Taurine-Mediated IDOL Contributes to Resolution of Streptococcus uberis Infection

Metabolic alterations occur in pathogenic infections, but the role of lipid metabolism in the progression of bacterial mastitis is unclear. Cross talk between lipid droplets (LDs) and invading bacteria occurs, and targeting of de novo lipogenesis inhibits pathogen reproduction.

ABSTRACT

Metabolic alterations occur in pathogenic infections, but the role of lipid metabolism in the progression of bacterial mastitis is unclear. Cross talk between lipid droplets (LDs) and invading bacteria occurs, and targeting of de novo lipogenesis inhibits pathogen reproduction. In this study, we investigate the role(s) of lipid metabolism in mammary cells during Streptococcus uberis infection. Our results indicate that S. uberis induces the synthesis of fatty acids and production of LDs. Importantly, taurine reduces fatty acid synthesis, the abundance of LDs and the in vitro bacterial load of S. uberis. These changes are mediated, at least partly, by the E3 ubiquitin ligase IDOL, which is associated with the degradation of low-density lipoprotein receptors (LDLRs). We have identified a critical role for IDOL-mediated fatty acid synthesis in bacterial infection, and we suggest that taurine may be an effective prophylactic or therapeutic strategy for preventing S. uberis mastitis.

INTRODUCTION

Streptococcus uberis is a major causative pathogen of clinical and subclinical bovine mastitis. It has been estimated that it is responsible for 9.6% to 22.1% of cases of clinically recognized mastitis in Europe (1). S. uberis was classified as an environmental pathogen (2); however, new molecular typing studies have demonstrated that there is also direct bacterial transmission and predominance of particular strains of bacteria, including sequence types (STs) 5, 6, and 20 (3). Because conventional mastitis control measures are often ineffective and may have adverse effects, it is urgent to find new ways to control mastitis.

Alterations of metabolism support pathogen survival in host cells (4, 5). In 1863, Rudolf Virchow identified lipid drop (LD)-laden macrophages in Mycobacterium leprae-infected patients (6). Current evidence indicates that many bacteria interact with host LDs, such as Mycobacterium tuberculosis (7), Mycobacterium bovis (8), and M. leprae (9). Lipid metabolism is associated with host-pathogen interactions, and LDs are required for growth and proliferation of invading bacteria (10). Although many bacteria share mechanisms targeting host LDs, different pathogens are associated with unique host LD-bacterial interactions. S. uberis, a Gram-positive bacterium, activates Toll-like receptor 2 (TLR2) receptors (11), which are essential in triggering LD formation in macrophages (12); however, the role of LDs in S. uberis infection is poorly understood.

Low-density lipoprotein receptors (LDLRs) are key determinants of circulating plasma lipoprotein levels (13). LDLRs are involved in lipid metabolism stabilization (14) and play a role in metabolic syndrome (15, 16). The E3 ubiquitin ligase IDOL is a posttranscriptional regulator of LDLR abundance (17). Inhibiting IDOL to enhance LDLR is a therapeutic strategy to target multiple metabolic syndrome manifestations, such as obesity, dyslipidemia, and hepatic lipidosis (18). Notwithstanding, there is little direct evidence indicating whether the IDOL-LDLR pathway plays a broader metabolic role in regulating LDs.

Taurine (a sulfur-containing amino acid) has a millimolar concentration in mammalian tissues (19). Taurine has been shown to effectively ameliorate metabolic diseases (20–22). Through metabolomics analysis, we found that taurine normalizes metabolic substances of the host, especially those related to fatty acid synthesis (unpublished data). Taurine regulates metabolism in a variety of ways, such as lipid metabolism, energy expenditure, suppression of oxidative stress, and inflammation (23). However, its effect on LDs is unclear.

Employing S. uberis to study the relationship among intracellular microbes, LDs, and lipid metabolism, we showed that taurine reduction of LDs in host cells contributes to reduction of bacterial loading. Additionally, we established IDOL-mediated fatty acid synthesis as a novel process reducing LDs. Thus, IDOL modulation of lipid metabolism may be exploited as a putative defense against intracellular pathogenic bacteria, by limiting pathogen access to host nutrients.

RESULTS

S. uberis infection induces LDs in host cells.

Pathogenic infection reshapes host metabolism, including lipid metabolism, to provide for its own growth and replication (24–26). Electron microscopic studies showed that S. uberis significantly induced LDs in mouse mammary epithelial cells (mMECs) (Fig. 1A). Similar results were obtained by labeling LDs with the neutral lipid marker BODIPY 493/503 for confocal imaging (Fig. 1B) and flow cytometric analyses (Fig. 1C). In addition, S. uberis enhanced the activity of acetyl coenzyme A (acetyl-CoA) carboxylase (ACC) (Fig. 1D) and gene expression (Scd1, Srebf1, and Fasn) related to lipid synthesis (Fig. 1E).

FIG 1.

S. uberis infection causes accumulation of LDs in mMECs. mMECs were infected with S. uberis at an MOI of 10 and incubated at 37°C for 3 h as described in Materials and Methods. (A) The ultrastructure of mMECs was observed by transmission electron microscopy (TEM). Images are representative of 10 samples/group. Scale bar, 1 μm. (B) BODIPY 493/503 staining for LD visualization (green) increased in S. uberis infection. mMECs were fixed with 4% paraformaldehyde and stained with BODIPY. Uninfected mMECs were incubated under identical conditions. Images are representative of 10 samples/group. Scale bar, 20 μm. (C) Intracellular LD content was evaluated by staining cells (10,000/sample) with BODIPY 493/503, followed by analysis with CellQuest Pro acquisition software and FlowJo software. Data are presented as geometric mean fluorescence intensity. The dotted line indicates background fluorescence in uninfected mMECs. (D) ACC enzyme activities. (E) Fasn, Scd1, and Srebf1 mRNA relative expression related to lipid synthesis in mMECs during S. uberis invasion. (F) Acox1 and Ehhadh mRNA relative expression relative to fatty acid β oxidation. (G) Cd36 mRNA relative expression related to fatty acid uptake. (H) The ultrastructure of mMECs was obtained by TEM. Images are representative of 10 samples/group. Scale bar, 1 μm. White arrows indicate S. uberis in mMECs. Panels C to G are representative of 3 independent experiments. Data are presented as means ± SEM (n = 6, unless otherwise indicated). *, P < 0.05, i.e., significantly different between the indicated groups by unpaired t test.

The increase in LDs is related to lipid synthesis, lipid catabolism, and fatty acid absorption (27). However, S. uberis infection lowered expression of genes associated with lipid catabolism, such as Acox1 and Ehhadh (Fig. 1F), and had little effect on Cd36 levels (Fig. 1G), associated with fatty acid transportation. Similar results were obtained with bovine mammary epithelial cells (MAC-T cells) (see Fig. S2 in the supplemental material). Collectively, these biological events in S. uberis infection are associated with the accumulation of LDs in mammalian cells.

LDs contribute to S. uberis burden.

Limiting the ability of obligate intracellular pathogens to acquire specific nutrients inhibits their growth or induces a dormant state (28). Because LDs directly interacted with bacteria (Fig. 1H), we hypothesized that host cell LDs provided beneficial conditions for S. uberis growth. To confirm this, mMECs were treated with C75 (FASN inhibitor) to inhibit fatty acid synthesis or incubated with oleic acid to stimulate LD formation. These drugs had no obvious effect on the growth of bacteria within 3 h (Fig. S1). We found that S. uberis burden decreased (Fig. 2A and B) after C75 treatment inhibition of LD formation in host cells (Fig. 2E and F). Oleic acid supplementation increased LD formation (Fig. 2E and H), stimulating S. uberis growth (Fig. 2A and C).

FIG 2.

Taurine reduction of LDs inhibits the burden of S. uberis. mMECs were treated with 10 μM C75 for 12 h to inhibit fatty acid synthesis, incubated with 360 μM oleic acid for 24 h or 45 mM taurine for 24 h, and then infected with S. uberis in mid-exponential phase (MOI = 10) for 3 h at 37°C. (A) Viable bacteria enumerated as CFU on THB agar. CFU were counted by the spread plate method after incubation for 12 h at 37°C. The efficacy of taurine, C75, or oleic acid on the elimination of S. uberis was confirmed by culturing mMEC lysates on THB agar plates. (B and C) The number of S. uberis colonies in mMECs. Data are presented as means ± SD (n = 6). *, P < 0.05, i.e., significantly different between the indicated groups by one-way ANOVA. (D) Representative flow cytometry histogram of FLI-H channel at 3 h postinfection (hpi) in FITC–d-Lys-S. uberis-infected mMECs of the indicated groups. The dotted line indicates background fluorescence in uninfected mMECs. Data are presented as means ± SEM (n = 6). *, P < 0.05, i.e., significantly different between the indicated groups by one-way ANOVA. (E) BODIPY 493/503 staining for LD visualization (green) in different groups. mMECs were fixed with 4% paraformaldehyde and stained with BODIPY. Images are representative of 10 samples/group. Scale bar, 20 μm. (F and H) Flow cytometry of BODIPY-stained mMECs. (Left) Representative histograms. (Right) Mean fluorescence intensities from 3 experiments. The dotted line indicates background fluorescence in uninfected mMECs. Data are presented as means ± SEM (n = 6). *, P < 0.05, i.e., significantly different between the indicated groups by one-way ANOVA. (G) Schematic diagram of flow cytometry to assess bacterial burden. Experiments were conducted in triplicate.

Metabolomic analysis showed that taurine reduced lipid metabolites (unpublished data). We speculated that taurine exerted its antibacterial effects by interfering with lipid metabolism. First, we determined that taurine had no significant effect on the growth of S. uberis in Fig. S1. We found that taurine reduced intracellular bacterial load (Fig. 2A to C) and diminished LD accumulation (Fig. 2E to H). Flow cytometric data indicated that taurine treatment decreased S. uberis-derived fluorescein isothiocyanate (FITC)–d-Lys and LDs (Fig. 2G and D). Similar results occurred with MAC-T cells, supporting the hypothesis that LD availability affects S. uberis loading (Fig. S3 and S4).

Taurine decreases LDs independently of sterol response binding protein 1 (SREBP1).

LDs participate in homeostatic lipid metabolism. We speculated that taurine-associated LDs were a part of this metabolism. Hence, we detected genes related to lipid metabolism pathways (29, 30), as metabolism and the synthesis of LDs are inseparable (31). Taurine effectively reduced ACC activity (Fig. 3A) and inhibited gene expression related to lipid synthesis (Fig. 3B). Taurine had no obvious effect on genes associated with lipid catabolism (Fig. 3C) and absorption (Fig. 3D).

FIG 3.

SREBP1 is not reliant on taurine to decrease LDs. mMECs were treated with 5 μM fatostatin for 24 h to inhibit SREBP1 expression and then infected with S. uberis in mid-exponential phase (MOI = 10) for 3 h at 37°C. (A) ACC enzyme activities. (B) Fasn, Scd1, and Srebf1 mRNA relative expression in mMECs during S. uberis invasion. (C) Acox1 and Ehhadh mRNA relative expressions. (D) Cd36 mRNA relative expression. (E) Immunofluorescence staining was performed for SREBP1 (red) between the control group and stimulated S. uberis group as well as the taurine pretreatment group. Images are representative of 10 samples/group. Scale bar, 20 μm. (F) The protein expression levels of SREBP1 and P-ACC with or without taurine or fatostatin during S. uberis infection were determined by Western blotting in mMECs. For quantitative analysis, bands were evaluated densitometrically with ImageJ analyzer software normalized for β-actin or ACC density. (G) Flow cytometry of BODIPY-stained mMECs. (Left) Representative histograms. (Right) Mean fluorescence intensities from 3 experiments. The dotted line indicates background fluorescence in uninfected mMECs. (H) BODIPY 493/503 staining for LD visualization (green). Images are representative of 10 samples/group. Scale bar, 20 μm. Experiments whose results are shown in panels A to G were repeated 3 times. Data are presented as the means ± SEM (n = 6, unless otherwise indicated). *, P < 0.05, i.e., significantly different between the indicated groups by one-way ANOVA.

To further explore the molecular mechanism of taurine activity, we studied the expression of SREBP1, a key regulatory transcription factor for lipid synthesis (32). To explore the role of SREBP1 in this process, fatostatin was added to inhibit SREBP1 expression. Immunofluorescent staining and Western blotting revealed that taurine significantly reduced SREBP1 expression with about a 1.3-fold change in protein levels (Fig. 3E and F). A similar result occurred in relation to ACC phosphorylation, whose level was enhanced obviously (more than 1.5-fold change) (Fig. 3F). Moreover, fatostatin, an inhibitor of both SREBP1 and SREBP2 (33), had an effect similar to that of taurine (Fig. 3A, B, E, and F).

Inhibition of SREBP1 did not reduce LD content; it significantly increased LD levels (Fig. 3G and H) as well as the S. uberis load in cells (Fig. S5), though it had no direct effect on the growth of bacteria in 3 h (Fig. S1). Although fatostatin had little effect on the genes involved in lipid catabolism (Fig. 3C), it robustly increased Cd36 expression (Fig. 3D). It is possible that by restraining de novo fatty acid synthesis, mMECs may take up exogenous fatty acids via Cd36. These results indicate that taurine does not reduce LDs through inhibition of SREBP1.

Taurine has anti-inflammatory potential via nuclear liver X receptor (LXR).

Given the known role of SREBP1 in inflammatory factor production (34, 35), we asked whether taurine relieved inflammation by this mechanism. Expression of inflammatory genes Cxcl9 and Nos2 increased in response to suppression of SREBP1 (Fig. 4A). The relative expression (Fig. 4B) and levels (Fig. S6) of inflammatory factors, including tumor necrosis factor alpha (TNF-α), interleukin 1β (IL-1β), and IL-6, further supported this conclusion. Oishi et al. have shown that SREBP1 produces unsaturated fatty acids that reduce the expression of inflammatory genes (36). As a result of addition of taurine or inhibition of SREBP1, Fads1 and Scd2 (related to encoding unsaturated fatty acids) increased significantly, resulting in a stronger anti-inflammatory effect (Fig. 4C).

FIG 4.

Taurine improves LXR-β-mediated anti-inflammation by enhancing unsaturated fatty acid gene expression. mMECs were treated with 5 μM fatostatin for 24 h to inhibit SREBP1 expression or 1 μM GW3965 for 24 h to activate LXRs and then infected with S. uberis in mid-exponential phase (MOI = 10) for 3 h at 37°C. (A) Cxcl9 and Nos2 mRNA expressions are related to inflammation. (B) Cytokine (TNF-α, IL-1β, and IL-6) levels in cell supernatants. (C) Fads1 and Scd2 mRNA expression relative to anti-inflammatory unsaturated fatty acid synthesis. (D) Lxra and Lxrb mRNA expression. (E) Abca1 mRNA expression. (F) Protein expression levels of LXR-β and ABCA1 with or without GW3965 during S. uberis infection determined by Western blotting in mMECs. For quantitative analysis, bands were evaluated densitometrically with ImageJ analyzer software normalized for β-actin density. (G) Fads1 and Scd2 mRNA expression. (H) Cxcl9 and Nos2 mRNA expression. All experiments were repeated 3 times. Data are presented as means ± SEM (n = 6). *, P < 0.05, i.e., significantly different between the indicated groups by one-way ANOVA.

LXR is upstream of SREBP1 (37) and primarily functions to synthesize anti-inflammatory fatty acids (38, 39). In the current study, taurine increased Lxra and Lxrb expression (Fig. 4D). Consistent results were obtained for Abca1, a gene downstream of Lxrs (Fig. 4E). Western blotting data indicated that taurine significantly increased LXR-β (nearly a 2-fold change) and downstream ABCA1 expression (around a 1.65-fold change), in a fashion similar to that for LXR agonist (Fig. 4F). Activation of LXR genes encoding unsaturated fatty acids increased significantly (Fig. 4G), while inflammatory genes decreased accordingly (Fig. 4H). Taken together, our results indicate that the anti-inflammatory potential of taurine is dependent on LXR upregulating gene expression relative to unsaturated fatty acid levels.

Taurine increases LDLR expression via IDOL.

LXR agonist increased SREBP1 expression with about a 1.13-fold change (Fig. 5A), in contrast to the effect of taurine (over a 1.5-fold change). The changes in ACC protein level also confirmed this result (Fig. 5A). Two proteins, IDOL and LDLR, communicate upstream with LXR and SREBP1, respectively, to jointly stabilize lipid metabolism (40, 41). Taurine significantly decreased Idol (also known as Mylip) and had little effect on Ldlr gene expression (Fig. 5B). Different infection times (1 to 3 h) showed no differences in Ldlr transcripts, but there was lower Idol gene expression with taurine (Fig. 5C).

FIG 5.

Taurine inhibits IDOL expression enhancing LDLR levels. mMECs were treated with 5 μM fatostatin for 24 h to inhibit SREBP1 expression or 1 μM GW3965 to for 24 h activate LXRs and then infected with S. uberis in mid-exponential phase (MOI = 10) for 3 h at 37°C. (A) Protein expression levels of SREBP1 and P-ACC, with or without GW3965 during S. uberis infection, were determined by Western blotting in mMECs. For quantitative analysis, bands were evaluated densitometrically with ImageJ analyzer software and normalized for β-actin or ACC density. (B) Idol and Ldlr mRNA expression. (C) Changes in Idol and Ldlr gene expressions at different incubation times with S. uberis. Experiments were repeated 3 times. Data are presented as means ± SEM (n = 6). *, P < 0.05, i.e., significantly different from the 0-h group; #, P < 0.05, i.e., significantly different between the S. uberis group and taurine-plus-S. uberis group at the same time points by two-way ANOVA. (D) Idol and Ldlr mRNA expression with or without GW3965 and taurine. (E) Idol and Ldlr mRNA expression with or without fatostatin and taurine. (F) Protein expression levels of IDOL and LDLR with or without GW3965 and fatostatin during S. uberis infection were determined by Western blotting in mMECs. For quantitative analysis, bands were evaluated densitometrically with ImageJ analyzer software and normalized for β-actin density. Experiments whose results are shown in panels A, B, and D to F were repeated 3 times. Data are presented as means ± SEM (n = 6). *, P < 0.05, i.e., significantly different between the indicated groups by one-way ANOVA.

LXR agonist significantly increased Idol gene expression; taurine reduced this change (Fig. 5D). Correspondingly, SREBP1 inhibitor (fatostatin) decreased Ldlr expression, but taurine had little effect on this gene (Fig. 5E). It is possible that a decrease in Idol has no significant inhibitory effect on Ldlr mRNA levels, due to IDOL suppression of LDLR posttranscription (17). Western blotting results (Fig. 5F) confirmed this supposition and suggested that taurine increases the expression of LDLR (around a 1.3-fold change) by inhibiting IDOL (about a 2-fold change).

IDOL decreases S. uberis load by reducing LDs.

To demonstrate whether IDOL is a target protein for this process, specific interference RNA (siRNA) was used to significantly reduce IDOL expression (Fig. 6A). Compared with the infection group, siIdol increased LDLR (about a 1.3-fold change) and LXR (over a 2-fold change) and significantly decreased SREBP1 (around a 1.4-fold change) expression (Fig. 6B). Idol inhibition reduced expression of genes related to lipid synthesis (Fig. 6C). Genes involved in fatty acid oxidation (Fig. 6D) and Cd36 (Fig. 6E) did not significantly change. Furthermore, data from flow cytometry (Fig. 6F) and fluorescence microscopy (Fig. 6G) showed that LDs diminish significantly after interference from Idol. Similar results occurred with MAC-T cells (Fig. S3 and S4). Since LXR protein increased, siIdol groups had a potential anti-inflammatory function (Fig. 6H and I). Interference by Idol also significantly reduced the S. uberis load (Fig. 6J and K). Together, these results indicate that inhibition of IDOL to reduce LDs restricts S. uberis burden.

FIG 6.

Inhibiting IDOL limits LD formation restricting S. uberis loading. mMECs were transfected with 50 nM siIdol for 48 h at 37°C using Lipofectamine 3000 reagent (Invitrogen) to inhibit Idol expression and then infected with S. uberis in mid-exponential phase (MOI = 10) for 3 h at 37°C. (A) Idol mRNA expression in control and interference groups. (B) Protein expression levels of LDLR, SREBP1, LXR-β, and P-ACC with or without siIdol during S. uberis infection were determined by Western blotting in mMECs. For quantitative analysis, bands were evaluated densitometrically with ImageJ analyzer software and normalized for β-actin or ACC density. (C) Fasn, Scd1, and Srebf1 mRNA relative expressions. (D) Acox1 and Ehhadh mRNA expressions. (E) Cd36 mRNA expression. (F) Flow cytometry of BODIPY-stained mMECs. (Left) Representative histograms. (Right) Mean fluorescence intensities from 3 experiments. The dotted line indicates background fluorescence in uninfected mMECs. Data are presented as means ± SEM (n = 6). *, P < 0.05, i.e., significantly different between the indicated groups by one-way ANOVA. (G) BODIPY 493/503 staining for LD visualization (green). Images are representative of 10 samples in each group. Scale bar, 20 μm. (H) Fads1 and Scd2 mRNA relative expressions. (I) Cxcl9 and Nos2 mRNA relative expressions. (J) The number of S. uberis colonies in mMECs after interference with Idol on THB agar plates. Data are presented as means ± SD (n = 6). *, P < 0.05, i.e., significantly different compared with the S. uberis groups by one-way ANOVA. (K) Representative flow cytometry histogram of FLI-H channel at 3 hpi in FITC–d-Lys-S. uberis-infected mMECs of the indicated groups. Data are presented as means ± SEM (n = 6). #, P < 0.05, i.e., significantly different between the treated versus untreated groups; *, P < 0.05, i.e., significantly different between the mock versus siIdol groups by two-way ANOVA. Experiments whose results are shown in panels A to F and H to K were repeated 3 times. Data in panels A to F and H and I are presented as means ± SEM (n = 6). *, P < 0.05, i.e., significantly different between the indicated groups by one-way ANOVA.

DISCUSSION

Host LD increase is a well-known metabolic change during pathogen invasion (42, 43). Here, we demonstrate that IDOL mediates LD and lipid metabolism remodeling in a specific bacterial infection. Inhibition of IDOL restricts fatty acid synthesis and host LD increase, attenuating bacterial load. Moreover, modulating lipid metabolism limits bacterial access to host resources, thus restricting pathogen growth. This is consistent with evidence that shows that LDs play an important role in support of bacterial pathogenesis (44). These data provide a rationale for metabolic defense as a putative therapy for S. uberis as well as other intracellular pathogens.

Although our work focuses on S. uberis, an intracellular bacterium, competition between hosts and pathogens for limited nutrients is a process common to most intracellular infections. LDs are intracellular lipid reservoirs (45), and many lipid synthesis enzymes (e.g., ergosterol) localize to them, suggesting a close association between lipid synthesis pathways and LDs (46). Infection-driven LD biogenesis is a complex process involving innate immune receptors, transcriptional and posttranscriptional regulation, increased lipid uptake, and lipid synthesis. Accumulating evidence demonstrates that intracellular pathogens are able to exploit LDs as an energy source, a replication site, or a place to escape immune responses (47). The most well-known example of LD interactions with a pathogen is that of hepatitis C virus, whose core protein localizes to LDs (48). Here, we show that S. uberis significantly increases LDs in mammalian cells; via electron microscopy, we found that S. uberis directly interacts with LDs. Inhibition of fatty acid synthesis effectively decreases the intracellular bacterial load. These findings are consistent with the fact that many bacteria, such as Chlamydia trachomatis (49) and Mycobacterium tuberculosis (50), complete their life cycle in host LDs. Notably, in this study, the change of the colony number did not show a significant exponential change, usually a linear relationship. It is possible that compared with the complex environment of the organism, the single-culture condition in the laboratory has certain influence on the biological characteristics of pathogens (51), as the mastitis model in vivo constructed in our laboratory also presents a significant antibacterial effect (11, 52).

In noncancerous cells, LDLR expression is suppressed by high cellular cholesterol levels via inactivation of SREBPs and activation of the LXR-IDOL axis (53). Besides cholesterol, LDL also contains Apo B-100, fatty acids, and phospholipids (54), suggesting that this signaling pathway is also involved in metabolic lipid stabilization. S. uberis increases SREBP1 expression and upregulates fatty acid synthesis. Taurine significantly reduces SREBP1 and LDs by increasing LDLR abundance by inhibiting IDOL. It seems likely that high LDLR expression reduces SREBP1 levels via a negative-feedback inhibition and reducing enhanced fatty acid synthesis, as there are many negative-feedback mechanisms in cells (55). Additionally, by lowering IDOL, the expression of LXR and its downstream ABCA1 (cholesterol transporter) increases. It is possible that the cells export cholesterol and fatty acids by this method (56), but further study is warranted.

SREBP1 mediates many inflammatory pathways (34, 35). Expression of inflammatory genes and inflammatory factors increases significantly after inhibition of SREBP1. Next, we detected the expression levels of Fads1 and Scd2 related to the synthesis of unsaturated fatty acids. These genes have been shown to promote the synthesis of unsaturated fatty acids, thereby inhibiting inflammatory pathways like NF-κB and showing an anti-inflammatory effect (36). Because the expression of these genes also significantly increases, we studied the LXR protein next. It lies upstream of SREBP1, regulates many enzymes involved in unsaturated fatty acid synthesis (57), and inhibits inflammation by antagonizing proinflammatory transcription factors, such as NF-κB (58). We speculate that taurine may also promote the genes related to the synthesis of unsaturated fatty acids by activating LXR, thereby exerting an anti-inflammatory effect, although Dai et al. have demonstrated that taurine has anti-inflammatory effects by regulating the phospholipase C (PLC)-g1-protein kinase C a-Ca²⁺-NF-κB/nuclear factor of activated T cells (NFAT) signaling pathway during S. uberis infection (59). Unlike LXR agonists that increase SREBP1 expression, taurine seems to increase LXR expression by inhibiting IDOL. Consistent with this, much drug-induced relief occurs through negative feedback on signal pathways, and they are viewed as major targeted therapies (60, 61).

Our laboratory discovered that taurine activates adenosine 5′-monophosphate (AMP)-activated protein kinase, reducing energy metabolism, including the glycolysis pathway and tricarboxylic acid (TCA) cycle (unpublished data). Notably, the production of citric acid in the tricarboxylic acid cycle is involved in fatty acid synthesis (62). Glucose delivers carbon for various lipid classes (63, 64). An ATP and pyruvate milieu triggers fatty acid biosynthesis and LD formation (65). In the current study, we found that during S. uberis infection, IDOL is a target protein that functions to reduce lipid synthesis. Although the relationship between lipid metabolism, glycolysis, and TCA is unclear, rescuing the aberrant metabolism of the host may be a therapeutic aid for creating disease resistance.

In summary, our findings provide evidence for a role of taurine in reducing LDs and thus decreasing bacterial load via inhibition of IDOL (Fig. S8). Taurine is an important regulator of lipid metabolism that opens a number of therapeutic opportunities; while an LD decrease leads to bacterial load reduction, the mechanism(s) by which this occurs remains to be determined. The competition between microbes and host cells for metabolites opens the possibility that modulation of host metabolism may be able to limit pathogen access to nutrients and ultimately control infection.

MATERIALS AND METHODS

Bacterial strain, cell culture, and treatment.

S. uberis (strain 0140J) was inoculated into Todd-Hewitt broth (THB) and incubated at 37°C in an orbital shaker to log-phase growth (optical density at 600 nm [OD₆₀₀] = 0.4 to 0.6).

mMECs were grown at 37°C in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum (FBS; Gibco, USA) in 6-well plates. After they reached 70 to 80% confluence, the monolayers were treated or not with 45 mM taurine (Sangon Biotech, Shanghai, China) for 24 h, 5 μM fatostatin (inhibitor of SREBPs; SellecK Chemicals, Houston, TX) for 24 h, 1 μM GW3965 (agonist of LXRs; SellecK Chemicals) for 24 h, 360 μM oleic acid (Sigma-Aldrich, USA) for 24 h, or 10 μM C75 (inhibitor of fatty acid synthase; Sigma-Aldrich) for 12 h, all at 37°C, or transfected with 50 nM siIdol (Mylip) for 48 h at 37°C using Lipofectamine 3000 reagent (Invitrogen). Transfection reagents and siRNA (siIdol) were purchased from Guangzhou Ruibo Biotechnology Co., Ltd. (Guangzhou, Guangdong, China). The sequences of siRNA were designed and listed as follows: siIdol for mMECs, CCACACCAGTCTCCTCAAT. The interference of the Idol gene was identified by quantitative real-time PCR (qPCR) (Fig. 6A) for cell models.

MAC-T cells were incubated in DMEM with 10% FBS, plated at 70 to 80% confluence into 6-well plates, and then treated according to the different test conditions with siIdol (Mylip) for MAC-T cells (CCGAATACCAAGTGTTGCA). The interference of Idol gene was identified by fluorescence quantitative PCR (Fig. S7) for cell models.

The treated cells were infected with S. uberis at a multiplicity of infection (MOI) of 10 for 1 h, 2 h, or 3 h at 37°C according to different test conditions. The supernatant and cells were collected separately and stored at −80°C until use.

Determination of S. uberis strain growth curves.

Bacterial growth was measured as described previously (66). Briefly, S. uberis was cultured in THB until the OD₆₀₀ reached 0.4 to 0.6. Then an equal amount of each bacterial culture was transferred into 100 ml of THB with 45 mM taurine, 5 μM fatostatin, 1 μM GW3965, 360 μM oleic acid, and 10 μM C75 at a ratio of 1:100 (vol/vol) and incubated at 37°C and 20 × g. The OD₆₀₀ of bacteria was monitored at 1-h intervals (Fig. S1).

RNA extraction and qPCR.

PCR was conducted as previously described (11). Total RNA was extracted by TRIzol reagent (TaKaRa, Dalian, China). Corresponding cDNA was obtained using reverse transcriptase and oligo(dT) 18 primer (TaKaRa). An aliquot of the cDNA was mixed with 25 μl of SYBR green PCR master mix (TaKaRa) and 10 pmol of each specific forward and reverse primer. All mixed systems were analyzed in an ABI Prism 7300 sequence detection system (Applied Biosystems, Waltham, MA). Fold changes were calculated as threshold cycle (2^−ΔΔCT) values. All primer sequences (Table S2) were synthesized by Tsinke Company (Beijing, China).

Total protein extraction and Western blotting.

Intracellular protein levels were determined by Western blotting. An anti-β-actin antibody (Bioworld, USA) was used as a loading control. Cells were washed twice in ice-cold phosphate-buffered saline (PBS) and lysed by incubation in radioimmunoprecipitation assay (RIPA) buffer (Beyotime, Nantong, China) containing phenylmethylsulfonyl fluoride (PMSF; Beyotime) on ice for 30 min. Supernatants were collected by centrifugation at 5,000 × g for 10 min at 4°C, and protein concentrations were determined using a bicinchoninic acid assay kit (Beyotime) and detected with a spectrophotometer (Tecan, Männedorf, Switzerland). Proteins were separated by electrophoresis on a polyacrylamide gel and transferred to polyvinylidene fluoride membranes (Millipore, USA). The membranes were blocked with 5% nonfat milk diluted in Tris-buffered saline with Tween 20 (TBST) for 2 h at room temperature (approximately 10 to 25°C) and hybridized overnight with an appropriate primary antibody at 4°C. The following primary antibodies were diluted in TBST: β-actin (1:10,000), SREBP1 (1:1,000), ACC (1:1,000), phosphorylation-ACC (P-ACC) (1:1,000), LXR-β (1:1,000), LDLR (1:1,000), and IDOL (1:1,000). The membranes were washed 3 times with TBST before and after incubation with horseradish peroxidase (HRP)-linked anti-rabbit IgG (CST, MA; 1:10,000) secondary antibody at room temperature (approximately 10 to 25°C) for 2 h. Signals were detected using an ECL Western blot analysis system (Tanon, Shanghai, China). Bands were quantified using ImageJ software (NIH, USA).

Lipid droplet staining and counting.

Lipid droplets were evaluated by staining mMECs and MAC-T cells with BODIPY 493/503, a neutral lipid dye, to facilitate quantification of neutral lipid content by flow cytometry (67). Briefly, cells were quickly rinsed using 3 ml of PBS to remove medium. After incubation with 2 μM BODIPY in the dark for 15 min at 37°C, cells were washed 3 times in PBS and detached. The cells were centrifuged at 400 × g for 5 min, resuspended in PBS, and immediately analyzed by flow cytometry using FACSCanto (BD, NJ). Ten thousand cells per sample were analyzed using CellQuest Pro acquisition and FlowJo software.

BODIPY staining and microscopy.

Coverslips were washed with 3 ml of PBS 3 times and fixed with 4% formaldehyde for 15 min at room temperature. They were again washed 3 times with PBS after incubation with 2 μM BODIPY in the dark for 15 min at 37°C. They were then incubated with 10 μM DIPY in the dark for 20 min at room temperature and washed 3 times for 5 min in PBS. Finally, the plate was flooded with an antifluorescence quenching sealer and observed and photographed with a laser confocal microscope (67).

ELISA.

TNF-α, IL-1β, and IL-6 levels in mMECs were quantified using enzyme-linked immunosorbent assay (ELISA) kits (mouse TNF-α ELISA kit, mouse IL-1β ELISA kit, and mouse IL-6 ELISA kit) according to the manufacturer’s instructions (Rigor Bioscience, Beijing, China). Briefly, prepared standards (50 μl) were reacted for 60 min at 37°C with HRP-linked antibodies (40 μl), and each microplate was washed 5 times with the wash solution included in the kit. Chromogen solutions A (50 μl) and B (50 μl) were added the wells and incubated for 10 min at 37°C. Stop solution (50 μl) was then added to the wells, and ODs were measured at a wavelength of 450 nm within 10 min after addition of stop solution. TNF-α and IL-1β levels were expressed as nanograms per gram of protein. Qualitative differences or similarities between the control and experimental groups were consistently observed throughout the study.

Detection of relative enzyme activities.

The activities of ACC in mMECs were measured by commercial kits (ACC assay kit) according to the manufacturer’s instructions (Solarbio, Beijing, China).

Viable bacterial count assay.

Viable bacteria were enumerated as CFU on THB agar. CFU were counted by the spread plate method after incubation for 12 h at 37°C.mMECs and MAC-T cells with or without taurine, fatostatin, GW3965, siIdol, and C75 were incubated in DMEM with 10% FBS and plated at 80% confluence in 6-well plates. After infection with S. uberis for 3 h at mid-exponential phase (OD₆₀₀ = 0.4 to 0.6), S. uberis-infected cells were washed 3 times with PBS containing 100 mg/ml of gentamicin, followed by gentamicin-free PBS. Cells were pelleted at 1.4 × g for 10 min. Equal numbers of cells were lysed with sterile triple-distilled water, and CFU were counted by the spread plate method after incubation for 12 h at 37°C (52).

Flow cytometric analysis of bacterial burden.

S. uberis was grown at 37°C in THB medium until reaching an OD₆₀₀ of 0.6. The medium was diluted to an OD₆₀₀ of 0.3 with fresh medium containing 0.1 mM FITC–d-Lys (Shengguang, Xiamen, China). The diluted bacteria were further incubated at 37°C until achieving an OD₆₀₀ of 0.4 to 0.6. The bacteria were centrifuged, washed with THB medium 3 times, and resuspended in cell culture medium. Cells were then incubated in 20 μM FITC–d-Lys (a chemical biology approach that enables rapid and covalent incorporation and detection of a fluorescently reprivatized peptidoglycan component during cell wall synthesis in real time) for 3 h at 37°C, washed 3 times in PBS, and detached with trypsin. Then cells were centrifuged at 400 × g for 5 min, resuspended in PBS, and immediately analyzed by flow cytometry using a FACSCanto instrument; 10,000 cells/sample were analyzed using CellQuest Pro acquisition software and FlowJo software.

Immunofluorescence staining.

Cell-filled slides were washed 3 times with PBS, fixed with 4% formaldehyde for 15 min, and washed an additional 3 times with PBS, and cell membranes were permeated with 0.5% PBS with Tween 20 (PBST) for 20 min. They were washed 3 more times with PBS before blocking with 5% goat serum for 1 h at room temperature. They were again washed 3 times with PBS and incubated with diluted primary antibody (1:100) overnight at 4°C. After 3 washings with PBS, secondary antibody (1:10,000) was added and the slides were incubated for 2 h at room temperature. They were again washed 3 times with PBS, and nuclei were stained with 10 μM 4′,6-diamidino-2-phenylindole (DAPI) for 20 min. After 3 washings with PBS, the plate was flooded with an antifluorescence quenching sealer and examined and photographed with a laser confocal microscope.

Transmission electron microscopy.

Cells were trypsinized and harvested, washed 3 times with PBS, fixed with a buffer containing 2.5% glutaraldehyde for 24 h, and postfixed in 1% osmium tetroxide for 2 h. The cells were dehydrated in a graded ethanol series, washed with propylene oxide, and encased in embedding medium. The samples were sectioned on an ultramicrotome at a thickness of 90 nm. The ultrathin sections were stained with uranyl acetate and lead citrate. Images were obtained using a transmission electron microscope (FEI T12; FEI, USA) at 80 kV.

Statistical analyses.

Statistical analyses were performed using GraphPad Prism 8 software. Sample numbers and repetitions are indicated in the figure legends. All data were analyzed using unpaired t test, one-way analysis of variance (ANOVA), or two-way ANOVA, as indicated in the figure legends. All data are presented as means ± standard deviations (SD) or standard errors of the means (SEM). For all experiments, P values of <0.05 were considered significant.