Leuconostoc mesenteroides mediates an electrogenic pathway to attenuate the accumulation of abdominal fat mass induced by high fat diet

Abstract

Although several electrogenic bacteria have been identified, the physiological effect of electricity generated by bacteria on host health remains elusive. We found that probiotic Leuconostoc mesenteroides (L. mesenteroides) can metabolize linoleic acid to yield electricity via an intracellular cyclophilin A-dependent pathway. Inhibition of cyclophilin A significantly abolished bacterial electricity and lowered the adhesion of L. mesenteroides to the human gut epithelial cell line. Butyrate from L. mesenteroides in the presence of linoleic acid were detectable and mediated free fatty acid receptor 2 (Ffar2) to reduce the lipid contents in differentiating 3T3-L1 adipocytes. Oral administration of L. mesenteroides plus linoleic acid remarkably reduced high-fat-diet (HFD)-induced formation of 4-hydroxy-2-nonenal (4-HNE), a reactive oxygen species (ROS) biomarker, and decreased abdominal fat mass in mice. The reduction of 4-HNE and abdominal fat mass was reversed when cyclophilin A inhibitor-pretreated bacteria were administered to mice. Our studies present a novel mechanism of reducing abdominal fat mass by electrogenic L. mesenteroides which may yield electrons to enhance colonization and sustain high amounts of butyrate to limit ROS during adipocyte differentiation.

Figure 1

Characterization of electricity produced by L. mesenteroides EH-1 bacteria in the presence of linoleic acid. (a) Chronoamperometry results were obtained from an electrochemical chamber with L. mesenteroides EH-1 bacteria alone (B), 2% linoleic acid (LA), bacteria plus linoleic acid (B-LA). The changes in voltage (mV) against time (min) were recorded by Lutron software. (b) Ferrozine and ferric ammonium citrate were added into media with linoleic acid (LA), L. mesenteroides EH-1 bacteria (B), bacteria plus linoleic acid (B-LA). The dark brown complexes of ferrozine-chelatable irons (mmol/l) were photographed. (c) The OD₅₆₂ values of ferrozine-chelatable irons (mmol/l) were quantified. (d) Linoleic acid fermentation of L. mesenteroides EH-1. L. mesenteroides EH-1 (B) was incubated in rich media with/without linoleic acid (LA) for 24 h. Rich media plus linoleic acid (LA) without L. mesenteroides EH-1 were included as controls. (e) Fermentation was detected by OD₅₆₂. (f) The concentrations of nine SCFAs were detected by GC–MS analysis. Data shown represent the mean ± SD of experiments performed in triplicate. ***p < 0.001 (two-tailed t-test by GraphPad Prism 5).

Figure 2

Effect of linoleic acid fermentation of L. mesenteroides EH-1 and butyrate on adipocyte differentiation. (a) From left to right: 3T3-L1 preadipocytes were treated with culture media (CM), differentiation media (DM), differentiation media with rich media containing linoleic acid (M-LA) or supernatant from the culture of L. mesenteroides EH-1 bacteria alone (M-B), bacteria plus linoleic acid (M-B-LA). Lipids (arrows) were stained with Oil Red O and extracted by isopropanol for quantification by absorbance at OD₅₁₀. (b) From left to right: 3T3-L1 preadipocytes were treated with supernatant from the culture of bacteria plus linoleic acid in the presence of GLPG-0974 (GLPG), a Ffar2 antagonist (M-B-LA-GLPG), or DMSO (M-B-LA-DMSO). Preadipocytes in differentiation media treated with butyrate (Butyrate-DMSO) or a Ffar2 antagonist (Butyrate-GLPG) were included. Data are the mean ± SD of experiments performed in triplicate. **p < 0.001. ***p < 0.0001. ns non-significant (two-tailed t-test by GraphPad Prism 5). Bars = 100 µm.

Figure 3

Reduction of the production of ROS and 4-HNE by linoleic acid fermentation of L. mesenteroides EH-1. (a) Cells loaded with DCFH-DA were treated with media (CM, DM), linoleic acid (M-LA) or supernatants (M-B or M-B-LA) as described in Fig. 2a. Green fluorescence (arrows) derived from DCFH-DA reaction was quantified by measurement with excitation and emission at 485 and 530 nm. (b) ICR mice were fed a SCD or HFD at a 3-day interval for 6 weeks. The levels of 4-HNE and β-actin in abdominal fat mass by western blot analysis were quantified by Image J software. The 4-HNE production in abdominal fats in mice administered with linoleic acid alone (LA), L. mesenteroides EH-1 bacteria alone (B), bacteria plus linoleic acid (B-LA) was examined. The ratio intensities of 4-HNE to β-actin were quantified by Image J software. Full-length western blot images were presented in Fig. S4. Data are the mean ± SD of experiments performed in triplicate. *p < 0.05. **p < 0.01. ***p < 0.001 (two-tailed t-test by GraphPad Prism 5). Bars = 100 µm.

Figure 4

Requirement of L. mesenteroides EH-1 cyclophilin A for bacterial electricity and adhesion, as well as reduction of 4-HNE. (a) mRNA expression of cyclophilin A (CypA) in L. mesenteroides EH-1 pretreated with (I-B) or without TMN355 (B) was detected by real-time PCR. (b) Chronoamperometry results were obtained from an electrochemical chamber with L. mesenteroides EH-1 pretreated with (I-B-LA) or without (B-LA) TMN355 in the presence of linoleic acid. (c) Ferrozine and ferric ammonium citrate were added into media with L. mesenteroides EH-1 pretreated with or without TMN355 in the presence of linoleic acid. The dark brown complexes of ferrozine-chelatable irons were photographed. (d) The OD₅₆₂ values of ferrozine-chelatable irons (mmol/l) were quantified. (e) Adhesion of L. mesenteroides EH-1 pretreated with or without TMN355 in the presence of linoleic acid on Caco-2 cells. Bacterial CFUs were counted by plating serial dilutions (1∶10⁰ to 1∶10⁵) of Caco-2 cells with adherent bacteria on a TSB agar plate and (f) the number (log₁₀ CFU/ml) of adherent bacteria was shown. (g) The 4-HNE production in abdominal fats in mice administered L. mesenteroides EH-1 plus linoleic acid (B-LA) or TMN355-pretreated bacteria plus linoleic acid (I-B-LA) was examined. The levels of 4-HNE and β-actin in abdominal fat mass were detected by western blot analysis. The ratio intensities of 4-HNE to β-actin were quantified by Image J software. Full-length western blot images were presented in Fig. S5. Data shown represent the mean ± SD of experiments performed in triplicate. ***p < 0.001 (two-tailed t-test by GraphPad Prism 5).

Figure 5

Essential role of electricity produced by bacteria plus linoleic acid in reduction of abdominal fat mass. (a) The abdominal fat (arrows) in whole body or isolated abdominal fat masses in mice fed with SCD or HFD was shown. (b) Change in body weight was recorded and (c) abdominal fat masses were dissected and weighed. HFD-fed ICR mice were administered linoleic acid (LA) alone, L. mesenteroides EH-1 bacteria alone (B), bacteria plus linoleic acid (B-LA) or TMN355-pretreated bacteria plus linoleic acid (I-B-LA) for 6 weeks. (d) Representative mice with abdominal fat in whole body or isolated abdominal fat masses were shown. (e) Change in body weight was recorded and (f) abdominal fat masses were dissected and weighted. Scale bars (Black) = 10 mm. Scale bars (Red) = 5 mm. Data shown represent the mean ± SD of experiment in triplicate using five mice per group. *p < 0.05. ***p < 0.001. (two-tailed t-tests by GraphPad Prism 5).

Figure 6

An outline of the actions of L. mesenteroides EH-1 on reduction of HFD-induced accumulation of abdominal fat mass. The L. mesenteroides EH-1 metabolized linoleic acid to yield electrons (e⁻) via an intracellular cyclophilin A (CypA)-dependent pathway. Electrons enhance bacterial colonization and sustain high amounts of butyrate in the gut. The butyrate may circule through the bloodstream, bind to the Ffar2 on the surface of adipocytes, limit the formation of 4-HNE during adipocyte differentiation, and lower the HFD-induced accumulation of abodominal fat mass. The Adobe Illustrator 2020 (Adobe, San Jose, CA, USA) was used to make this outline.