1. Volume 163, Issue 7, Pages 1585-1595 (December 2015)
Non-lethal Inhibition of Gut Microbial Trimethylamine Production for the Treatment of Atherosclerosis 
Zeneng Wang, Adam B. Roberts, Jennifer A. Buffa, Bruce S. Levison, Weifei Zhu, Elin Org, Xiaodong Gu, Ying Huang, Maryam Zamanian-Daryoush, Miranda K. Culley, Anthony J. DiDonato, Xiaoming Fu, Jennie E. Hazen, Daniel Krajcik, Joseph A. DiDonato, Aldons J. Lusis, Stanley L. Hazen 
Cell 
Volume 163, Issue 7, Pages (December 2015)
DOI: /j.cell
Copyright © 2015 Elsevier Inc. Terms and Conditions

2. Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

3. Figure 1
The Choline Analog DMB Inhibits Microbial Choline TMA Lyase Activity
(A) Effect of the indicated choline analogs on microbial TMA lyase activity (measured as d9-TMA production from 100 μM of the indicated d9-labeled substrate) from the lysate of E. coli Top10 cells transformed (pBAD vector) with cutC/D genes (from P. mirabilis).
(B) Effect of DMB on choline TMA lyase activity in intact P. mirabilis incubated with the indicated concentrations of d9-choline substrate with or without DMB (1 mM) at 37°C. NA, no addition.
(C) DMB effect on choline uptake. P. mirabilis (OD600 nm = 0.5) cells were pelleted and then re-suspended in minimal media supplemented with the indicated concentrations of d9-choline with or without DMB (2 mM) for 15 min at 37°C. Intracellular d9-choline was then quantified as described in the Experimental Procedures.
(D) Choline TMA lyase activity from intact E. coli Top10 cells transformed with the indicated constructs in the presence versus absence of DMB.
(E) DMB dose-response curves for inhibition of choline TMA lyase activity in intact E. coli Top10 cells transformed with cutC/D genes from either D. alaskensis (pUC57 vector) or P. mirabilis (pBAD vector).
(F) TMA lyase activity for the indicated substrates in P. mirabilis lysate with or without DMB.
Data are presented as mean ± SE from three independent replicates.
See also Figures S1, S2, and S6.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

4. Figure 2
Inhibitory Effect of DMB on Alternative Microbial TMA Lyases and Both Mouse Cecal and Human Fecal Microbial TMA Lyase Activities with Multiple Substrates
(A) TMA lyase activity for the indicated substrates (375 μM each) in combined lysates from E. coli BL21 cells transformed (pET22 vector) with cntA and cntB genes (from A. baumannii) incubated in the presence or absence of DMB.
(B) TMA lyase activity with the indicated substrates (375 μM, with or without DMB) in intact E. coli BL21 cells transformed (pET22 vector) with yeaW/X genes (from E. coli DH10B).
(C–E) TMA lyase activity with the indicated substrates (with or without DMB) incubated with (C) mouse cecum lysate (1.3 mg/ml protein) or (D and E) human fecal microbes (equivalent to 100 mg/ml). The DMB concentration in all reaction systems was 10 mM (in A and B) or 2 mM (in C–E).
Data are presented as mean ± SE from three independent replicates.
See also Figures S3, S6C, and S6D.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

5. Figure 3 DMB Serves as a Non-lethal Inhibitor
(A–C) P. mirabilis (A), P. penneri (B), and E. fergusonii (C) cells were grown in nutrient broth ± DMB (0.1% v/v) and both the growth rate (OD600) and total amount of d9-TMA produced from d9-choline (100 μM) (insert) were monitored. NA, no addition.
(D and E) Demonstration that DMB decreases plasma TMAO levels in male (D) and female (E) C57BL/6J Apoe−/− mice placed chronically on either a 1.0% choline-supplemented or a 1.0% carnitine-supplemented diet in the presence versus absence of DMB (1%, v/v in drinking water).
(F) Three groups of male C57BL/6J Apoe−/− mice (8 weeks old) were placed on a chemically defined diet equivalent to chow supplemented with choline (1.0% total choline). Two groups of mice were administered 1.2 μmol of DMB daily (provided in corn oil; total volume, 150 μl) by either (oral) gastric gavage or subcutaneous (SC) route, and the third group of mice received sham (vehicle) gavages daily. After 2 weeks, plasma TMAO levels were determined by LC/MS/MS.
Data are presented as mean ± SE from three independent replicates (A–C) or the indicated numbers of independent replicates (D–F). See also Figure S4C and Tables S1 and S2.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

6. Figure 4
DMB Attenuates Choline-Enhanced Foam Cell Formation and Atherosclerosis
(A and B) Representative Oil-Red-O/hematoxylin staining of peritoneal macrophages (A) and foam cell quantification (B) from 20-week-old male C57BL/6J Apoe−/− mice fed chemically defined chow (0.07% total choline) or 1% choline-supplemented diets from the time of weaning (4 weeks). Scale bars in (A), 50 μm. Data are presented as mean ± SE.
(C) Representative Oil-Red-O/hematoxylin staining of aortic root sections from 20-week-old male C57BL/6J Apoe−/− mice fed chemically defined chow (0.07% total choline) or choline-supplemented (1.0% total choline) diets in the presence versus absence of DMB provided in the drinking water, as described in the Experimental Procedures. Scale bars, 200 μm.
(D) Aortic root lesion area was quantified in 20-week-old male C57BL/6J Apoe−/− mice from the indicated diet and DMB treatment groups.
Data are presented as mean ± SE. p values shown were calculated by ANOVA. See also Figures S4 and S5.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

7. Figure 5 DMB Alters Gut Microbial Composition
(A–E) Unweighted UniFrac distances plotted in principal-component analysis (PCoA) space comparing cecal microbial communities in male (A) or female (C) C57BL/6J Apoe−/− mice fed chow versus choline-supplemented diet in the presence versus absence of DMB. Each data point represents a sample from a distinct mouse projected onto the first two principal coordinates (percent variation explained by each principal component [PCo] is shown in parentheses). (B, D, and E) Impact of diet and DMB on the proportion of several taxa.
Data are presented as mean ± SE (n ≥ 9 mice per group). See also Tables S3 and S4.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

8. Figure 6
Schema Showing Use of DMB to Inhibit Gut Microbial TMA Production for the Treatment of Atherosclerosis
DMB is a structural analog of choline and an inhibitor of microbial TMA production (a choline TMA lyase inhibitor). Provision of DMB in the drinking water of atherosclerosis-prone Apoe−/− mice inhibits choline-diet-dependent enhancement in TMAO, endogenous macrophage foam cell formation, and atherosclerosis development.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

9. Figure S1
Characterization of Choline TMA Lyase Activity of P. mirabilis and Recombinant CutC/D from P. mirabilis and D. alaskensis, Related to Figure 1
(A) E. coli Top10 cells were transformed with empty pBAD vector (negative control), pBAD-cutC/D (from P. mirabilis), or the indicated catalytically inactive cutC point mutants, as described in Experimental Procedures. Cells were grown, CutC/D expression was induced with 0.2% arabinose, and then cells collected by centrifugation, re-suspended in 2x culture volumes of PBS and incubated at 37°C in gas tight reaction vials with 100 μM of d9-choline substrate. At the indicated time points, TMA lyase activity was measured as d9-TMA production using stable isotope dilution LC/MS/MS.
(B) Effect of the indicated choline analogs on choline TMA lyase activity of P. mirabilis lysate. Protein lysate (3 mg) was incubated with d9-choline (100 μM) in 2 ml PBS with the indicated concentrations of choline analogs (P-choline, betaine, TMSi-ETOH, and DMB) at 37°C for 23 hr in gas tight reaction vials and choline TMA lyase activity measured as the production of d9-TMA, as described under Experimental Procedures.
(C) E. coli Top10 cells were transformed with pUC57 plasmid (negative control) or pUC57 expression vectors constitutively expressing D. alaskensis cutC alone, cutD alone, cutC- intergenic region-cutD, or the indicated catalytically inactive cutC point mutants, as described in Experimental Procedures. Bacterial cells at stationary phase (OD600 nm = 1.5), were pelleted, re-suspended in 6x original culture volume of cold argon-sparged PBS containing d9-choline (100 μM) substrate ± DMB (2 mM) and incubated in gas-tight reaction vials under argon blanket at 37°C for 12 hr. Choline TMA lyase activity was quantified by the production of d9-TMA by stable isotope dilution LC/MS/MS.
Data represent mean ± SE from three independent replicates (A-C). NA, no addition.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

10. Figure S2
Expression of P. mirabilis CutC/D in E. coli Stellar Cells and the Effects of DMB on CutC/D TMA Lyase Activity, Related to Figure 1
(A) E. coli Stellar cells were transformed with pGK::nucMCS plasmid harboring constructs with P. mirabilis cutC/D genes under the control of the Staphylococcus nuclease promoter, Pnuc. CutC was N-terminally tagged with an 8x-His-tag, and CutD was C-terminally tagged with Strep-Tag II, as described under Experimental Procedures. As an additional negative control, cells were also transformed with the empty vector. Shown is the SDS-PAGE with Coomassie blue staining of bacterial lysates from the indicated transformed cells. Identity of bands as CutC and CutD was confirmed by excising of each band, and performing LC/MS/MS analysis on in-gel tryptic digests (data not shown).
(B–E) P. mirabilis cutC/D-transformed, or empty-vector transformed E. coli Stellar cells were grown up to stationary phase (OD600 = 1.5), harvested by centrifugation and re-suspended in 1x PBS with a volume of 6:1 (original culture volume), and then incubated with 100 μM of the indicated d9- labeled substrates ± DMB (2 mM) at 37°C for 24 hr in gas-tight reaction vials, as described under Experimental Procedures. As an additional (negative) control, isotope-labeled substrates in PBS without addition of cells were incubated in sealed reaction vials. d9-TMA produced was measured by LC/MS/MS.
(F) SDS-PAGE analysis of purified CutC after Ni-NTA IMAC affinity chromatography, and purified CutD by Strep-Tactin affinity chromatography.
(G and H) Purified CutC (30 μg) and CutD (30 μg) were incubated together in PBS (1 ml) supplemented with Na2S2O4 (2 mM) and S-adenosyl-L-methionine (1 mM) in the presence of isotope-labeled substrate (100 μM of either d9-choline (G) or d9-GPC (H)) and the indicated concentrations of DMB, in gas-tight reaction vials under anaerobic conditions at 37°C for 12 hr. d9-TMA produced was then quantified by LC/MS/MS. Reactions without CutC or CutD were used as blank controls for each individual d9-TMA containing compound (d9-choline or d9-GPC) and showed negligible (∼0.25 nM) d9-TMA production.
Data (B-E,G,H) represent mean ± SE from 3 independent replicates.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

11. Figure S3
Characterization of the Effect of DMB on TMA Lyase Activity of the Reiske-type Proteins CntA/B from A. baumannii and YeaW/X from E. coli, Related to Figure 2
(A) E. coli BL21(DE3) cells were transformed with pET plasmid harboring constructs with A. baumannii (ATCC 19606) synthesized genes (GenScript) for cntA alone, cntB alone, or the catalytically inactive E205D cntA point mutant, as described under Experimental Procedures. As an additional negative control, E. coli were also transformed with empty vector. The indicated bacterial lysates (3 mg protein each) were then assayed for carnitine TMA lyase activity using d9-carnitine (100 μM) as substrate in the presence versus absence of DMB (10 mM). Reactions were performed in PBS in the presence of NADPH (200 μM) and ammonium iron(II) sulfate hexahydrate (1 mM) at 37°C for 16 hr. d9-TMA produced was quantified as described under Experimental Procedures.
(B) E. coli BL21 (DE3) cells were transformed with pET plasmid harboring constructs with E. coli DH10B genes for yeaW alone, yeaX alone, yeaW + yeaX, or the catalytically inactive E208D yeaW point mutant + yeaX, as described under Methods. As an additional negative control, E. coli were also transformed with empty pET plasmid. Bacterial cells at stationary phase (OD600 nm = 1.5) were harvested by centrifugation and re-suspended in PBS containing d9-choline (100 μM) substrate in the presence versus absence of DMB (10 mM) and incubated in gas-tight reaction vials at 37°C for 16 hr. Choline TMA lyase activity was quantified by the production of d9-TMA by stable isotope dilution LC/MS/MS as described under Experimental Procedures.
(C) SDS-PAGE analysis of purified YeaW and YeaX after Ni-NTA IMAC chromatography.
(D) Purified YeaW + YeaX (30 μg each) were incubated in PBS supplemented with NADH (250 μM) and the indicated d9-TMA containing substrates (200 μM) in gas-tight vials at 37°C for 8 hr. d9-TMA produced was then quantified as described under Experimental Procedures.
Data represent mean ± SE from 3 independent replicates (A, B, D).
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

12. Figure S4
DMB Inhibits Atherosclerosis in Female Apoe−/− Mice on a High-Choline Diet, Affects Flavin Monooxygenase Activity, and Is Not Detected in Urine if Orally Administered, Related to Figures 3 and 4
(A) C57BL/6J Apoe−/− mice at time of weaning (4 weeks) were placed on chemically defined chow diet (0.07% total choline) or chow diet supplemented with choline (1.0% total choline), with or without 1% DMB added to the drinking water. At 20 weeks of age, mice were sacrificed. Hearts/aorta from mice were fixed and processed for aortic lesion area quantification, as described in Experimental Procedures. Data shown represent mean ± SE from the indicated number of female mice per treatment group (data for male mice is shown in main manuscript).
(B) Liver tissue homogenate from male mice (1 mg/ml protein) was incubated with d9-TMA (100 μM) in PBS supplemented with and without NADPH (100 μM) at 37°C for 8 hr. Reactions were stopped by adding 0.1 N formic acid, and d9-TMAO formed quantified by LC/MS/MS using d4-choline as internal standard. Total hepatic FMO activity was calculated as the NADPH-dependent d9-TMAO formed. Data represent mean ± SE from the indicated number of mice.
(C) Identification of DMB in urine by GC/MS. GC chromatograms of blank (violet), DMB standard spiked into urine (green), urine recovered 30 min after gastric gavage of mice with 1.9 μl (15 μmol) DMB in 200 μl saline (red), urine recovered 30 min after intraperitoneal injection (ip) of mice with 1.9 μl (15 μmol) DMB in 200 μl saline (14.2 μM DMB detected) (blue). The red dotted line indicates the retention time of DMB (3,3-dimethyl-1-butanol). p values shown were calculated by ANOVA.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

13. Figure S5 Liver Metabolism of DMB, Related to Figure 4
The indicated units of liver alcohol dehydrogenase were incubated in sodium phosphate buffer (50 mM, pH 7.0) supplemented with NAD+ (7.5 mM) and either ethanol or DMB (each 3.5%) at 25°C with spectra (  nm) serially acquired over time (Spectra Max Plus, Molecular Devices, Sunnyvale, CA). Representative optical absorption spectra shown are from comparable time points for ethanol (A) and DMB (B) incubations, and represent the formation of the product, NADH, with the oxidation of alcohol.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions

14. Figure S6
Analysis of Synthesized P-Choline and d9-GPC by High-Resolution Mass Spectrometry, Related to Figures 1 and 2
(A) Positive ion TOF mass spectrum of synthesized (2-hydroxyethyl)(trimethyl) phosphonium (P-choline).
(B) Collision induced dissociation (CID) mass spectrum of the parent ion at m/z = The prepared compound, P-choline, was dissolved in water at a concentration of 50 μM, and was injected onto an AB SCIEX TripleTOF 5600 LC-MS/MS System at a flow rate of 10 μl/min with an electrospray interface operating in positive TOF scan mode or product scan mode, using the following operation parameters: GS1 25; GS2 15; Curtain gas 25; ISVF 5500; CE 30; and the accurate mass spectra were recorded. An automated Calibrant Delivery System (ABSCIEX) using a dual nebulizer APCI source that introduces the flow from APCI Positive Calibration Solution (AB SCIEX, Cat# ) to Triple TOF was used to calibrate TOF MS and product ions.
(C) Positive ion TOF mass spectrum of synthesized [d9-N,N,N-trimethyl]glycerophosphocholine (d9-GPC).
(D) Collision induced dissociation (CID) mass spectrum of the parent ion at m/z = The prepared compound, d9-GPC, was dissolved in water at a concentration of 50 μM, and was injected to AB SCIEX TripleTOF 5600 LC-MS/MS System as above.
Cell  , DOI: ( /j.cell )
Copyright © 2015 Elsevier Inc. Terms and Conditions