The TMAO-Producing Enzyme Flavin-Containing Monooxygenase 3 Regulates Obesity and the Beiging of White Adipose Tissue

Abstract

Emerging evidence suggests that microbes resident in the human intestine represent a key environmental factor contributing to obesity-associated disorders. Here, we demonstrate that the gut microbiota-initiated trimethylamine N-oxide (TMAO)-generating pathway is linked to obesity and energy metabolism. In multiple clinical cohorts, systemic levels of TMAO were observed to strongly associate with type 2 diabetes. In addition, circulating TMAO levels were associated with obesity traits in the different inbred strains represented in the Hybrid Mouse Diversity Panel. Further, antisense oligonucleotide-mediated knockdown or genetic deletion of the TMAO-producing enzyme flavin-containing monooxygenase 3 (FMO3) conferred protection against obesity in mice. Complimentary mouse and human studies indicate a negative regulatory role for FMO3 in the beiging of white adipose tissue. Collectively, our studies reveal a link between the TMAO-producing enzyme FMO3 and obesity and the beiging of white adipose tissue.

Keywords: FMO3; adipose; diabetes; flavin-containing monooxygenase 3; microbiota; nutrition; obesity.

Conflict of interest statement

CONFLICTS OF INTEREST

R.C.S., D.M.S., M.W. R.N.H, A.B., D.F., A.L.B., A.D.G., M.H., A.C., L.L., X.S.L., B.W., Y.H.M, H.K., N.C., C.P., R.E.M., C.D.L., S.K.D., L.L.R., N.Z., A.J.M., S.D., W.H.W.T., B.O.E, C.A.F, M.L., M.C., S.V.N.P., J.H., A.J.L, and J.M.B. have no conflicts of interest to declare. S.L.H. and Z.W. are named as co-inventors on pending and issued patents held by the Cleveland Clinic relating to cardiovascular diagnostics and therapeutics. S.L.H. reports he has been paid as a consultant by the following companies: Esperion, and Procter & Gamble. S.L.H. also reports he has received research funds from Astra Zeneca, Procter & Gamble, Roche, and Takeda. S.L.H. has the rights to receive royalty payments for inventions or discoveries related to cardiovascular diagnostics from Cleveland Heart Lab Inc., Frantz Biomarkers, and Siemens Healthcare. R.G.L., R.M.C., and M.J.G. are employees at Ionis Pharmaceuticals, Inc. (Carlsbad, CA).

Copyright © 2017 The Authors. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1. Elevated Circulating Levels of TMAO are Associated with Type 2 Diabetes Mellitus in Humans

We recruited two separate cohorts of stable subjects in preventative cardiology (n=187) or hepatology clinics (n=248) to evaluate associations between fasting circulating choline or trimethylamine-N-oxide (TMAO) levels with prevalent type 2 diabetes (T2DM). The total number of subjects recruited in both studies was n=435. Patient demographics, laboratory values, and clinical characteristics are provided in Tables S1–S3 and Figure S1. (A) Relationship of fasting plasma TMAO concentrations and prevalent T2DM. Boxes represent the 25th, 50th and 75th percentiles of plasma TMAO concentration, and whiskers represent the 10th and 90th percentiles. (B) Relationship of fasting plasma choline concentrations and prevalent T2DM. Boxes represent the 25th, 50th and 75th percentiles of plasma choline concentration, and whiskers represent the 10th and 90th percentiles. (C) Forest plots of the odds ratio of prevalent T2DM and quartiles of TMAO; bars represent 95% confidence intervals. (D) Forest plots of the odds ratio of prevalent T2DM and quartile of choline; bars represent 95% confidence intervals.

Figure 2. Plasma TMAO Levels in Mice and FMO3 mRNA Expression in Men Demonstrate Positive Correlations with Obesity

(Panels A–D) Correlation of plasma trimethylamine-N-oxide (TMAO) levels with obesity-related traits in 180 male mice from 92 inbred strains within the hybrid mouse diversity panel (HMDP) after 8 week feeding of a high fat and high sucrose diet. Correlation coefficient (r) and p value (p) are indicated for each obesity trait. (A) Correlation between plasma TMAO and body weight (B) Correlation between plasma TMAO and fat mass (C) Correlation between plasma TMAO and mesenteric fat weight (D) Correlation between plasma TMAO and subcutaneous fat weight (E) Correlations between human white adipose tissue flavin monooxygenase 3 (FMO3) mRNA expression and metabolic traits or brown/beige adipocyte marker gene expression (n=770). The gene name and probeset ID is provided for each of the brown/beige adipocyte marker genes.

Figure 3. FMO3 Knockdown Protects Mice from High Fat Diet-Induced Obesity by Stimulating the Beiging of White Adipose Tissue

At 6–8 weeks of age, female C57BL/6 mice were treated with either a non-targeting control ASO or or Fmo3 ASO in conjunction with either standard rodent chow or high fat diet (HFD) feeding for the indicated times. (A) Hepatic Fmo3 mRNA expression was quantified by qPCR after 10 weeks (B) Plasma levels of TMA after 6 weeks (C) Plasma levels of TMAO after 6 weeks (D) Body weight changes over 12 weeks (E) Gonadal white adipose tissue (WAT) weight at necropsy Panels (F–J) MRI images and subsequent quantification of adiposity in control and Fmo3 ASO-treated mice maintained on diets for 14 weeks. (F) MRI images of control and Fmo3 ASO-treated mice (G) Peritoneal adipose tissue mass (H) Subcutaneous adipose tissue mass (I) % Fat mass (J) % Lean mass All data represent the mean ± S.E.M. for n=5–10 mice per group; *, p≤0.05, **, p≤0.01, ***, p≤0.001, ****, p≤0.0001 vs. control ASO-treated mice fed the same diet. +, p≤0.05, ++, p≤0.01, +++, p≤0.001, ++++, p≤0.0001 vs. chow-fed mice treated with the same ASO.

Figure 4. FMO3 Knockdown Stimulates the Beiging of White Adipose Tissue

Panels (A–E) At 6–8 weeks of age, female C57BL/6 mice were treated with either a non-targeting control ASO or Fmo3 ASO in conjunction with high fat diet feeding (HFD) for 8–12 weeks. (A) Microscopic examination of hematoxylin and eosin stained gonadal white adipose tissue; scale bar equals 200µm. (B) Gonadal white adipose tissue (WAT) mRNA expression of β1-adrenergic receptor (Adrb1), uncoupling protein 1 (Ucp1), and transmembrane protein 26 (Tmem26) quantified by qPCR. (C) Surface density of β1-adrenergic receptor (β1-AR) in gonadal white adipose tissue (WAT) quantified by radio-ligand binding assay. (D) Cyclic AMP (cAMP) levels in gonadal white adipose tissue (WAT). (E) Peroxisome proliferator-activated receptor gamma coactivator 1-alpha (Ppargc1a) mRNA expression was quantified by qPCR in gonadal (gWAT), inguinal (iWAT) and brown (BAT) adipose tissue. Panels (F–H) Mice were housed in metabolic cages for indirect calorimetry measurements. Gray background denotes dark cycle. (F) Oxygen consumption (VO2) (G) Heat production (H) Respiratory exchange ratio (RER) Panels (I–K) At 6–8 weeks of age, female C57BL/6 mice were treated with either a non-targeting control ASO or Fmo3 ASO in conjunction with feeding of HFD or HFD supplemented with 0.02% w/w TMAO for 12 weeks. See also Figure S4. (I) Body weight changes over 12 weeks (J) Gonadal white adipose tissue (WAT) expression of uncoupling protein 1 (Ucp1) (K) Gonadal white adipose tissue (WAT) expression of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (Ppargc1a), PR domain-containing 16 (Prdm16), T-box transcription factor (Tbx1), and transmembrane protein 26 (Tmem26) All data represent the mean ± S.E.M. for n=5–10 mice per group; *, p≤0.05, **, p≤0.01, ***, p≤0.001, ****, p≤0.0001 vs. control ASO-treated mice fed the same diet. +++, p≤0.001 vs. chow-fed mice treated with the same ASO. &&&, p≤0.001 vs. Fmo3 ASO-treated mice fed HFD.

Figure 5. Genetic Deletion of FMO3 Protects Mice From Diet-Induced Obesity

(A) Top panel: CRISPR-Cas9 strategy for generating Fmo3−/− mice. The sequence of exon 2 of the murine Fmo3 coding sequence is shown. The target sequence (underlined) used for construction of the guide RNA is shown with arrows indicating predicted cleavage sites by Cas9. Bottom panel: immunoblotting analysis of FMO3 protein levels in the livers of wild-type (WT) and Fmo3−/− (KO) mice. (B) Decreased adiposity in Fmo3−/− mice. Fmo3+/+ (n=9), Fmo3+/− (n=6), and Fmo3−/− (n=11) mice were fed a 1.3% choline chloride (w/w) diet for 12 weeks before tissue collection. Plasma TMAO and TMA levels (top left), food intake (top right), body weight (bottom left) and 4 fat pads/body weight (%; bottom right) are shown. The four fat pads included in the 4 fat pads/body weight measurement were gonadal, mesentery, perirenal, and subcutaneous. *: p≤0.05 between Fmo3+/+ and Fmo3−/− groups. &: p≤0.05 between Fmo3+/− and Fmo3−/− genotype groups. Panels (C–H) Ldlr−/−; Fmo3−/− mice are more resistant to obesity than Ldlr−/− littermates when fed a Western diet for 12 weeks. (C) Liver FMO activity (D) Plasma TMAO levels (E) Body weight changes over 12 weeks (F) Fat mass/body weight (%) (G) 4 fat pads weight/body weight (%); the 4 fat pads measured were gonadal, mesentery, perirenal, and subcutaneous (H) Gene expression analysis of subcutaneous fat pads of Ldlr−/− (n=17) and Ldlr−/−; Fmo3−/− (n=9) mice. Cell death-inducing DFFA-like effector A (Cidea), Cytochrome C oxidase subunit 8b (Cox8b), Elongation of very long chain fatty acids protein 3 (Elovl3), Uncoupling protein 1 (Ucp1), Diglyceride acyltransferase 1 (Dgat1), Leptin (Lep), Stearoyl CoA desaturase-1 (Scd1), Transducin-like enhancer of split 3 (Tle3)*, p≤0.05, **, p≤0.01, and ***, p≤0.0001 between the two genotype groups.