fat mass, and REE in both groups
				-RER decreased in high dairy (-0.02, P < 0.05) and tended to increase in low dairy (0.02, P < 0.10) (adjusted group effect P = 0.006)
				-Relative abundance of Papillibacter in wk 24 correlated with fat mass loss (rs = 0.61, P = 0.017)
				-Low dairy decreased Veillonella (P = 0.014)
				-High dairy had no significant taxonomic change
Canfora et al., 2017 [32]	-People with overweight or obesity with prediabetes (n = 44) -12 wk	-Fiber (galactooligosaccharides, 5g 3x/d) compared to placebo	-REE and RER (metabolic cart) -Plasma and fecal SCFAs -Microbiome (16S rRNA)	-No alterations in plasma or fecal SCFAs, REE and substrate oxidation (RER)
				-Fiber increased Bifidobacterium (P = 0.009), Prevotella oralis (P = 0.010), Prevotella melaninogenica (P = 0.008), Bacteroides stercoris (P = 0.011) and Sutterella wadsworthia (P = 0.002)
Karl et al., 2017 [33]	-Soldiers under intense training (n = 73) -4 d	-Control diet -Protein-supplemented diet (+ 4 whey protein-based snack bars) -Carbohydrate-supplemented diet (+ 4 carbohydrate-based snack bars)	-TEE (DLW) -Microbiome (16S rRNA) -Plasma and fecal metabolomics -Intestinal permeability (sucralose excretion)	-No differences between diet groups for any of the following outcomes observed: weight loss, higher TEE, increase in intestinal permeability, increase in α-diversity, increased F/B ratio.
				-Decrease in metabolites of amino acid, fatty acid, carbohydrate, and energy metabolism
				-Amino acid and nucleotide metabolites were predictive for microbiome composition
				-Intestinal permeability correlated with pretraining Actinobacteria (rs = -0.53, P = 0.01) Proteobacteria (rs = 0.64, P = 0.002) and Sutterella (rs = 0.68, P = 0.001) relative abundance
Karl et al., 2017 [34]	-Men and postmenopausal women (40–65 y with BMI 20–35, n = 81) -6 wk	-Whole grains-based diet (16 ± 0.8 g of fiber) compared to refined grains-based diet (8 ± 0.4 g of fiber)	-REE (metabolic cart) -Microbiome (16S rRNA) -Energy and SCFAs in stool	-Whole grains increased REE and stool energy content whereas refined–decreased (P = 0.04 and P < 0.001)
				-Refined grains decreased propionate (P = 0.05) and acetate (P = 0.02) compared to whole grains
				-Whole grains tended to decrease Enterobacteriaceae and to increase Lachnospira and Roseburia. No significant associations.
Most et al., 2017 [35]	-People with overweight or obesity (n = 37) -12 wk	-Polyphenols (epigallocatechin-3-gallate 282 mg/d and resveratrol 80 mg/d) compared to placebo	-TEE and fat oxidation (metabolic cart) -Mitochondrial oxidative capacity -Microbiome (RT-PCR)	-Men had higher Bacteroidetes than women at baseline (P < 0.01)
				-Polyphenols decreased Bacteroidetes compared to placebo in men (P = 0.05) but not in women (P = 0.15)
				-Polyphenols increased fat oxidation (fasting P = 0.03, postprandial P = 0.02) and mitochondrial oxidative capacity (P = 0.01)
				-Baseline Bacteroidetes correlated with postprandial fat oxidation in men (r = 0.855, P = 0.01)
Müller et al., 2020 [36]	-People with normal weight or overweight (n = 48) -12 wk	-Fiber (arabinoxylan-oligosaccharides, 15 g/d) compared to placebo	-Substrate oxidation and TEE (metabolic cart) -Microbiome (16S rRNA) -Fecal and plasma SCFAs	-Fiber tended to increase fat oxidation (P = 0.073) whereas placebo tended to decrease it (P = 0.089) (group effect P = 0.008).
				-No effect on TEE and carbohydrate oxidation
				-Fiber reduced α-diversity (P < 0.001)
				-Fiber changed gut microbiome composition (P = 0.05), exemplified by increased abundance of Bifidobacterium, Akkermansia, Prevotellaceae NK3831 group Lactobacillus, and decreased abundance of Blautia, Eubacterium hallii, Coriabacteriaceau and Dorea
				-No differences between fiber and placebo in fecal and plasma SCFAs
Yu et al., 2020 [37]	-People with obesity (n = 22) -12 wk (6 wk receiving intervention + 6 wk follow-up)	-FMT from healthy donors with normal weight to those with obesity, compared to placebo	-REE (metabolic cart) -Microbiome (16S rRNA and metagenomic)	-No change in REE, weight, and body composition
				-Compared to placebo, the microbiome profile of FMT recipients approached the microbiome of donors (P < 0.001) and were less similar to their own baseline microbiome (P < 0.05). Engraftment persisted during the 12 wk.
				-No significant correlations
Crossover randomized controlled trials
Basolo et al., 2020 [38]	-Healthy adults (n = 25) -17 d (4 d run-in + 3 d of intervention + 3 d wash-out + 3 d intervention + 4 d wash-out)	-Overfeeding (OF, 150% of weight-maintaining diet, WMD) or underfeeding (UF, 50% of WMD) with provided diets	-Energy in stool and urine -Microbiome (metagenomic) -Plasma SCFAs and bile acids	-OF excreted more absolute calories in stool (P < 0.001) and urine
				-UF excreted more calories as a % of ingested calories in stool (P < 0.001) and urine (P < 0.001)
				-Plasma butyrate increased in OF (P < 0.001) and decreased in UF (P = 0.007)
				-UF decreased plasma deoxycholic acid (P = 0.012)
				-UF increased total colonization (P = 0.005) and α-diversity (P = 0.004)
				-UF increased Verrucomicrobia (p=0.01)
Jumpertz et al., 2011 [39]	-Individuals with normal weight or obesity (n = 21) -15 d (3 d run-in + 3 d intervention + 3 d wash-out +3 d intervention + 3 d wash-out)	-Run-in and wash-out diet with WMD -Overfeeding (OF) with either 2400 kcal or 3400 kcal per d	-Energy in stool, and urine -Microbiome (16S rRNA)	-Calories consumed in 3400 kcal intervention correlated with relative abundance of Firmicutes (r = 0.47, P = 0.04) and Bacteroidetes (r = -0.47, P = 0.04)
				-Changes in stool calories (% of ingested calories) in OF was correlated with change in relative abundance of Firmicutes (r = -0.50, P = 0.02) and Bacteroidetes (r = 0.52, P = 0.01)
				-In subjects with normal weight, there was decrease in energy absorption (P < 0.05)
Kaczmarek et al., 2019 [40]	-Healthy adults (n = 18) -60 d (18 d intervention + 24 d wash-out + 18 d intervention)	-WMD with no Brassica diet (control) or same diet with broccoli (200 g + 20 g of raw daikon radish as a source of myrosinase to hydrolyze the glucosinolates)	-Plasma metabolites -Microbiome and funtional potential (16S rRNA)	-Broccoli consumption reduced F/B ratio
				-No differences in α-diversity, β-diversity was smaller in broccoli than in control (P = 0.03)
				-Broccoli increased Bacteroidetes (P = 0.03), reduced Firmicutes (P = 0.05), and, thus, F/B ratio (P = 0.01)
				-People with BMI <26 kg/m2 increased plasma metabolites, and BMI >26 kg/m2 decreased it
				-People with BMI <26 kg/m2 decreased F/B ratio (P = 0.04) in broccoli.
				-In broccoli, the peak in plasma glucosinolate metabolites was correlated with change in Bacteroidetes (r = 0.69, P = 0.04) and change in Firmicutes (r = -0.66, P = 0.05).
				-Broccoli had increased pathways related to endocrine system (P = 0.05), transport and catabolism (P = 0.04), and energy metabolism (P = 0.01), and decreased membrane transport (P = 0.03) compared to control.
Karusheva et al., 2019 [41]	-Type 2 diabetes patients (n = 12) -28 d (7 d run-in + 7 d intervention + 7 d wash-out + 7 d intervention)	-Intervention: diet with low branched-chain amino acids (<60% BCAA-) or with all essential amino acids (BCAA+)	-TEE (estimated) -Mitochondrial oxidative capacity -Microbiome (metabolomic)	-No difference in TEE and skeletal muscle mitochondrial oxidative capacity
				-BCAA- increased adipose tissue mitochondrial efficiency (P < 0.05)
				-BCAA+ increased indicators of β-oxidation in adipose tissue (P < 0.05)
				-BCAA- had reduced Firmicutes (P < 0.05) and increased Bacteroidetes (P < 0.05) compared to BCAA+
Papadopoulou et al., 2020 [42]	-Healthy adults (n = 17) -4 wk (1-wk intervention + 2 wk wash-out + 1-wk intervention)	-Intervention: 60% of diet consumed either at lunch or dinner	-Stool energy and SCFAs -Microbiome (16S rRNA)	-No difference in stool characteristics, energy, and SCFAs
				-Microbiome clustered by enterotype, not meal timing
				-When participants had large lunch there was higher concentration of E. coli (P = 0.03)
Single-arm interventions
Kelder et al., 2014 [43]	-Healthy men (n = 10) -4 wk	-High-fat high-caloric diet (HFHC) was provided to all participants	-REE (metabolic cart) -Fecal SCFAs -Microbiome (16S rRNA)	-HFHC increased body weight (P < 0.001), carbohydrate oxidation (P < 0.001), and REE (P < 0.05)
				-HFHC did not change overall microbiome composition
				-Correlation network show that carbohydrate oxidation had a positive correlation with taxa in Firmicutes phylum and negative correlation with taxa in Bacteroidetes phylum (r and p values not shown)
				-Multiple SCFAs correlated with Porphyromonadaceae and Sutterellaceae
				-REE had a positive correlation with P/B ratio and Prevotellaceae
Ott et al., 2018 (44)	-Healthy males with normal weight (n = 24) -4 wk (7 d run-in + 7 d intervention + 14 d wash-out	-Run-in and wash-out participant’s usual WMD -Intervention HFHC	-REE (metabolic cart) -Gut permeability -Microbiome (metagenomic)	-No difference in REE
				-No consistent difference in gut permeability
				-No difference in α-diversity
				-HFHC decreased Bacteroidaceae and increased Betaproteobacteria

Abbreviations: BCAA: Branched-chain amino acids; BMI: Body mass index; DLW: Doubly labeled water; F/B ratio: Firmicutes to Bacteroidetes ratio; FMT: Fecal microbiome transplantation; HFHC: high-fat high-caloric diet; OF: Overfeeding; REE: Resting energy expenditure; P/B ratio: Prevotella to Bacteroides ratio; RER: respiratory exchange ratio; RT-PCR: Reverse transcription polymerase chain reaction; SCFA: Short-chain fatty acid; TEE: Total energy expenditure; UF: Underfeeding; WMD: weight-maintaining diet.