Lactate promotes Salmonella intracellular replication and systemic infection via driving macrophage M2 polarization

ABSTRACT Salmonella is one of the most important enteric pathogens worldwide, which is able to cause lethal systemic infection via survival and replication in host macrophages. Lactate, a byproduct of anaerobic or aerobic glycolysis, can induce macrophage M2 polarization, but the relationship between lactate-mediated macrophage M2 polarization and bacterial infection is poorly understood. Here, we evaluated the role of lactate and lactate-mediated macrophage M2 polarization in the pathogenicity of Salmonella. We found that lactate levels were significantly increased in Salmonella-infected macrophages, and the increased lactate was derived from the host. Macrophage and mouse infection assays showed that the addition of lactate enhanced Salmonella replication within macrophages and the colonization of mouse systemic loci, while pharmacological or genetic inhibition of host lactate production impaired Salmonella intracellular replication and its virulence in mice. Further analysis revealed that lactate promotes M2 polarization of Salmonella-infected macrophages, and the induction of macrophage M2 polarization by lactate is responsible for lactate-mediated Salmonella growth promotion. Moreover, we showed that macrophage-derived lactate induces the Salmonella pathogenicity island (SPI)-2 type III secretion system, leading to increased translocation of the SPI-2 effector SteE, which is responsible for driving M2 polarization. Overall, these findings suggest that lactate promotes Salmonella intracellular replication and systemic infection via driving macrophage M2 polarization and highlight the complex interactions between Salmonella and macrophages. IMPORTANCE The important enteropathogen Salmonella can cause lethal systemic infection via survival and replication in host macrophages. Lactate represents an abundant intracellular metabolite during bacterial infection, which can also induce macrophage M2 polarization. In this study, we found that macrophage-derived lactate promotes the intracellular replication and systemic infection of Salmonella. During Salmonella infection, lactate via the Salmonella type III secretion system effector SteE promotes macrophage M2 polarization, and the induction of macrophage M2 polarization by lactate is responsible for lactate-mediated Salmonella growth promotion. This study highlights the complex interactions between Salmonella and macrophages and provides an additional perspective on host-pathogen crosstalk at the metabolic interface.

Macrophages are innate immune cells constituting the first line of defense against an infection (16).The physiological state of macrophages is plastic and regulated by microenvironmental stimuli (17).Depending on the activation state, the macrophages are divided into classically activated M1 macrophages and alternatively activated M2 macrophages, which generally exert host defense and tissue repair functions, respectively (18).After stimulation by pathogen-associated molecular patterns [such as lipopolysaccharide (LPS)] and/or by interferons (such as IFN-γ), macrophages are polarized to M1 phenotype, which produce high levels of nitric oxide and proinflamma tory cytokines against microbial infection.In contrast, stimulation of macrophages using cytokines such as interleukin (IL)-4 and IL-13 promotes antiinflammatory response, and macrophages are polarized to the M2 phenotype to promote tissue repair (19,20).Salmonella mainly resides and grows in M2 macrophages rather than in M1 macrophages (21)(22)(23).Moreover, Salmonella can actively promote macrophage M2 polarization by activating signal transducers and activators of transcription 3 (STAT3), via the SPI-2 T3SS effector SteE (24,25).
Lactate or lactic acid, a byproduct of anaerobic or aerobic glycolysis (the Warburg effect), is produced from the conversion of pyruvate by lactate dehydrogenase A (LDHA) (26,27).During bacterial infection, lactate production is significantly increased in infected macrophages, due to the reprogramming of cellular glucose metabolism from oxidative phosphorylation (OXPHOS) to aerobic glycolysis (28).Several intracellu lar bacterial pathogens, such as Mycobacterium tuberculosis and Brucella abortus, can exploit macrophage-derived lactate as a nutrient to grow and survive in the intracellular niche (29,30).Although lactate levels are increased in Salmonella-infected macrophages, lactate is not exploited by intracellular Salmonella as a nutrient; instead, Salmonella senses host-derived lactate as a host cue to induce its SPI-2 T3SS expression (31).However, there is still no direct evidence to support that lactate promotes Salmonella replication in macrophages.
Emerging evidence suggests that lactate can promote macrophage polarization to M2 phenotype, via mechanisms such as activation of hypoxia-inducible factor-1α and ERK/STAT3 signaling pathway (32,33).Given that Salmonella mainly grows in M2 macrophages, it is unclear whether lactate can promote Salmonella replication in macrophages by promoting macrophage M2 polarization.Both lactate and Salmonella can promote macrophage M2 polarization, but the relationship between lactate-and Salmonella-mediated M2 polarization is unknown.
In this study, we aimed to investigate the effects of lactate and lactate-mediated macrophage M2 polarization on Salmonella intracellular replication and pathogenic ity.Through in vitro and in vivo infection assays along with seahorse analysis, flow cytometry, and many other molecular techniques, we report that macrophage-derived lactate promotes Salmonella replication within macrophages and colonization of mouse systemic loci.During Salmonella infection, lactate via the Salmonella SPI-2 T3SS effector SteE promotes macrophage M2 polarization, and the induction of macrophage M2 polarization by lactate is responsible for lactate-mediated Salmonella growth promotion.Our findings illustrate the important role of host-derived lactate in Salmonella pathoge nicity and highlight the complex interactions between Salmonella and macrophages.

Lactate promotes Salmonella replication within macrophages
Previous studies have reported that the lactate levels of macrophages were increased in response to Salmonella infection (31), which was also confirmed here.At 20 h post-infec tion with wild-type Salmonella (Salmonella enterica serovar Typhimurium ATCC 14028s, WT), the lactate levels of mouse primary bone-marrow-derived macrophages (BMDMs) increased 4.8-fold relative to those of uninfected BMDMs (Fig. 1A).Lactate levels in WT-infected BMDMs are comparable to that of the BMDMs infected with three mutant strains, ΔlldD (cannot produce L-lactate), ΔldhAΔdld (cannot produce D-lactate), and ΔlldDΔldhAΔdld (cannot produce D-and L-lactate) (Fig. 1B and C), indicating that the increased lactate in infected BMDMs is derived from the macrophages.
We analyzed the effect of lactate on Salmonella replication in macrophages.The replication of Salmonella in BMDMs was significantly (P < 0.01) increased with the addition of 3 mM lactate (Fig. 1D).Immunofluorescence analysis showed that the addition of lactate did not influence the number of bacteria in each infected BMDM at initial infection stage (2 h) but increased the number of bacteria in each infected BMDM at 20 h post-infection (Fig. 1E).When we inhibited lactate production in BMDMs by adding the competitive inhibitor FX11 (23.3 µM) of LDHA, Salmonella replication decreased 2.2-fold, and lactate addition relieved the bacterial replication defect medi ated by FX11 (Fig. 1F).siRNA knockdown of LDHA also decreased Salmonella replication in mouse RAW264.7 macrophage cell line, and lactate addition relieved the bacterial replication defect mediated by ldhA siRNA (Fig. 1G).These results collectively indicate that macrophage-derived lactate promotes Salmonella replication within macrophages.

Lactate promotes Salmonella colonization of systemic loci in mice
To assess whether lactate contributes to Salmonella infection in vivo, we injected Salmonella-infected mice with lactate (0.6 mg lactate/g mice) or phosphatebuffered saline (PBS) (control).The livers and spleens of infected mice were collected on day 3 post-infection to quantify the bacterial burden.The results showed that lactate injection significantly (P < 0.01) increased the bacterial burden in the mouse liver and spleen (Fig. 2A).Moreover, the bacterial burden was significantly (P < 0.01) less in the livers of liver specific LDHA knockout mice (LDHA −/− ) compared to that in C57BL/6 wild-type mice (Fig. 2B).In contrast, the bacterial burden in the spleens of LDHA −/− mice is comparable to that of wild-type mice (Fig. S1), suggesting that the decreased bacterial colonization in LDHA −/− mice is a liverspecific phenotype, as a result of ldhA knock out in the liver.These results indicate that lactate promotes Salmonella colonization of systemic loci in mice.

Lactate promotes M2 polarization of Salmonella-infected macrophages
As lactate can promote macrophage M2 polarization, we then assessed whether lactate promotes M2 polarization of Salmonella-infected macrophages.Energy metabolism is closely related to the polarized phenotype of macrophages, with M1 macrophages exhibiting enhanced glycolysis and reduced OXPHOS, whereas M2 macrophages display enhanced OXPHOS and reduced glycolytic metabolism (34).Seahorse analysis showed that extracellular acidification rate (ECAR), an indicator of glycolytic flux, was significantly (P < 0.05) increased in Salmonella WT-infected BMDMs compared to the uninfected BMDMs (Fig. 3A), while basal oxygen consumption rate (OCR), an indicator of mitochon drial respiration, was significantly (P < 0.01) decreased in Salmonella WT-infected BMDMs compared to that of the uninfected BMDMs (Fig. 3B), indicating a metabolic reprogram ming of macrophages from OXPHOS to glycolysis upon Salmonella infection, which are in agreement with previous results (31).The addition of 3 mM lactate decreased the ECAR and increased the OCR of Salmonella WT-infected BMDMs (Fig. 3A and B).These results suggest that lactate inhibits glycolysis and promotes OXPHOS in Salmonella-infected macrophages.The repression of glycolysis and promotion of OXPHOS by lactate indicate an M2 bias in Salmonella-infected macrophages in the presence of lactate.(G) Replication of Salmonella WT in ldhA siRNA-treated or control siRNA-treated RAW264.7 cells, in the presence or absence of 3 mM lactate.Data are presented as mean ± SD of three independent experiments (A, C-F).P values were determined using two-tailed unpaired Student's t-test (A, D, and E) or one-way analysis of variance (ANOVA) (C, F, and G).** P < 0.01, *** P < 0.001; ns, not significant.Moreover, we analyzed the expression levels of three M1 macrophage marker genes (Il1b, Tnf, and iNOS) and two M2 macrophage marker genes (Il4ra and Il10) in Salmonella WT-infected RAW264.7 macrophages, in the presence or absence of 3 mM lactate.The results showed that upon the addition of lactate, the expression of M1 marker genes was significantly decreased, while the expression of M2 marker genes was significantly (P < 0.05) increased in Salmonella WT-infected RAW264.7 macrophages (Fig. 4A and B), indicating that lactate inhibits M1 polarization gene expression and promotes M2 polarization gene expression of Salmonella-infected macrophages.
Collectively, these results indicate that lactate promotes M2 polarization of Salmo nella-infected macrophages.

Lactate promotes Salmonella intracellular replication via driving macrophage M2 polarization
Next, we assessed the relationship between lactate-mediated macrophage M2 polarization and Salmonella growth promotion.We treated RAW264.7 macrophages with IL4 to induce M2 polarization, and then, the IL4-stimulated macrophages were infected with Salmonella WT to test bacterial replication ability in the presence or absence of the LDHA inhibitor FX11, which inhibits macrophage lactate production.Addition of FX11 inhibited Salmonella replication in RAW264.7 cells (Fig. 5A, left panel).If lactate promotes Salmonella intracellular replication via driving macrophage M2 polarization, the decreased replication ability of Salmonella due to decreased intracellular lactate levels could be offset by the pre-induction of macrophages into the M2 phenotype.As expected, the addition of FX11 did not influence the replication ability of Salmonella in IL4-stimulated RAW264.7 cells (Fig. 5A, right panel).Moreover, siRNA knockdown of LDHA decreased Salmonella replication in RAW264.7 cells (Fig. 5B, left panel), but upon pre-stimulation with IL4, the replication ability of Salmonella WT in ldhA siRNAtreated RAW264.7 cells was comparable to that of the control siRNA-treated cells (Fig. 5B, right panel), indicating that the pre-stimulation of macrophages to the M2 pheno type abolished the bacterial replication defect mediated by ldhA siRNA.These results suggest that macrophage-derived lactate promotes Salmonella intracellular replication via driving macrophage M2 polarization.

Lactate promotes macrophage M2 polarization via the SPI-2 T3SS effector SteE
We investigated the relationship between lactate-and Salmonella-mediated M2 macrophage polarization.Salmonella promotes macrophage M2 polarization via the SPI-2 T3SS effector SteE, which activates the STAT3 signaling pathway to stimulate the production of the critical antiinflammatory cytokine IL10 (35).Interestingly, our previous study showed that macrophage-derived lactate could activate Salmonella SPI-2 gene expression, including steE (31).Consistent with the result, using a steE-lux transcriptional fusion and bioluminescent reporter assays, we showed that the addition of lactate induced the expression of steE in RAW264.7 cells (Fig. S2).Thus, we speculated that SteE is probably involved in the lactate-mediated macrophage M2 polarization during Salmonella infection.Macrophage infection assays showed that lactate addition increased Salmonella WT replication in RAW264.7 but did not influence the intracellular replication of ΔsteE (Fig. 6A), indicating that the mutation of steE abolished bacterial growth promotion mediated by lactate.Thus, Salmonella intracellular replication by lactate is SteE dependent.Quantitative real-time PCR (qRT-PCR) assays showed that lactate addition increased the expression of M2 macrophage marker genes (Il4ra and Il10) in Salmonella WT-infected RAW264.7 but did not influence the expression of M2 macrophage marker genes in ΔsteE-infected RAW264.7 macrophages (Fig. 6B), indicating that the mutation of steE abolished the induction of macrophage M2 polarization by lactate.Thus, polarization of macrophages by lactate is SteE dependent.
Moreover, we analyzed the lactate levels and M2 marker gene expression in Salmo nella WT-and ΔsteE-infected mouse livers at day 3 post-infection.The results showed that lactate levels had increased in the liver of both WT-infected mice and ΔsteE-infected mice compared to that of the uninfected mice (Fig. 6C, left panel), but relative expression of M2 marker genes was only increased in the livers of WT-infected mice but not in the livers of ΔsteE-infected mice (Fig. 6C, right panel).Consistent with the expression of M2 marker genes, the percentage of M2 macrophages (CD163+, CD163 is commonly associated with M2 polarization) in the liver of WT-infected mice was significantly higher than that of ΔsteE-infected mice (Fig. 6D).These results further indicate that SteE is required to promote M2 gene expression by lactate.The percentage of the CD163+ macrophage population correlated well with the bacterial burden of WT and ΔsteE in the livers of infected mice (Fig. 6D), implying that SteE-mediated macrophage M2 polariza tion is required for Salmonella efficient colonization of mouse systemic tissues.Collec tively, these results suggest that lactate promotes macrophage M2 polarization via SteE during Salmonella infection.
The relationship among lactate and SPI-2 expression, SteE translocation, and phosphorylation of STAT3 during Salmonella infection of macrophages was further investigated.Bioluminescent reporter assays revealed that lactate induced the higher expression of the Salmonella SPI-2 regulon (ssaG-lux) in RAW264.7 cells at 8 and 20 h post-infection, especially at 20 h post-infection (Fig. S3A).The addition of lactate increased the translocation of steE-FLAG to macrophage cytosol (Fig. S3B), as revealed by immunofluorescence analysis.Moreover, the translocation of SteE was quantified by Western blotting, using GAPDH as a loading control.The results showed that lactate increased the protein levels of SteE-FLAG in RAW264.7 cells (Fig. S3C and S4), further suggesting that lactate increased SteE translocation to RAW264.7 cells.In line with the increased SPI-2 expression and SteE translocation, Western blotting analysis showed that lactate increased the phosphorylation of STAT3 of RAW264.7 cells (Fig. S3D and S4).These results reveal that lactate drives higher expression of SPI-2 regulon, increased translocation of SteE, and, therefore, increased phosphorylation of STAT3, which in turn drives macrophage M2 polarization and promotes Salmonella intracellular replication.

DISCUSSION
Replication within macrophages represents a critical step during the induction of systemic infection by Salmonella and requires complex crosstalk and interactions between Salmonella and host macrophages (36).More recent studies suggest that interactions with macrophages at the metabolic interface are important for Salmonella pathogenicity (37)(38)(39)(40).In this study, we demonstrated that lactate promotes Salmonella intracellular replication and systemic infection by driving macrophage M2 polarization.The lactate-mediated macrophage M2 polarization depends on the function of the T3SS effector of Salmonella, SteE, which has been shown to activate the STAT3 signaling pathway (35,41).Therefore, our results illustrate the mechanisms by which host-derived lactate promotes Salmonella pathogenicity and provides an additional perspective on host-pathogen crosstalk at the metabolic interface.
We propose a model for lactate-mediated Salmonella growth in macrophages (Fig. 7).After Salmonella enters the macrophages, macrophages reprogram their glucose metabolism, leading to an increase in glycolytic metabolism and lactate levels of the infected macrophages (42).Host-derived lactate is employed by intracellular Salmonella as a cue to promote SPI-2 T3SS expression (31), which leads to the increased transloca tion of effector SteE from SCV to macrophage cytoplasm.SteE induces the phosphoryla tion of STAT3, and then, the phosphorylated, activated STAT3 drives macrophage M2 polarization (24).The replication of Salmonella in M2 macrophages leads to bacterial dissemination and systemic infection.This model builds a direct link between lactate-mediated and Salmonella-directed macrophage M2 polarization and further highlights the complex interactions between Salmonella and macrophages.
Our findings suggest that lactate inhibits M1 bactericidal responses and increases M2 polarization of Salmonella-infected macrophages, promoting Salmonella replica tion in infected macrophages.However, lactate also accumulates in the extracellular environment during Salmonella infection, and its role in the surrounding macrophages, especially uninfected macrophages, remains unclear.Notably, during in vivo infection, a large proportion of macrophages are not infected with Salmonella (43).A previous study showed that lactate produced by tumor cells can promote the M2 polarization of tumor-associated surrounding macrophages to attenuate the engulfment of tumor cells by macrophages, contributing to tumor growth (44).Presumedly, Salmonella and other intracellular pathogens employ similar mechanisms to modulate M2 polarization of uninfected macrophages.Further investigations are necessary to determine the effects of lactate derived from Salmonella-infected macrophages on the polarized phenotype and immune function of the surrounding uninfected macrophages.
We showed that lactate-mediated macrophage M2 polarization depended on the SPI-2 T3SS effector SteE.Mutation of SteE abolished the induction of macrophage M2 polarization by lactate and Salmonella growth promotion mediated by lactate.SPI-2 loci are present in Salmonella enterica but absent in the genome of Salmonella bongori, which is predominantly associated with cold-blooded animals, and other intestinal bacterial pathogens (45).In addition, SteE is present primarily in S. Typhimurium but absent in many other S. enterica serovars, such as S. Enteritidis, S. Typhi and S. Paratyphi.We confirmed that the addition of lactate did not influence the replication of S. Enteriti dis and S. Typhi in RAW264.7 macrophages (Fig. S5A and B).Thus, both lactate-and SteE-participating macrophage M2 polarization and the lactate-dependent increase in Salmonella replication are specific for S. Typhimurium but not for other Salmonella serovars, which lack the SteE effector.The mechanisms by which lactate promotes macrophage M2 polarization during infection with other bacterial pathogens require further study.
In addition to SteE, lactate also induces the expression of other SPI-2 effectors (31), for example, SifA, SseJ, SopD2, SseF, SseG, and SrfH, which are important for stimulating Salmonella replication in macrophages (46).However, it is unclear why the effect of lactate is specifically mediated via SteE but not mediated via other effectors.The different functions of these SPI-2 effectors lead to a plausible explanation that all these effectors are essential, while SteE acts in the "last step" for promoting Sal monella intracellular replication.First, SrfH inhibits the directed migration of infected macrophages to avoid their clearance by the host immune system (47).Second, SifA, SseJ, SopD2, SseF, and SseG induce the biogenesis and maintenance of the Salmonella replicative niche-SCV and/or the formation of Salmonella-induced filaments (SIFs) in infected macrophages (48)(49)(50)(51), which provide the requirements supporting Salmonella persistence.Last, SteE drives macrophage M2 polarization via the phosphorylation of STAT3 (24), decreasing host inflammation and increased Salmonella replication.In addition, our results revealed a dose-dependent activity of SteE, with higher levels of SteE driving more STAT3 phosphorylation and more Salmonella replication.In contrast, the inhibition of macrophage migration and the formation of SCV (and SIFs) probably do not occur in dose-dependent manners.Therefore, although lactate induces the expression of many critical SPI-2 effectors, we only observed that its effect is mediated explicitly via SteE.Further research is needed to validate this hypothesis and illustrate the complex actions among SPI-2 effectors.
Lactate is one of the most abundant metabolites in host cells, and its abundance increases in infected macrophages due to the metabolic shift to a Warburg-like metabolism (52).In contrast to the induction of antibacterial effects upon metabolic reprogramming, M2 polarization induced by lactate promotes the antiinflammatory functions of macrophages (53).Therefore, on the host side, lactate may balance the macrophage's pro-and antiinflammatory responses during bacterial infection to maintain tissue homeostasis.However, based on this study, we speculate that lactatemediated homeostasis is exploited by Salmonella, which participate in the lactate-medi ated M2 polarization pathway via the T3SS effector SteE.
Taken together, our results suggest that host-derived lactate promotes Salmonella intracellular replication and systemic infection by driving macrophage M2 polarization.In addition to the host STAT3 transcription factors, the Salmonella T3SS effector SteE is also required for lactate-directed M2 polarization, indicating that Salmonella actively regulates host cell polarization to expand the population of permissive macrophages for their survival.In addition to Salmonella, many other intracellular bacteria use macro phages for replication; links to macrophage M2 polarization and other bacterial infection require further investigations.

Bacterial strains, plasmids, and growth conditions
The bacterial strains and plasmids used in this study are listed in Table S1.Salmonella enterica serovar Typhimurium ATCC 14028s was used as the wild-type strain throughout this work.Mutant strains and FLAG-tagged strain (steE-FLAG) were generated using the λ Red-based recombination system (54,55).To generate the steE-lux transcriptional fusion, the PCR products of steE promoter region were digested with BamHI and XhoI and cloned into the XhoI-BamHI site of the plasmid pMS402, which carries a promoterless luxCDABE reporter gene cluster.The steE-lux fusion plasmid was then transformed into WT strain to generate the lux reporter strain WT+steE-lux.The primers used to construct the strains and plasmids are listed in Table S2.Salmonella strains were grown to stationary phase in Luria-Bertani (LB) broth at 37°C with aeration for infection of macrophages and mouse infection assays.When required, antibiotics were added at the following final concentrations: 25 µg/mL chloramphenicol, 100 µg/mL ampicillin, and 50 µg/mL kanamycin.

Cell culture
The mouse macrophage cell line RAW264.7 (ATCC TIB71) was obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37°C with 5% CO 2 .BMDMs were prepared by flushing the femurs and tibia of C57BL/6 wild-type mice with PBS.Harvested BMDMs were differentiated in DMEM containing 10% FBS and 10 ng/mL Macrophage Colony Stimulating Factor (M-CSF) (Proteintech, HZ-1192).Cells were incubated for 7 days at 37°C in a 5% CO 2 environment with medium change every 2-3 days.

Macrophage infections
BMDMs or RAW264.7 cells were seeded at a density of 1 × 10 5 cells per well in a 24well culture plate and infected with Salmonella WT or mutant strains grown to station ary phase at a multiplicity of infection of 10.When indicated, the cells were treated with 10 ng/mL recombinant mouse IL4 (Peprotech, 214-14) for 24 h before infection.Overnight-cultured bacterial strains were serially diluted to 1 × 10 6 colony-forming units (CFUs)/mL, opsonized in DMEM containing 10% serum for 30 min, and then added to macrophage monolayers.The plates were centrifuged at 500 × g for 5 min to synchronize bacterial uptake.After incubation for 30 min, the cells were washed thrice with PBS and incubated in DMEM containing 100 µg/mL gentamicin for 1 h to remove extracellular bacteria.The cells were then incubated in a DMEM containing 10 µg/mL gentamicin (Gm) for the remainder of the experiment.Infected cells were lysed using 1% Triton X-100 at 2 and 20 h post-infection, and intracellular bacteria were enumerated on LB plates for CFU analysis.Fold intracellular replication of Salmonella strains was determined by dividing the intracellular bacterial load recovered at 20 h by that recovered at 2 h.A 3 mM lactate and/or a 23.3 µM FX11 were added after 1 h of gentamicin treatment to avoid the possible influence of lactate (and/or FX11) on bacterial internalization.

Immunofluorescence microscopy
To enumerate intracellular bacteria, BMDMs seeded on 20-mm-diameter coverslips were infected with Salmonella WT as described above.A 3 mM lactate was added after 1 h of gentamicin treatment.At 2 and 20 h post-infection, the cells were washed thrice with PBS, fixed with 4% paraformaldehyde (PFA; Solarbio, P1110) at 4°C for 15 min and blocked with 5% bovine serum albumin (BSA; Solarbio, SW3015) at room temperature for 1 h.The cells were then incubated with the primary antibody mouse anti-Salmonella Typhimurium LPS (Abcam, ab8274; 1:100 dilution) at room temperature for 1 h.Next, the cells were washed thrice with PBS and incubated with fluorophoreconjugated secondary antibody Goat Anti-Mouse IgG H&L (Alexa Fluor 488) (Abcam, ab150113; 1:200 dilution) at room temperature for 1 h.After being washed thrice with PBS, the cell nuclei were stained with 5% DAPI (Bioss, C02-04002) at room temperature for 5 min.For detecting the translocation of SteE, RAW264.7 cells seeded on 20-mm-diameter coverslips were infected with WT steE-FLAG strain for 8 and 20 h.The cells were then incubated with the primary antibodies FITC Anti-Salmonella antibody (Abcam, ab20320; 1:50 dilution), Recombinant Alexa Fluor 647 Anti-LAMP1 antibody (Abcam, ab237307; 1:100 dilution), and Recombinant Anti-DDDDK tag (Binds to FLAG tag sequence) antibody (Abcam, ab205606; 1:100 dilution) at room temperature for 1 h.Next, the cells were washed thrice with PBS and incubated with fluorophoreconjugated secondary antibody Goat Anti-Rabbit IgG H&L (Alexa Fluor 405) (Abcam, ab175652; 1:500 dilution) at room temperature for 1 h.The stained cells were visualized using a Zeiss LSM800 confocal microscope (Zeiss, Germany).The ZEN 2.3 (blue edition) was used for the further image processing.

Mouse infections
To enumerate bacterial burden in mouse liver and spleen, mice were intraperitoneally (i.p.) infected with ~10 4 CFUs of the WT strain plus lactate (0.6 mg lactate/g mice) or bacteria only in PBS.The infected mice were sacrificed on day 3 post-infection.Their livers and spleens were collected, homogenized in PBS, serially diluted, and plated on LB agar plates to obtain CFUs per gram of tissue.To detect M2 marker gene expression in mouse livers, the infected mice were sacrificed on day 3 post-infection.The livers were homogenized with TRIzol reagent to extract total RNA for subsequent qRT-PCR experiments.

Measurement of lactate concentration
To assess the lactate levels of BMDMs, cells were left uninfected or infected with Salmonella WT for 20 h, and the supernatants were collected to measure lactate concentration.To assess lactate levels in the mouse liver, mice were left uninfected or intraperitoneally infected with ~10 4 CFU Salmonella WT or steE mutant.The infected mice were sacrificed on day 3 post-infection, and their spleens were collected and homogenized in 1 mL of PBS to measure lactate concentration.The lactate concentra tion was measured using a lactate assay kit (Sigma-Aldrich, MAK064) according to the manufacturer's instructions.

Seahorse analysis
BMDMs were seeded into Seahorse 24-well tissue culture plates (Agilent Technologies) and left uninfected or infected with Salmonella WT for 20 h in the presence or absence of 3 mM lactate.The OCR and ECAR of BMDMs were measured on the Seahorse XF24 Analyzer (Agilent Technologies), by using Seahorse XF Glycolysis Stress Test Kit (Agilent Technologies, 103020) and Seahorse XF Cell Mito Stress Test Kit (Agilent Technologies, 103015), respectively, according to manufacturer's instructions.The basal OCR was calculated by subtracting non-mitochondrial respiration from the values obtained before oligomycin addition.The ECAR was calculated by subtracting the normalized ECAR values after 2-deoxy-D-glucose injection from the ECAR values after glucose injection.

RNA isolation and quantitative real-time PCR
Total RNA was isolated using the RNAsimple Total RNA Kit (TIANGEN Biotech, DP419), and cDNA was synthesized using StarScript III RT Master Mix (Genstar, A233) according to the manufacturer's instructions.Each 20-µL qRT-PCR reaction system contained 1-µL cDNA, 10 µM of each primer, and 10 µL of 2× RealStar Power SYBR qPCR Mix (Genstar, A314).The qRT-PCR was run on a QuantStudio Real-Time PCR Systems (Thermo Fisher Scientific) using the following program: 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 30 s. Expression of target genes was normalized to Gapdh as a control.Relative changes in gene expression were analyzed using the comparative cycle threshold (ΔΔCt) method (56).
Small interfering RNA (siRNA) transduction ldhA and control siRNAs were designed and synthesized by Guangzhou RiboBio Co., Ltd.(GuangZhou, China).Primer for siRNA against ldhA:5′-GGAATCAATGAGGATGTCT-3′.RAW264.7 cells were transfected with ldhA siRNA or control siRNA at a final concentration of 50 nM using RiboFECT CP Transfection Kit (RiboBio, C10511) 48 h before bacterial infection.The interference efficiency was assessed via qRT-PCR at 24 h after transfection.

Bioluminescent reporter assays
To determine the effect of lactate on intracellular ssaG and steE expression, RAW264.7 cells were infected with the WT+ssaG-lux or WT+steE-lux.A 3-mM lactate was added after 1 h of gentamicin treatment.At 8 and 20 h post-infection, the cells were lysed with 1% Triton X-100.A 200-µL sample of the cell lysates was loaded into a 96-well black assay plate with a clear flat bottom (Corning 3603) to measure luminescence using the Spark multimode microplate reader (Tecan).A 100-µL sample of the cell lysates was serially diluted and plated on LB agar plates to enumerate intracellular bacterial CFUs.Luminescence values were normalized to intracellular bacterial CFUs.

Flow cytometry
Mice were i.p. infected with ~10 4 CFUs of Salmonella WT or ΔsteE, with two mice in each bacterial injection group.The WT-and ΔsteE-infected mice were sacrificed on day 3 post-infection.Their livers were collected and divided into two parts with scissors on each liver, one for detecting bacterial CFUs and the other for flow cytometry analysis.For flow cytometry analysis, liver from two mice (of one group) were pooled together and then dissociated into single cells through gentleMACS/Mouse liver dissociation kit (Miltenyi Biotec, 130-105-807).Cells were fixed with 4% PFA and permeabilized with 0.1% Triton X-100.The fixed cells were then stained with an anti-CD68 antibody (Abcam, ab221251), an anti-CD163 antibody (Abcam, ab182422), and goat anti-rabbit lgG H&L (FITC) (Abcam, ab6717).The cells were suspended in PBS and adjusted to a concentration of 1 × 10 5 cells/mL.Nonspecific antigens were blocked using a 5% BSA buffer.The cells were analyzed using a BD FACSAria Flow Cytometer (BD Biosciences).To determine bacterial CFUs, livers from two mice (of one group) were pooled and homogenized in PBS together, serially diluted, and plated on LB agar plates to obtain CFUs per gram.Three independent experiments were performed.
To prepare whole cell lysates, cells were washed once in PBS and lysed in SDS loading buffer before sonication.For post nuclear supernatant (PNS) and pellet analysis, cells were washed once in PBS and lysed in cell lysis buffer (10% glycerol, 20 mM Tris-Cl pH 7.4, 150 mM NaCl, 0.1% Triton X-100) on ice for 10 min before clarification by centrifugation.SDS loading buffer was added to the PNS and pellet samples so that both fractions were in the same final volume.All samples were then heated to 95°C and separated by SDS-PAGE using either 8%, 10%, or 12% polyacrylamide denaturing gels, transferred to polyvinylidene difluoride (PVDF) membrane (Millipore), and visualized by immunoblotting using ECL detection reagents (Dako) on a Chemidoc Touch Imaging System (Bio-Rad).

Data analysis
Data were obtained from three independent experiments and were presented as mean ± SD, unless otherwise indicated.Statistical significance was determined by two-tailed unpaired Student's t-test, Mann-Whitney U test, one-way analysis of variance (ANOVA), or two-way ANOVA using GraphPad Prism 8.0.1 software (GraphPad, Inc., San Diego, CA, USA).Significant differences between groups are represented as * P < 0.05, ** P < 0.01, and *** P < 0.001.ns represents no statistical significance.

FIG 1
FIG 1 Lactate promotes Salmonella replication within macrophages.(A) Lactate production by uninfected BMDMs (UN) or those infected with Salmonella WT (Salmonella Typhimurium ATCC 14028s) for 20 h.(B) Enzymes responsible for D-and L-lactate production in Salmonella.(C) Lactate production by uninfected BMDMs (UN) or those infected with Salmonella WT, ΔlldD, ΔldhAΔdld, and ΔlldDΔldhAΔdld for 20 h.(D) Replication of Salmonella WT in BMDMs, in the presence or absence of 3 mM lactate.(E) Number of intracellular bacteria per BMDM, in the presence or absence of 3 mM lactate.Infected cells were fixed at 2 and 20 h post-infection and prepared for immunofluorescence staining.The number of intracellular bacteria per infected cell was counted in random fields, n = 150 cells per group pooled from three independent experiments.Representative immunofluorescence images were shown in the left panel.Green, Salmonella; blue, nuclei.Scale bars, 20 µm.(F) Replication of Salmonella WT in BMDMs, in the presence or absence of 23.3 µM FX11 and 23.3 µM FX11+ 3 mM lactate.

FIG 2
FIG 2 Lactate promotes Salmonella colonization of systemic loci in mice.(A) Bacterial counts recovered from the liver and spleen of C57BL/6 mice intraperito neally (i.p.) infected with Salmonella WT.Mice were either i.p injected with ~10 4 CFU bacteria plus lactate (0.6 mg lactate/ g mice) or bacteria only (control).Liver and spleen were collected on day 3 post-infection to quantify bacterial burden; n = 6 mice per group.(B) Bacterial counts recovered from the liver of wild-type or liverspecific LDHA knockout (LDHA −/− ) C57BL/6 mice i.p. infected with Salmonella WT.Mice were i.p injected with ~10 4 CFU bacteria.The liver was collected on day 3 post-infection to quantify bacterial burden; n = 6 mice per group.Data are combined from two independent experiments (A, B).P values were determined using Mann-Whitney U test (A, B). ** P < 0.01.

FIG 7
FIG 7 Model of lactate-mediated Salmonella growth in macrophages.Salmonella infection promotes macrophage glycolysis, and lactate levels increased in the infected macrophages.Host-derived lactate induces Salmonella SPI-2 T3SS expression, and the translocation of effector SteE from SCV to macrophage cytoplasm increased.SteE induces the phosphorylation of STAT3.Activated STAT3 drives macrophage M2 polarization, and Salmonella replicates in M2 macrophages.MФ, macrophage.