Galectin-3 is a β-galactoside-binding lectin that plays an important role in inflammatory diseases. It also interacts with the surface carbohydrates of many pathogens, including LPS. However, its role in infection is not fully understood. Data presented herein demonstrate for the first time that galectin-3 is a negative regulator of LPS-induced inflammation. Galectin-3 is constitutively produced by macrophages and directly binds to LPS. Galectin-3-deficient macrophages had markedly elevated LPS-induced signaling and inflammatory cytokine production compared with wild-type cells, which was specifically inhibited by the addition of recombinant galectin-3 protein. In contrast, blocking galectin-3 binding sites by using a neutralizing Ab or its ligand, β-lactose, enhanced LPS-induced inflammatory cytokine expression by wild-type macrophages. In vivo, mice lacking galectin-3 were more susceptible to LPS shock associated with excessive induction of inflammatory cytokines and NO production. However, these changes conferred greater resistance to Salmonella infection. Thus, galectin-3 is a previously unrecognized, naturally occurring, negative regulator of LPS function, which protects the host from endotoxin shock but, conversely, favors Salmonella survival.

Lipopolysaccharide represents a major pathogen-associated molecular pattern (PAMP)3 from the outer membrane of Gram-negative bacteria (1, 2). It is a potent immune activator closely associated with many infectious and inflammatory diseases (1, 2). LPS consists of hydrophobic lipid A, the 0-polysaccharide chain, and a core oligosaccharide that may be recognized differently by the host immune system (2, 3, 4). The visualization of LPS requires the TLR4 complex and triggers MyD88-dependent and independent signaling pathways (2, 5). This leads to the activation of NF-κB and kinases, including MAPK, ERK1/2, p38, and JNK, which subsequently turn on the expression of many inflammatory genes including NADPH oxidase and inducible NO synthase (4). Since LPS is a powerful immune activator and may be fatal, the response to LPS must be tightly regulated to maintain the immune response at an appropriate level (6).

Galectin-3 is a 31-kDa chimeric galectin characterized by a single C-terminal carbohydrate recognition domain (CRD) for carbohydrate binding and an N-terminal aggregating domain that interacts with a noncarbohydrate ligand and allows the formation of oligomers (7, 8). It is predominantly expressed by innate cells including macrophages and is closely associated with inflammatory responses and cellular functions (9). It mainly exists in the cytosol due to the lack of a leader sequence, and it is involved in many cellular events (e.g., apoptosis) (10). However, galectin-3 can also be externalized through a nonclassical transport pathway (11). As an extracellular protein, it may interact with glycoproteins within the extracellular matrix to form a glycoprotein lattice or act as a soluble ligand to crosslink with the carbohydrates of surface proteins by N-terminal oligomerization, thus evoking signal transduction and cell functions (10, 11, 12, 13, 14). Mice lacking galectin-3 develop a dominant Th1 phenotype and exhibit abnormalities in several inflammatory disease models including asthma and diabetes, suggesting that galectin-3 may be involved in the regulation of inflammatory and Th1 responses (15, 16, 17, 18).

β-galactoside carbohydrates are common structures on many pathogens (19). It has been suggested that galectin-3 may serve as a pathogen pattern recognition receptor to visualize PAMPs from bacteria (19, 20), parasites (21, 22), and fungi (23, 24). Moreover, it has been reported that galectin-3 interacts with LPS via both N′ and C′ terminals (25). However, the immunological and pathophysiological significance related to these interactions has not been explored.

Data presented herein demonstrate that galectin-3 is a negative regulator for LPS function. Macrophages spontaneously express galectin-3, which specifically binds to LPS. Macrophages from galectin-3-deficient mice have elevated inflammatory cytokine production in response to LPS and lipid A compared with wild-type cells. This is accompanied by an increased phosphorylation of JNK, p38, ERK, and NF-κBp65. The increased inflammatory cytokine production by galectin-3 knockout cells could be normalized by recombinant galectin-3 protein. In vivo, mice lacking galectin-3 excessively produced inflammatory cytokines and NO and were more susceptible to LPS shock. On the other hand, such mice were more resistant to Salmonella infection due to the skewing of a Th1 response and increasing the levels of NO and hydrogen peroxide. Thus, galectin-3 is a novel natural negative regulator for LPS function, which protects against endotoxic shock but may be detrimental by helping in early Salmonella infection.

C57BL/6 mice were purchased from Harlan U.K. Galectin-3−/− mice were generated as described previously (16). All mice were maintained at the Biological Services Facilities of the University of Glasgow under U.K. Home Office guidelines. Polymyxin B and β-lactose, LPS from Escherichia coli 0111:B4, LPS (Re) from Salmonella minnesota Re 595, lipid A, and peptidoglycan (PGN) were purchased from Sigma-Aldrich. TLR agonists bacterial lipopeptide (BLP, Pam3Cys-SK4), poly(I:C), flagellin and R848 were from InvivoGen. Recombinant murine galectin-3, anti-galectin-3 Ab, and ELISA kit were obtained from R&D Systems. ELISA kits for cytokines, IL-12, IL-6, and TNF-α were from BD Biosciences.

Bone marrow-derived macrophages (BMMs) were generated as before (26). RPMI 1640 medium (Invitrogen) containing 10% FCS, penicillin/streptomycin, and glutamine (complete medium) were used for cell culture. Briefly, femurs were flushed with complete medium, and cells were plated in complete medium containing 104 U/ml recombinant human CSF-1 on 10-cm bacteriological plastic plates (Bibby Sterilin) for 7 days in a 37°C incubator containing 5% CO2. In some experiments, 20% L929 cell-conditioned medium was used as a source of CSF-1 instead of recombinant human CSF-1. For in vitro macrophage activation, BMMs were plated in 24-well plates at 1 × 106 cells/well in 1 ml complete medium plus CSF-1 overnight. The cells were then treated with LPS or different TLR ligands for 8 h, and the cytokines produced in the supernatants were determined by ELISA.

Two methods were used for the study of galectin-3 (R&D Systems) binding to E. coli LPS. ELISA plates (96-well) were coated overnight with LPS (50 μg/ml) (see Fig. 3A) or with recombinant galectin-3 (20 μg/ml) or PBS as a control in triplicates (see Fig. 3E). The plate was then washed and blocked with 10% FCS for 1 h at 37°C. In Fig. 3A, different doses of recombinant galectin-3 were added. In some wells, the galectin-3 was preincubated with anti-galectin-3 Ab or control IgG (2 μg/ml) for 30 min before being added into the wells. The samples were incubated for 2 h at 37°C, washed thoroughly, and reincubated with biotinylated anti-galectin-3 Ab (1 μg/ml). For Fig. 3E, different doses of biotinylated LPS (InvivoGen) were added for 1 h. ExtrAvidin peroxidase (2 μg/ml) was then added for 1 h followed by tetramethylbenzidine substrate. The plate was read using an ELISA reader at 630 nm.

BMMs (1 × 106 cells/tube) were harvested and blocked with Fc block (5 μg/ml) for 30 min. For surface staining, the conjugated anti-galectin-3, CD14-FITC, TLR4-PE, or isotype control Abs were then added and incubated for 30 min at 4°C. For intracellular galectin-3 staining, the cells were fixed with Cytofix/Cytoperm buffer (BD Biosciences), permeabilized with Perm/Wash buffer (BD Biosciences), and incubated with PE-conjugated anti-galectin-3 or isotype controls. Cells were then washed in FACS buffer (1× PBS supplemented with 2% FCS, 0.1% sodium azide) before being analyzed on a FACSCalibur flow cytometer (BD Biosciences).

BMMs were stimulated with LPS or LPS (Re) for different times. The cytoplasmic and nuclear proteins were extracted by using a nuclear extract kit (Active Motif). The proteins were separated on 12% SDS-PAGE and transferred onto polyvinylidene difluoride membrane followed by incubation with Abs specific for phosphorylated (P) p65, p38, JNK1/2, ERK1/2, and nonphosphorylated p65, p38, JNK1/2, and ERK1/2 as loading controls (Cell Signaling Technology).

BMMs were grown on culture slides and then stained for galectin-3 or TLR4 using specific conjugated Abs. Cells were analyzed with a Zeiss Axiovert microscope using a FITC and Texas Red specific filter set (Glen Spectra).

Fresh overnight cultures of Salmonella-GFP strain were washed once in distilled water and heat fixed on polylysine slides. Fixed Salmonella-GFP were preincubated with PBS or 5 μg of galectin-3 (Affinity BioReagents) on slides for 30 min, then incubated with anti-galectin-3 Ab (5 μg/ml) or IgG control, followed by anti-mouse Ig-Cy3 (1/200 dilution, Sigma- Aldrich) for 30 min and visualized by confocal microscopy.

Age- and sex-matched galectin-3−/− and control C57BL/6 mice (n = 10/group) were injected i.p. with LPS (23 mg/kg body weight). Animals were observed every 6 h for general health. Experiments were terminated on day 5 after LPS injection, as required by the guidelines for animal experimentation (Home Office, London, U.K.). Blood samples were taken from tail vein at indicated times. Serum NO production was quantified by the measurement of the accumulation of nitrite levels using a Griess reagent kit (Biotium) and cytokines produced by ELISA.

Attenuated Salmonella BRD509 (aroAaroD) strain was cultured as before (27, 28). Groups of female galectin-3−/− or wild-type mice were infected orally with 1 × 109 live bacteria in PBS (27). Mice were observed daily for up to 16 days. The mice were culled at different time points. The serum was taken and NO levels were determined by the Griess method, H2O2 by a peroxide assay kit (Pierce), and galectin-3 by ELISA. The spleens, livers, and lymph nodes of the mesentery were removed and homogenized. The serially diluted homogenates were put onto Luria-Bertani agar plates in triplicates and incubated at 37°C overnight. The numbers of bacteria were counted and expressed as CFU/organ. The spleen and draining lymph node cells were cultured in 24-well plates at 2 × 106 cells/ml complete medium and treated with medium alone or with different doses of Salmonella Ags (bacteria were lysed by freezing at −70°C and thawing out at 37°C for at least three cycles) for 48 and 72 h. The supernatants were collected and cytokine profile was determined by ELISA using paired Abs.

Statistical analysis was performed using Minitab software. For in vivo trials, the two-tailed log-rank test was used, and Student’s t test was used for analysis of cytokine concentrations. Statistical significance was accepted at p < 0.05.

Galectin-3 was originally identified on macrophages as a macrophage surface Ag (7). However, its role in these cells is not fully understood. First, we sought to systemically study the expression and localization of galectin-3 BMMs. These primary macrophages were generated from wild-type (WT) and galectin-3−/− mice and the expression of galectin-3 was determined. CD14+ BMMs from WT but not galectin-3−/− mice constitutively expressed intracellular galectin-3 detected by flow cytometry using a specific Ab (Fig. 1,A). Since TLR4 and CD14 are the important components in the LPS receptor complex, we established that galectin-3−/− expressed similar levels of TLR4 and CD14 as did the WT cells (Fig. 1,B). Immunofluorescence microscopy showed the coexpression of galectin-3 with TLR4 on BMMs (Fig. 1,C). Furthermore, BMMs from WT but not galectin-3−/− mice also spontaneously secreted galectin-3 detected within 2 h after culture and were further enhanced in a time-dependent manner (Fig. 1,D). Interestingly, the galectin-3 production could be significantly increased 2–4 h after LPS stimulation (Fig. 1 D).

FIGURE 1.

BMMs coexpress galectin-3, TLR4, and CD14. BMMs from WT and galectin-3−/− (Gal3−/−) mice were generated as described (48). A, The cells were stained for surface CD14 then permeabilized and stained for intracellular galectin-3 using specific Abs and detected by FACS. B, The surface TLR4 and CD14 levels on WT and Gal3−/− cells were detected by FACS. C, Cells were double stained with anti-TLR4-Texas Red and anti-galectin-3-FITC and visualized by immunofluorescence microscopy. D, WT and Gal3−/− BMMs (1 × 106/ml) were cultured with or without LPS (1 μg/ml) for indicated times, and secreted galectin-3 was detected by ELISA. The results represent one of three independent experiments.

FIGURE 1.

BMMs coexpress galectin-3, TLR4, and CD14. BMMs from WT and galectin-3−/− (Gal3−/−) mice were generated as described (48). A, The cells were stained for surface CD14 then permeabilized and stained for intracellular galectin-3 using specific Abs and detected by FACS. B, The surface TLR4 and CD14 levels on WT and Gal3−/− cells were detected by FACS. C, Cells were double stained with anti-TLR4-Texas Red and anti-galectin-3-FITC and visualized by immunofluorescence microscopy. D, WT and Gal3−/− BMMs (1 × 106/ml) were cultured with or without LPS (1 μg/ml) for indicated times, and secreted galectin-3 was detected by ELISA. The results represent one of three independent experiments.

Close modal

To define the role of galectin-3 as a pathogen pattern recognition receptor and the immunological significance related to the interplay between galectin-3 and LPS, we tested the effects of several well-documented TLR agonists on galectin-3−/− and WT BMMs: LPS (E. coli 0111:B4) and lipid A for TLR4, bacteria BLP and PGN for TLR2, poly(I:C) for TLR3, flagellin for TLR5, and R848 for TLR7 and TLR8. BMMs (1 × 106/ml/well) were rested on 24-well plates overnight before being activated with different TLR ligands directly without washing. As illustrated in Fig. 2,A, these TLR agonists induced the expected levels of IL-12, IL-6, and TNF-α production from WT BMMs. However, LPS-stimulated galectin-3−/− cells exaggerated this and markedly increased amounts of all inflammatory cytokines tested compared with WT controls (Fig. 2,A). It was noted that the cytokines induced by TLR2 agonists, BLP and PGN were selectively affected in galectin-3-null cells, since only IL-6 but not IL-12 and TNF-α production was significantly enhanced (Fig. 2,A). In contrast, galectin-3 deficiency had no significant influence on the effect of other tested TLR ligands (Fig. 2,A). We further assessed the role of galectin-3 on LPS and its derivative LPS (Re) from rough strain (lacking 0-polysaccharide) of Salmonella and lipid A. BMMs from WT and galectin-3−/− mice were cultured overnight as above and stimulated with graded doses of LPS, LPS (Re), or lipid A directly without washing for 8 h. As illustrated in Fig. 2 B, compared with WT, galectin-3−/− BMMs exhibited exaggerated response to LPS, LPS (Re), and lipid A and produced significantly higher amounts of IL-12, IL-6, and TNF-α in a dose-dependent manner. No significant changes in the IL-10 and TGF-β production had been detected between WT and galectin-3−/− BMMs (data not shown). Thus, our results demonstrate that galectin-3 is a novel regulator that mainly controls the response to TLR4 ligands and, partially, TLR2 ligands.

FIGURE 2.

Galectin-3−/− BMMs accelerate LPS-induced inflammatory cytokines. The overnight-cultured BMMs (1 × 106/ml) from WT and galectin-3−/− mice were directly activated with LPS (0.1 μg/ml), BLP (0.1 μg/ml), PGN (10 μg/ml), poly(I:C) (5 μg/ml), flagellin (0.5 μg/ml), and R848 (1 μg/ml) (A) or different dose of LPS, LPS (Re), and lipid A as indicated (B). The cells were cultured for 8 h and IL-12, IL-6, and TNF-α production was determined by ELISA. ∗, p < 0.05; ∗∗, p < 0.01 TLR agonists treated vs control cells. Data are the means ± SEM of three individual experiments.

FIGURE 2.

Galectin-3−/− BMMs accelerate LPS-induced inflammatory cytokines. The overnight-cultured BMMs (1 × 106/ml) from WT and galectin-3−/− mice were directly activated with LPS (0.1 μg/ml), BLP (0.1 μg/ml), PGN (10 μg/ml), poly(I:C) (5 μg/ml), flagellin (0.5 μg/ml), and R848 (1 μg/ml) (A) or different dose of LPS, LPS (Re), and lipid A as indicated (B). The cells were cultured for 8 h and IL-12, IL-6, and TNF-α production was determined by ELISA. ∗, p < 0.05; ∗∗, p < 0.01 TLR agonists treated vs control cells. Data are the means ± SEM of three individual experiments.

Close modal

The results from galectin-3-deficient cells (Fig. 2) reveal an apparent modulatory role of this molecule for LPS responsiveness. To investigate whether galectin-3 affects LPS function by binding to LPS as suggested before (25), using two different binding assays, we confirmed that galectin-3 bound to LPS (Fig. 3, A and E). Importantly, this binding could be specifically abolished by the preincubation of galectin-3 with anti-galectin-3 Ab but not control IgG (Fig. 3,A). To determine whether the interaction between galectin-3 and LPS could affect LPS-induced inflammatory cytokine production, the LPS or lipid A was pretreated with different doses of recombinant galectin-3 for 30 min before being added into the overnight cultured galectin-3−/− BMMs, and the subsequent cytokine production was measured as illustrated (Fig. 3,B). The WT and galectin-3−/− BMMs cultured in the presence or absence of LPS or lipid A were also included as controls. As noted previously, galectin-3−/− cells had exaggerated IL-12, IL-6, and TNF-α production in response to LPS or lipid A compared with WT cells. The increased effect was significantly reduced by the preincubation of LPS or lipid A with galectin-3 in a dose-dependent manner (Fig. 3,B). Reversely, pre-adding neutralizing anti-galectin-3 Ab but not control IgG into the WT BMMs culture significantly enhanced LPS-induced IL-12, IL-6, and TNF-α production (Fig. 3 C). Furthermore, it was noted that LPS elicited a less significant increase in inflammatory cytokine production when the galectin-3 containing supernatants from overnight cultured WT BMMs were replaced with fresh medium before adding LPS, suggesting that both secreted and surface galectin-3 contribute to the regulatory effect (data not shown). Thus, our results suggest that galectin-3 inhibits LPS and lipid A function through a direct interaction.

FIGURE 3.

Galectin-3 binds to LPS and blocks LPS-induced inflammatory cytokine production. A and E, Galectin-3 specifically bound to LPS in two different binding assays (Materials and Methods). B, Galectin-3−/− and WT BMMs were cultured with or without LPS (0.1 μg/ml) or lipid A (LA, 0.5 μg/ml) alone or with LPS or lipid A pretreated with different concentrations of recombinant galectin-3 for 30 min. All cells were cultured for 8 h, and supernatant levels of IL-12, IL-6, and TNF-α were measured by ELISA. C, WT BMM cultures were pretreated with anti-galectin-3 Ab or control IgG (2 μg/ml) for 30 min before adding LPS for 8 h. D, WT BMM cultures were preincubated with or without variable doses of β-lactose (Lact) for 30 min before being activated with or without LPS for 8 h and cytokine production was determined. ∗, p < 0.05; ∗∗, p < 0.01 compared with control. Data are presented as means ± SEM and are representative of three independent experiments.

FIGURE 3.

Galectin-3 binds to LPS and blocks LPS-induced inflammatory cytokine production. A and E, Galectin-3 specifically bound to LPS in two different binding assays (Materials and Methods). B, Galectin-3−/− and WT BMMs were cultured with or without LPS (0.1 μg/ml) or lipid A (LA, 0.5 μg/ml) alone or with LPS or lipid A pretreated with different concentrations of recombinant galectin-3 for 30 min. All cells were cultured for 8 h, and supernatant levels of IL-12, IL-6, and TNF-α were measured by ELISA. C, WT BMM cultures were pretreated with anti-galectin-3 Ab or control IgG (2 μg/ml) for 30 min before adding LPS for 8 h. D, WT BMM cultures were preincubated with or without variable doses of β-lactose (Lact) for 30 min before being activated with or without LPS for 8 h and cytokine production was determined. ∗, p < 0.05; ∗∗, p < 0.01 compared with control. Data are presented as means ± SEM and are representative of three independent experiments.

Close modal

Structurally, the N-terminal of galectin-3 binds to lipid A and the C-terminal CRD binds to the β-galactoside of LPS (25). Therefore, it is possible that galectin-3 blocks LPS function via both sites. BMMs lacking galectin-3 were hypersensitive to lipid A, which lacks β-galactoside, suggesting that the N-terminal of galectin-3 has a regulatory effect on LPS (Figs. 2,B and 3,B). To investigate whether the C′-terminal lectin-binding domain of galectin-3 was also involved, we pretreated WT BMM cultures with or without variable doses of the galectin-3 ligand, β-lactose, for 30 min to block the C-terminal carbohydrate-binding sites of galectin-3 before adding LPS and culturing the cells for 8 h. Preincubating with β-lactose dramatically increased the LPS-induced inflammatory cytokines, IL-12, IL-6, and TNF-α production by WT BMMs in a dose-dependent fashion (Fig. 3 D). Therefore, our data suggest that galectin-3 may suppress LPS function by interacting with LPS via lipid A and/or β-lactose motifs using N-terminal nonlectin and/or C-terminal lectin domain, respectively. Polymyxin B neutralizes LPS activity by binding to the lipid A domain, which also blocks the interaction between LPS and galectin-3 (25, 29). Preincubation of LPS and lipid A with polymyxin B almost completely abolished their effect on galectin-3−/− BMMs (data not shown), confirming that the effects observed by LPS on galectin-3−/− BMMs are specific.

We further investigated whether the enhanced inflammatory cytokine production by galectin-3−/− cells is due to accelerated LPS signaling. WT and galectin-3−/− BMMs were activated as indicated (Fig. 4, A and B). The LPS-induced phosphorylation of JNK1/2, ERK1/2, p38, and NF-κBp65 was more persistent, and JNK1/2 appeared earlier and stronger in galectin-3−/− BMMs than that in WT cells (Fig. 4,A). It was also the case in the LPS (Re)-stimulated cells (Fig. 4,B), which is consistent with the inflammatory cytokine profile (Fig. 2,B). Importantly, the enhanced phosphorylation of JNK1/2, ERK1/2, p38, and NF-κBp65 by LPS in galectin-3−/− cells could be normalized by adding recombinant galectin-3 (Fig. 4 C). Thus, galectin-3 affects LPS function by binding to the molecule and subsequently interfering with its signaling cascades.

FIGURE 4.

Galectin-3 affects LPS function by sequestrating LPS signaling. BMMs were activated with LPS (0.1 μg/ml) (A) or LPS (Re) (0.5 μg/ml) (B) for indicated times. Cytosol proteins were separated and the levels of phosphorylated (P) JNK1/2, ERK1/2, p38, p65, and nonphosphorylated JNK1/2, ERK1/2, p38, and p65 (loading control) were detected by Western blot using specific Abs. C, BMMs were incubated with or without galectin-3 (500 ng/ml) for 30 min and activated with LPS for indicated times. The cytosol proteins were isolated and Western blot performed as in A and B.

FIGURE 4.

Galectin-3 affects LPS function by sequestrating LPS signaling. BMMs were activated with LPS (0.1 μg/ml) (A) or LPS (Re) (0.5 μg/ml) (B) for indicated times. Cytosol proteins were separated and the levels of phosphorylated (P) JNK1/2, ERK1/2, p38, p65, and nonphosphorylated JNK1/2, ERK1/2, p38, and p65 (loading control) were detected by Western blot using specific Abs. C, BMMs were incubated with or without galectin-3 (500 ng/ml) for 30 min and activated with LPS for indicated times. The cytosol proteins were isolated and Western blot performed as in A and B.

Close modal

Endotoxin shock is the most severe form of LPS-mediated inflammatory disease. The in vitro results above encouraged us to further explore the in vivo effect of galectin-3 in a shock model. WT and galectin-3−/− mice were given a minimal lethal dose of LPS. The development of shock syndrome and mortality were monitored at regular intervals. Galectin-3−/− mice were more susceptible to LPS-induced shock compared with the controls. No mice survived in the galectin-3−/− group 25 h post-LPS challenge, while 70% of WT mice were still alive (Fig. 5,A). Forty percent of the WT mice eventually completely recovered (Fig. 5,A). NO and TNF-α are critically involved in LPS shock (30, 31). We noted that galectin-3−/− mice produced more serum nitrite than did WT mice, and the levels of nitrite were further increased after LPS injection (Fig. 5,C). Consistent with these in vitro findings (Fig. 2), the elevated mortality rate in galectin-3−/− mice was also accompanied by the significantly enhanced serum TNF-α levels 4 h after LPS challenge (Fig. 5,D). As observed in vitro (Fig. 1,D), LPS also induced high levels of serum galectin-3 in WT but not in galectin-3−/− mice, indicating that the induction of galectin-3 may be involved in a feedback regulation of the LPS-mediated inflammatory response (Fig. 5 B).

FIGURE 5.

Galectin-3−/− mice are more susceptible to endotoxin shock. Groups of 10 WT and galectin-3−/− mice were challenged with LPS (25 mg/kg of body weight). A, The mortality rate was monitored regularly and represented as percentage of survival. B, Serum galectin-3 was detected by ELISA. C, Nitrite was measured by the Griess method (Biotium) and TNF-α by ELISA (D). ∗, p < 0.05; ∗∗, p < 0.01. Data are the means ± SEM of three individual experiments.

FIGURE 5.

Galectin-3−/− mice are more susceptible to endotoxin shock. Groups of 10 WT and galectin-3−/− mice were challenged with LPS (25 mg/kg of body weight). A, The mortality rate was monitored regularly and represented as percentage of survival. B, Serum galectin-3 was detected by ELISA. C, Nitrite was measured by the Griess method (Biotium) and TNF-α by ELISA (D). ∗, p < 0.05; ∗∗, p < 0.01. Data are the means ± SEM of three individual experiments.

Close modal

LPS is a major PAMP of Salmonella and plays a critical role in salmonellosis (1, 2, 32). To further understand the in vivo regulatory role of galectin-3 in LPS function, we next evaluated its role in Salmonella infection. First, we found that galectin-3 binds to Salmonella (Fig. 6,A), suggesting that galectin-3 may play a direct role in Salmonella-mediated infection. To test its role in vivo, the galectin-3−/− and WT control mice were infected orally with Salmonella typhimurium as described (27, 28). Salmonella replicated gradually in the organs of WT mice, which peaked at day 10, then declined and were completely eliminated from the organs after 20 days of infection (Fig. 6,B). By contrast, in galectin-3−/− mice, the bacteria replicated poorly, showing dramatically reduced bacterial burdens in all organs tested, especially in the spleen and draining lymph node compared with that of WT mice (Fig. 6 B). We found that there was no clear difference in the bacterial uptake and viability between the galectin-3−/− and WT BMMs (data not shown), suggesting that the reduced bacterial numbers in galectin-3-deficient mice was not attributed to an attenuated ability of bacterial phagocytosis or an enhanced susceptibility to apoptosis after infection.

FIGURE 6.

Galectin-3−/− mice are highly resistant to Salmonella infection. A, Fixed Salmonella-GFP (Sal-GFP) bacteria were incubated with or without galectin-3 for 30 min before being labeled with anti-galectin-3 or isotype control followed by anti-IgG-Cy3 on slides and then visualized by confocal microscopy. B, Groups of 10 WT and galectin-3−/− mice were infected with Salmonella (BRD509) (1 × 109) orally and bacterial numbers in the infected organs were determined at regular intervals. C, The serum nitrite was detected by the Griess method and H2O2 levels by a peroxide assay kit. D, The mesenteric lymph node cells (2 × 106 cells/ml) were cultured with Salmonella Ag (Ag, 107/ml) or medium alone (Med) for 72 h. The cytokine levels in the supernatant were assessed by ELISA. E, The serum level of galectin-3 was detected by ELISA. ∗, p < 0.05; ∗∗, p < 0.01. Data are the means ± SEM and are representative of three individual experiments.

FIGURE 6.

Galectin-3−/− mice are highly resistant to Salmonella infection. A, Fixed Salmonella-GFP (Sal-GFP) bacteria were incubated with or without galectin-3 for 30 min before being labeled with anti-galectin-3 or isotype control followed by anti-IgG-Cy3 on slides and then visualized by confocal microscopy. B, Groups of 10 WT and galectin-3−/− mice were infected with Salmonella (BRD509) (1 × 109) orally and bacterial numbers in the infected organs were determined at regular intervals. C, The serum nitrite was detected by the Griess method and H2O2 levels by a peroxide assay kit. D, The mesenteric lymph node cells (2 × 106 cells/ml) were cultured with Salmonella Ag (Ag, 107/ml) or medium alone (Med) for 72 h. The cytokine levels in the supernatant were assessed by ELISA. E, The serum level of galectin-3 was detected by ELISA. ∗, p < 0.05; ∗∗, p < 0.01. Data are the means ± SEM and are representative of three individual experiments.

Close modal

It is well known that NO and H2O2 are the major bactericidal reactive oxygen species (ROS) (33, 34). We noted that galectin-3−/− mice produced more serum nitrite and H2O2 than did WT mice (Fig. 6,C). After Salmonella infection, the serum levels of nitrite and superoxide in galectin-3−/− mice were further enhanced on days 5 and 8 compared with WT mice (Fig. 6,C). Importantly, the magnitude of NO and H2O2 levels was the reciprocal of the bacterial counts in the infected organs of the null and WT mice. In agreement with the LPS shock result (Fig. 5,B), Salmonella infection also elevated serum galectin-3 production (Fig. 6,E), whose levels were correlated with bacteria number in the organs (Fig. 6 B).

A Th1 response and Th1 cytokines IFN-γ and TNF-α are also required by the host to clear Salmonella infections (35, 36). We further verified the role of Salmonella-induced IFN-γ and TNF-α production in host protection. The draining lymph node and spleen cells from the infected mice were treated with medium alone or Salmonella Ags for 72 h. No significant difference in the T cell proliferation was observed (data not shown). However, 10 days after infection, the draining LN cells from galectin-3−/− mice produced strikingly higher levels of Salmonella-specific IFN-γ and TNF-α than did the cells from WT mice (Fig. 6,D). No significant levels of IL-4, IL-5, and IL-10 were detected in the cultures (data not shown), indicating that galectin-3−/− mice developed an increased Th1 response. The Salmonella-specific Th1-like response was sustained for at least 7 days until the bacterial infection had been eliminated. Interestingly, the levels of Ag-specific Th1 cytokine production from both WT and galectin-3−/− cells were the reciprocal of the organ bacterial load, which indicated that the reduced replication of Salmonella in galectin-3−/− mice may also be due to an elevated protective Th1-type response (Fig. 6, B and D).

It has been reported that galectin-3 also binds to several macrophage membranous proteins, including CD98 and CD13 (37, 38). To investigate the possibility that galectin-3 may modulate LPS function via alternative interaction with other membrane proteins, galectin-3−/− BMMs were preincubated with recombinant galectin-3 for 1 h, then washed with prewarmed medium or left unwashed before being activated with LPS. We found that removing the unbound galectin-3 from the culture abolished its regulatory effect on LPS, suggesting that crosslinking of galectin-3 with other ligands on BMMs is not responsible for the function of galectin-3 (Fig. 7,A). Furthermore, anti-CD98 or CD13 antibody-treated galectin-3−/− BMMs produced similar levels of cytokines as did the untreated or IgG-treated cells in response to LPS or LPS preincubated with galectin-3 (Fig. 7 B). Additionally, we also found that replacing galectin-3-containing supernatants from overnight-cultured WT BMMs with fresh medium significantly enhanced LPS-induced inflammatory cytokine production by the cells (data not shown). Thus, although the interaction of galectin-3 with other surface ligands is important for macrophage function in some contexts, this is unlikely to be responsible for the regulatory effect of galectin-3 on LPS function in our experimental setting.

FIGURE 7.

Galectin-3 interacting with other surface ligands on BMMs is unlikely to be responsible for its regulatory effect on LPS. A, WT and galectin-3−/− BMMs were preincubated with or without galectin-3 (0.5 μg/ml) for 1 h and then washed or unwashed with medium before being activated with or without LPS (0.1 μg/ml). B, WT and galectin-3−/− BMMs were pretreated with or without anti-CD98 (eBioscience), CD13 (Serotec) Ab, or control IgG (2 μg/ml) for 1 h before being stimulated with or without LPS alone or LPS-preincubated with galectin-3. All cells were cultured for 8 h and supernatant levels of TNF-α and IL-6 were measured by ELISA. Data are presented as means ± SEM and are representative of two independent experiments. ∗, p < 0.05 compared with control.

FIGURE 7.

Galectin-3 interacting with other surface ligands on BMMs is unlikely to be responsible for its regulatory effect on LPS. A, WT and galectin-3−/− BMMs were preincubated with or without galectin-3 (0.5 μg/ml) for 1 h and then washed or unwashed with medium before being activated with or without LPS (0.1 μg/ml). B, WT and galectin-3−/− BMMs were pretreated with or without anti-CD98 (eBioscience), CD13 (Serotec) Ab, or control IgG (2 μg/ml) for 1 h before being stimulated with or without LPS alone or LPS-preincubated with galectin-3. All cells were cultured for 8 h and supernatant levels of TNF-α and IL-6 were measured by ELISA. Data are presented as means ± SEM and are representative of two independent experiments. ∗, p < 0.05 compared with control.

Close modal

We have demonstrated herein that galectin-3 directly interacts with LPS and blocks LPS-induced inflammatory response. LPS-stimulated galectin-3−/− macrophages elevated LPS signaling and secreted higher levels of inflammatory cytokines compared with WT cells. This enhanced cytokine profile and signaling by the gene knockout cells could be normalized by recombinant galectin-3 protein. In reverse, blocking galectin-3 with a neutralizing Ab or its high-affinity ligand, β-lactose, led to a dramatically increased response to LPS and inflammatory cytokine production by WT macrophages. In vivo, galectin-3−/− mice were more vulnerable to LPS-induced shock, exhibiting a significantly enhanced mortality rate that was accompanied by an elevated profile of proinflammatory cytokines and NO production. Finally, galectin-3 binds to Salmonella. Salmonella-infected galectin-3-deficient mice increased Th1 response and NO burst, which protected the mice against Salmonella infection. Therefore, galectin-3 is a novel repressor and checkpoint for LPS function and LPS-mediated disease.

It is still not entirely known how galectin-3 represses LPS function. Galectin-3 is the only chimera lectin in the family that has a unique ability to interact with both carbohydrate and noncarbohydrate ligands (10). The lipid A part of LPS is essential for LPS-mediated bioactivity, and galectin-3 is capable of interacting with lipid A through its N-terminal (25). We found that galectin-3−/− cells were hypersensitive to lipid A stimulation (Figs. 2 and 3). Therefore, it is possible that galectin-3 blocks LPS function by binding to the lipid A portion of LPS. However, its C-terminal CRD may also contribute to the effect of galectin-3 by binding with β-lactose of LPS. Indeed, β-lactose, which antagonizes the interaction between galectin-3 and the β-lactose of LPS, dramatically enhanced LPS-induced inflammatory cytokine production by the WT macrophage (Fig. 3,D). Therefore, galectin-3 may interfere with LPS function by interacting with LPS via both lipid A and β-lactose portions (25). Since no obvious change has been found in the expression and function of LPS receptor complex related to the galectin-3 deficiency, it is likely that galectin-3 affects LPS function by binding to LPS and subsequently interferes with the interaction of LPS with its receptor complex, thereby repressing the downstream signaling cascades of LPS (Fig. 4).

It is noteworthy that the intracellular, membrane-bound, and secreted galectin-3 may all contribute to LPS regulation but by different mechanisms. The serum galectin-3 may antagonize LPS function during an acute bacterial infection via both N- and C-terminals. The membrane-bound molecule, which is coexpressed with TLR4, could also block LPS function by interacting with the lipid A motif of LPS through its free N-terminal. Moreover, the intracellular galectin-3 may play a critical role in intracellular bacterial infection. Thus, given that it is constantly expressed by all cells in high levels and induced by LPS stimulation, galectin-3 may represent a critical factor against excessive LPS-triggered inflammation.

NO and H2O2 play an important role in inflammation and are the major bactericidal ROS (33). We noted that galectin-3−/− mice spontaneously produced more serum nitrite and H2O2 than did WT mice (Figs. 5,C and 6,C). After LPS challenge and Salmonella infection, the serum levels of NO and superoxide in galectin-3−/− mice were further enhanced compared with WT mice (Figs. 5,C and 6,C). Furthermore, these ROS appeared much earlier than inflammatory cytokine production (Fig. 6, C and D), further suggesting that galectin-3 may directly modulate NO and superoxide production. Consistent with these observations, LPS and Salmonella also induced increased serum galectin-3 production in WT mice whose kinetic expression was inversely correlated with ROS and inflammatory cytokine production (Figs. 5 and 6). However, its production was positively correlated with survival rate in LPS shock and the bacterial load in the Salmonella infection model (Figs. 5 and 6). These results further suggest that the purpose of the galectin-3 induced by LPS was to suppress the subsequent inflammatory gene activation, including NO and superoxide production. Whereas the molecular mechanism by which galectin-3 affects NO and superoxide is still unknown, since JNK and p38 are associated with the induction of NO and superoxide (39, 40, 41) and galectin-3−/− cells enhanced LPS-induced JNK and p38 activity (Fig. 4), it is possible that galectin-3 modulates NO and superoxide by inhibiting these components of TLR4 signaling pathway. More work is needed to clarify this issue.

It is intriguing that mice lacking galectin-3 showed enhanced Th1 skewing during Salmonella infection. This is consistent with other reports that galectin-3 deficiency promotes a predominantly Th1 response (15, 16, 17, 18). However, the mechanism is not fully understood. Our results may provide an explanation for the preferential induction of inflammatory and Th1 responses in the mice, at least in LPS-mediated responses and Salmonella infection. Since LPS is a potent Th1 inducer by promoting IL-12 production (42), and, consequently, losing the regulation of galectin-3, APCs may become hypersensitive to LPS and produce more IL-12, which ultimately affects the Th1 polarization (43). Given that LPS is constantly present in the commensal flora and is widely distributed in the environment, this regulatory pathway may be critical for homeostasis during infection and may contribute to the pathogenesis of other inflammatory and Th1-mediated diseases (15, 16, 17, 18).

The outcome of salmonellosis depends on the interaction between the host and bacteria. Salmonella have evolved the ability to escape immune surveillance and expand in the host (44, 45). Our results demonstrate that galectin-3 is required for the improved survival of Salmonella in vivo via at least two mechanisms: first, by suppressing ROS release, and second, by inhibiting Th1 responses. While the detailed mechanism is still unresolved, the regulatory role of galectin-3 in Salmonella infection may be in addition to its binding of LPS, since it also partially modulates TLR2 function (Fig. 2 A). BLP and PGN are abundantly produced by all bacteria and are important PAMPs in bacterial-mediated diseases, including salmonellosis (46, 47). It is unclear at present why galectin-3 selectively modulates some but not all TLR agonists’ function. Since LPS contains the ligands for galectin-3, it is likely that only the microbial product that expresses galectin-3 ligand can be regulated. This hypothesis is under investigation. Nevertheless, it is likely that bacteria utilize host galectin-3 to mask their surface danger signals, LPS, and perhaps BLP and PGN as well, thereby silencing the immune system and allowing Salmonella to multiply. Thus, galectin-3 may play an important role in host-bacteria interaction and may be a new target for therapeutic interventions in bacteria infections. Finally, β-galactosides are common surface components associated with many infectious agents, and as such will be important to determine what role galectin-3 may play in other infections.

We thank Dr. David Holden (Department of Molecular Microbiology and Infection, Imperial College London, U.K.) for kindly providing the Salmonella-GFP strain and Dr. Charles McSharry (Division of Immunology, Infection and Inflammation, University of Glasgow, U.K.) for critical reading of the manuscript.

The authors have no financial conflicts of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This study received financial support from the Medical Research Council U.K., the Wellcome Trust, and the Chief Scientist’s Office, Scotland.

3

Abbreviations used in this paper: PAMP, pathogen-associated molecular pattern; BLP, bacterial lipopeptide; BMM, bone marrow-derived macrophage; CRD, carbohydrate recognition domain; PGN, peptidoglycan; ROS, reactive oxygen species; WT, wild type.

1
Ulevitch, R. J., P. S. Tobias.
1995
. Receptor-dependent mechanisms of cell stimulation by bacterial endotoxin.
Annu. Rev. Immunol.
13
:
437
-457.
2
Raetz, C. R. H..
1996
. Bacterial lipopolysaccharides: a remarkable family of bioactive macroamphiphiles. F. C. Neidhardt, ed.
Escherichia coli and Salmonella
1035
-1063. American Society for Microbiology, Washington, D.C..
3
Jiang, Z., P. Georgel, X. Du, L. Shamel, S. Sovath, S. Mudd, M. Huber, C. Kalis, S. Keck, C. Galanos, et al
2005
. CD14 is required for MyD88-independent LPS signaling.
Nat. Immunol.
6
:
565
-570.
4
Takeda, K., T. Kaisho, S. Akira.
2003
. Toll-like receptors.
Annu. Rev. Immunol.
21
:
335
-376.
5
Miller, S. I., R. K. Ernst, M. W. Bader.
2005
. LPS, TLR4 and infectious disease diversity.
Nat. Rev. Microbiol.
3
:
36
-46.
6
Liew, F. Y., D. Xu, E. K. Brint, L. A. O'Neill.
2005
. Negative regulation of Toll-like receptor-mediated immune responses.
Nat. Rev. Immunol.
5
:
446
-458.
7
Ho, M. K., T. A. Springer.
1982
. Mac-2, a novel 32,000 Mr mouse macrophage subpopulation-specific antigen defined by monoclonal antibodies.
J. Immunol.
128
:
1221
-1228.
8
Brewer, C. F., M. C. Miceli, L. G. Baum.
2002
. Clusters, bundles, arrays and lattices: novel mechanisms for lectin-saccharide-mediated cellular interactions.
Curr. Opin. Struct. Biol.
12
:
616
-623.
9
Liu, F. T..
2005
. Regulatory roles of galectins in the immune response.
Int. Arch. Allergy Immunol.
136
:
385
-400.
10
Rabinovich, G. A., L. G. Baum, N. Tinari, R. Paganelli, C. Natoli, F. T. Liu, S. Iacobelli.
2002
. Galectins and their ligands: amplifiers, silencers or tuners of the inflammatory response?.
Trends Immunol.
23
:
313
-320.
11
Hughes, R. C..
1999
. Secretion of the galectin family of mammalian carbohydrate-binding proteins.
Biochim. Biophys. Acta
1473
:
172
-185.
12
Cherayil, B. J., S. J. Weiner, S. Pillai.
1989
. The Mac-2 antigen is a galactose-specific lectin that binds IgE.
J. Exp. Med.
170
:
1959
-1972.
13
Sano, H., D. K. Hsu, L. Yu, J. R. Apgar, I. Kuwabara, T. Yamanaka, M. Hirashima, F. T. Liu.
2000
. Human galectin-3 is a novel chemoattractant for monocytes and macrophages.
J. Immunol.
165
:
2156
-2164.
14
Henderson, N. C., A. C. Mackinnon, S. L. Farnworth, F. Poirier, F. P. Russo, J. P. Iredale, C. Haslett, K. J. Simpson, T. Sethi.
2006
. Galectin-3 regulates myofibroblast activation and hepatic fibrosis.
Proc. Natl. Acad. Sci. USA
103
:
5060
-5065.
15
Pugliese, G., F. Pricci, C. Iacobini, G. Leto, L. Amadio, P. Barsotti, L. Frigeri, D. K. Hsu, H. Vlassara, et al
2001
. Accelerated diabetic glomerulopathy in galectin-3/AGE receptor 3 knockout mice.
FASEB J.
15
:
2471
-2479.
16
Zuberi, R. I., D. K. Hsu, O. Kalayci, H. Y. Chen, H. K. Sheldon, L. Yu, J. R. Apgar, T. Kawakami, C. M. Lilly, F. T. Liu.
2004
. Critical role for galectin-3 in airway inflammation and bronchial hyperresponsiveness in a murine model of asthma.
Am. J. Pathol.
165
:
2045
-2053.
17
Bernardes, E. S., N. M. Silva, L. P. Ruas, J. R. Mineo, A. M. Loyola, D. K. Hsu, F. T. Liu, R. Chammas, M. C. Roque-Barreira.
2006
. Toxoplasma gondii infection reveals a novel regulatory role for galectin-3 in the interface of innate and adaptive immunity.
Am. J. Pathol.
168
:
1910
-1920.
18
Nomoto, K., K. Tsuneyama, H. O. Abdel Aziz, H. Takahashi, Y. Murai, Z. G. Cui, M. Fujimoto, I. Kato, K. Hiraga, D. K. Hsu, et al
2006
. Disrupted galectin-3 causes non-alcoholic fatty liver disease in male mice.
J. Pathol.
210
:
469
-477.
19
Mandrell, R. E., M. A. Apicella, R. Lindstedt, H. Leffler.
1994
. Possible interaction between animal lectins and bacterial carbohydrates.
Methods Enzymol.
236
:
231
-254.
20
Gupta, S. K., S. Masinick, M. Garrett, L. D. Hazlett.
1997
. Pseudomonas aeruginosa lipopolysaccharide binds galectin-3 and other human corneal epithelial proteins.
Infect. Immun.
65
:
2747
-2753.
21
Pelletier, I., S. Sato.
2002
. Specific recognition and cleavage of galectin-3 by Leishmania major through species-specific polygalactose epitope.
J. Biol. Chem.
277
:
17663
-17670.
22
van den Berg, T. K., H. Honing, N. Franke, A. van Remoortere, W. E. Schiphorst, F. T. Liu, A. M. Deelder, R. D. Cummings, C. H. Hokke, I. van Die.
2004
. LacdiNAc-glycans constitute a parasite pattern for galectin-3-mediated immune recognition.
J. Immunol.
173
:
1902
-1907.
23
Kohatsu, L., D. K. Hsu, A. G. Jegalian, F. T. Liu, L. G. Baum.
2006
. Galectin-3 induces death of Candida species expressing specific β-1,2-linked mannans.
J. Immunol.
177
:
4718
-4726.
24
Fradin, C., D. Poulain, T. Jouault.
2000
. β-1,2-linked oligomannosides from Candida albicans bind to a 32-kilodalton macrophage membrane protein homologous to the mammalian lectin galectin-3.
Infect. Immun.
68
:
4391
-4398.
25
Mey, A., H. Leffler, Z. Hmama, G. Normier, J. P. Revillard.
1996
. The animal lectin galectin-3 interacts with bacterial lipopolysaccharides via two independent sites.
J. Immunol.
156
:
1572
-1577.
26
Sweet, M. J., B. P. Leung, D. Kang, M. Sogaard, K. Schulz, V. Trajkovic, C. C. Campbell, D. Xu, F. Y. Liew.
2001
. A novel pathway regulating lipopolysaccharide-induced shock by ST2/T1 via inhibition of Toll-like receptor 4 expression.
J. Immunol.
166
:
6633
-6639.
27
Xu, D., S. J. McSorley, L. Tetley, S. Chatfield, G. Dougan, W. L. Chan, A. Satoskar, J. R. David, F. Y. Liew.
1998
. Protective effect on Leishmania major infection of migration inhibitory factor, TNF-α, and IFN-γ administered orally via attenuated Salmonella typhimurium.
J. Immunol.
160
:
1285
-1289.
28
Xu, D., S. J. McSorley, S. N. Chatfield, G. Dougan, F. Y. Liew.
1995
. Protection against Leishmania major infection in genetically susceptible BALB/c mice by gp63 delivered orally in attenuated Salmonella typhimurium (AroAAroD).
Immunology
85
:
1
-7.
29
Ruberto, F., F. Pugliese, A. D'Alio, S. Martelli, K. Bruno, V. Marcellino, S. Perrella, A. Cappannoli, V. Mazzarino, A. Tosi, et al
2007
. Clinical effects of use of polymyxin B fixed on fibers in liver transplant patients with severe sepsis or septic shock.
Transplant. Proc.
39
:
1953
-1955.
30
MacMicking, J. D., C. Nathan, G. Hom, N. Chartrain, D. S. Fletcher, M. Trumbauer, K. Stevens, Q. W. Xie, K. Sokol, N. Hutchinson, et al
1995
. Altered responses to bacterial infection and endotoxic shock in mice lacking inducible nitric oxide synthase.
Cell
81
:
641
-650.
31
Beutler, B., I. W. Milsark, A. C. Cerami.
1985
. Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effect of endotoxin.
Science
229
:
869
-871.
32
Vazquez-Torres, A., B. A. Vallance, M. A. Bergman, B. B. Finlay, B. T. Cookson, J. Jones-Carson, F. C. Fang.
2004
. Toll-like receptor 4 dependence of innate and adaptive immunity to Salmonella: importance of the Kupffer cell network.
J. Immunol.
172
:
6202
-6208.
33
Mastroeni, P., A. Vazquez-Torres, F. C. Fang, Y. Xu, S. Khan, C. E. Hormaeche, G. Dougan.
2000
. Antimicrobial actions of the NADPH phagocyte oxidase and inducible nitric oxide synthase in experimental salmonellosis: II. Effects on microbial proliferation and host survival in vivo.
J. Exp. Med.
192
:
237
-248.
34
Shiloh, M. U., J. D. MacMicking, S. Nicholson, J. E. Brause, S. Potter, M. Marino, F. Fang, M. Dinauer, C. Nathan.
1999
. Phenotype of mice and macrophages deficient in both phagocyte oxidase and inducible nitric oxide synthase.
Immunity
10
:
29
-38.
35
Lehmann, J., S. Bellmann, C. Werner, R. Schroder, N. Schutze, G. Alber.
2001
. IL-12p40-dependent agonistic effects on the development of protective innate and adaptive immunity against Salmonella enteritidis.
J. Immunol.
167
:
5304
-5315.
36
Vazquez-Torres, A., G. Fantuzzi, C. K. Edwards, III, C. A. Dinarello, F. C. Fang.
2001
. Defective localization of the NADPH phagocyte oxidase to Salmonella-containing phagosomes in tumor necrosis factor p55 receptor-deficient macrophages.
Proc. Natl. Acad. Sci. USA
98
:
2561
-2565.
37
Mackinnon, A. C., S. L. Farnworth, P. S. Hodkinson, N. C. Henderson, K. M. Atkinson, H. Leffler, U. J. Nilsson, C. Haslett, S. J. Forbes, T. Sethi.
2008
. Regulation of alternative macrophage activation by galectin-3.
J. Immunol.
180
:
2650
-2658.
38
Mina-Osorio, P., I. Soto-Cruz, E. Ortega.
2007
. A role for galectin-3 in CD13-mediated homotypic aggregation of monocytes.
Biochem. Biophys. Res. Commun.
353
:
605
-610.
39
Hsu, Y. W., K. H. Chi, W. C. Huang, W. W. Lin.
2001
. Ceramide inhibits lipopolysaccharide-mediated nitric oxide synthase and cyclooxygenase-2 induction in macrophages: effects on protein kinases and transcription factors.
J. Immunol.
166
:
5388
-5397.
40
Uto, T., M. Fujii, D. X. Hou.
2005
. 6-(Methylsulfinyl)hexyl isothiocyanate suppresses inducible nitric oxide synthase expression through the inhibition of Janus kinase 2-mediated JNK pathway in lipopolysaccharide-activated murine macrophages.
Biochem. Pharmacol.
70
:
1211
-1221.
41
Lin, M. W., L. T. Tsao, L. C. Chang, Y. L. Chen, L. J. Huang, S. C. Kuo, C. C. Tzeng, M. R. Lee, J. P. Wang.
2007
. Inhibition of lipopolysaccharide-stimulated NO production by a novel synthetic compound CYL-4d in RAW 264.7 macrophages involving the blockade of MEK4/JNK/AP-1 pathway.
Biochem. Pharmacol.
73
:
1796
-1806.
42
Trinchieri, G..
2003
. Interleukin-12 and the regulation of innate resistance and adaptive immunity.
Nat. Rev. Immunol.
3
:
133
-146.
43
Hsieh, C. S., S. E. Macatonia, C. S. Tripp, S. F. Wolf, A. O'Garra, K. M. Murphy.
1993
. Development of TH1 CD4+ T cells through IL-12 produced by Listeria-induced macrophages.
Science
260
:
547
-549.
44
Galan, J. E., H. Wolf-Watz.
2006
. Protein delivery into eukaryotic cells by type III secretion machines.
Nature
444
:
567
-573.
45
Waterman, S. R., D. W. Holden.
2003
. Functions and effectors of the Salmonella pathogenicity island 2 type III secretion system.
Cell. Microbiol.
5
:
501
-511.
46
O'Brien, G. C., J. H. Wang, H. P. Redmond.
2005
. Bacterial lipoprotein induces resistance to Gram-negative sepsis in TLR4-deficient mice via enhanced bacterial clearance.
J. Immunol.
174
:
1020
-1026.
47
Quintela, J. C., M. A. de Pedro, P. Zollner, G. Allmaier, F. Garcia-del Portillo.
1997
. Peptidoglycan structure of Salmonella typhimurium growing within cultured mammalian cells.
Mol. Microbiol.
23
:
693
-704.