SslE (YghJ), a Cell-Associated and Secreted Lipoprotein of Neonatal Septicemic Escherichia coli, Induces Toll-Like Receptor 2-Dependent Macrophage Activation and Proinflammation through NF-κB and MAP Kinase Signaling

SslE (YghJ), a cell surface-associated and secreted lipoprotein, was identified as a potential vaccine candidate for extraintestinal pathogenic Escherichia coli, providing nearly complete protection from sepsis in a mouse model. We earlier found that SslE from neonatal septicemic E. coli could trigger the secretion of various proinflammatory cytokines in murine macrophages, the signaling pathway of which is still obscure.

ABSTRACT

SslE (YghJ), a cell surface-associated and secreted lipoprotein, was identified as a potential vaccine candidate for extraintestinal pathogenic Escherichia coli, providing nearly complete protection from sepsis in a mouse model. We earlier found that SslE from neonatal septicemic E. coli could trigger the secretion of various proinflammatory cytokines in murine macrophages, the signaling pathway of which is still obscure. In this study, we showed that SslE specifically binds to Toll-like receptor 2 (TLR2)/TLR1 heterodimers and recruits downstream adaptors MyD88, TIRAP, and TRAF6. In addition, SslE stimulates nuclear translocation of NF-κB and activates different mitogen-activated protein (MAP) kinase signaling cascades specific to the secretion of each cytokine in murine macrophages, which becomes impaired in TLR2 small interfering RNA (siRNA)-transfected cells and in cells blocked with a monoclonal antibody (MAb) against TLR2, suggesting the involvement of TLR2 in NF-κB and MAP kinase activation and subsequent cytokine secretion. Furthermore, our study is the first to show that SslE can stimulate TLR2-dependent production of other proinflammatory hallmarks, such as reactive nitrogen and oxygen species as well as type 1 chemokines, which contribute to the anti-infection immune response of the host. Also, the overexpression of major histocompatibility complex class II (MHC II) and other costimulatory molecules (CD80 and CD86) in macrophages essentially indicates that SslE promotes macrophage activation and M1 polarization, which are crucial in framing the host's innate immune response to this protein, and hence, SslE could be a potent immunotherapeutic target against E. coli sepsis.

INTRODUCTION

Sepsis is the systemic inflammatory response of a host against an infection. Bacteriological isolation of an infectious agent from blood or cerebrospinal fluid in the neonatal period, i.e., the first 4 weeks of life, is known as neonatal septicemia (1). Among the broad spectrum of pathogens implicated in sepsis of newborns, Escherichia coli demands critical attention specifically in developing countries. E. coli has emerged as a major pathogen (24.5%) for sepsis-related mortality among infants with very low birth weight (2) and is considered to be the most important agent for early-onset neonatal sepsis (3, 4). Increasing resistance of septicemic isolates to the currently used antibiotics, even to the newest ones, is a serious holdup in treatment of neonatal sepsis (5). Hence, development of a potent and inexpensive vaccine would be of critical research importance.

In search for a novel vaccine candidate, Moriel et al. have identified ECOK1-3385 by reverse vaccinology, which is broadly represented in diverse E. coli pathotypes, including extraintestinal pathogenic E. coli (ExPEC), and has been found to confer almost complete protection in a murine model of sepsis, making it a potent immunogenic vaccine candidate for ExPEC strains causing sepsis (6, 7). This protein has been designated SslE, a cell-associated and secreted lipoprotein of E. coli (8) possessing an M60-like metalloprotease domain (9), and is secreted via a type II secretion system (T2SS), a well-known export apparatus of Gram-negative bacteria to deliver various proteins, including diverse virulence determinants (10). Among the two T2SSs of E. coli, the T2SSβ operon is composed of three genes (yghJ, pppA, and yghG). yghJ encodes the SslE protein (11). Hence, SslE was formerly named YghJ (12, 13) and was reported to be secreted from diverse intestinal pathogenic E. coli strains, including enterotoxigenic E. coli (ETEC) and enteropathogenic E. coli (EPEC), in which it has been found to contribute in the virulence of the producer organisms. In EPEC, it is required for biofilm formation and virulence (14). In ETEC, it was found to be actively involved in degradation of intestinal mucins, including MUC2, MUC3, and bovine submaxillary mucin, which facilitate E. coli penetration of mucus layer and enhances access to apical epithelial cells (8, 15, 16). Importantly, immunization with SslE was found to protect mice against both urinary tract infection (UTI) and intestinal infection, causing it to be proposed as a broadly protective E. coli vaccine antigen (8). Furthermore, in patients infected with ETEC, SslE was identified as an immunogenic antigen (17, 18). In a previous study, we have identified SslE from a clinical neonatal septicemic E. coli (NSEC) isolate (19). We further cloned, expressed, and purified SslE and showed that SslE could trigger the production of various proinflammatory cytokines in murine macrophages (19). However, the signaling pathways involved in SslE-mediated proinflammation are yet unexplored. Moreover, as SslE is considered a potent antigen candidate for a vaccine against NSEC, a detailed understanding of the molecular mechanism of SslE-mediated activation of the innate immune defense during an NSEC infection would provide new insight for its use as an immunotherapeutic target against E. coli sepsis.

In neonates, the innate immune system is primarily destined to be at the forefront of defense to an infection (20, 21). Toll-like receptors (TLRs) are critical in instigating the innate immune response to invading pathogens. TLRs, an evolutionarily conserved family of pattern recognition receptors (PRRs), are type I transmembrane proteins of the interleukin-1 (IL-1) receptor family which possess an N-terminal leucine-rich repeat (LRR) domain for ligand binding, a single transmembrane domain, and a C-terminal intracellular signaling Toll/IL-1 receptor (TIR) domain and are critical in the host innate immune defense. So far, 13 mammalian TLRs have been identified (10 in humans and 13 in mice), each having distinct ligand specificity (22, 23). TLRs are widely expressed in many cell types. However, most of the TLRs are expressed in hematopoietically derived sentinel cells such as neutrophils, dendritic cells, and macrophages (22).

Macrophages, the sentinel of innate immune defense, are recruited to the site of infection for efficient and robust elimination of the pathogens (24, 25). Macrophages express TLRs which become activated by specific pathogen-associated molecular patterns (PAMPs) (22, 26) to trigger activation of mitogen-activated protein (MAP) kinase and NF-κB signaling pathways (23, 27, 28). This leads to the induction of various genes that function in host defense, including proinflammatory cytokines, chemokines (29, 30), reactive oxygen and nitrogen species, major histocompatibility complex class II (MHC II), and costimulatory molecules (31, 32), reflecting macrophage activation and M1 polarization. Macrophages undergo two kinds of functional polarization, M1 or M2 (32). M1 macrophages secrete proinflammatory cytokines such as IL-12, tumor necrosis factor alpha (TNF-α), IL-6, and IL-1 (31, 32) and also proinflammatory chemokines such as CCL3/MIP-1α, CCL4/MIP-1β, CCL5/RANTES, CXCL9/MIG, and CXCL-10/IP-10 (30), whereas M2 macrophages are associated with the secretion of anti-inflammatory cytokines such as IL-10 (31, 32) and a distinct chemokine repertoire such as TARC/CCL17, MDC/CCL22, eotaxin-2/CCL24, etc. (30). The proinflammatory cytokines and chemokines secreted from M1 macrophages attract several types of immune cells, including leukocytes and T cells, to the site of infection in order to clear invading microorganisms. Hence, M1 or proinflammatory macrophages have potent microbicidal activity and participate in the host defense against invading pathogens. The upregulated MHC II and costimulatory molecules on M1 macrophages subsequently activate both T and B lymphocytes, linking the innate response with adaptive immunity (33). All these factors together make TLR ligands ideal vaccine candidates (34).

In the present study, we have identified the cell surface receptor of SslE and also investigated the signaling mechanism for SslE-induced proinflammatory cytokine secretion in mouse macrophages, which we have further validated in HEK293 cells. We found that recombinant SslE (rSslE) from NSEC isolate activates TLR2 and subsequent NF-κB and MAP kinase signaling pathways to secrete proinflammatory cytokines in mouse macrophages. Furthermore, our study is the first to show that SslE can trigger production of various other proinflammatory markers and increases expression of MHC II and costimulatory molecules such as CD80 and CD86 on mouse macrophages. All these results clearly indicate that SslE induces macrophage activation and polarization toward the M1 or proinflammatory type, which is critical in framing the host's innate immune defense.

RESULTS

SslE induces overexpression of TLR2 and TLR1 in RAW 264.7 cells and specifically interacts with TLR2/TLR1 heterodimers through the canonical pathway.

We intended to identify the cell surface receptor of SslE. As SslE is a secreted lipoprotein and lipoproteins are potent agonists for TLR2, we speculated that being a lipoprotein, SslE may signal through TLR2. First, we studied whether SslE induces overexpression of TLR2. To do that, RAW 264.7 cells were inoculated or not with SslE and left for 6 h, and then the expression levels of different TLRs in control and treated cells were analyzed and compared by Western blotting, which showed an increased level of expression of TLR2 in SslE-treated cells compared to the untreated controls (Fig. 1A). It is a well-known fact that TLR2 signaling depends on the coexpression of other TLRs, either TLR1 or TLR6, the expression levels of which in untreated and SslE treated RAW 264.7 cells were also compared. The expression of TLR1 was upregulated in treated cells, whereas the expression level of TLR6 remained unchanged between control and treated cells (Fig. 1A). As lipopolysaccharide (LPS) is common contaminant of recombinant His-tagged proteins and is a known ligand for TLR4, expression of TLR4 in control and treated cells was also checked, which showed no difference in expression of TLR4 (Fig. 1A), indicating that SslE-mediated overexpression of TLR2 is specific and not due to any contaminating LPS if present. Though all the treatments with SslE were done in the presence of polymyxin B (PMB) at a concentration of 50 μg/ml (see Materials and Methods), since PMB can bind and neutralize any contaminating LPS in SslE preparations, we have also shown the expression of TLR2 and TLR4 in both the presence and absence of PMB (Fig. 1B). We found that the level of expression of TLR2 was similar in both the presence and absence of PMB in SslE-treated cells and was higher than that in the untreated control, whereas TLR4 shows no alteration in expression level in the SslE-treated (with or without PMB) or untreated cells (Fig. 1B). Surface overexpression of TLR2 was further validated by flow cytometry, which showed a 5-fold-higher expression (5- ± 0.2-fold; P < 0.001) on RAW 264.7 cells treated with SslE (Fig. 1B).

FIG 1.

SslE induces overexpression of TLR2 and TLR1 in RAW 264.7 cells and specifically interacts with TLR2/TLR1 heterodimers via a MyD88-dependent pathway. To check overexpression of TLRs, cells were treated with 50 μg/ml polymyxin B (PMB) and after 1 h were stimulated or not with SslE (150 nM) and incubated for 6 h. (A) Cell lysates were immunoblotted with anti-TLR2, anti-TLR1, anti-TLR4, and anti-TLR6 MAbs. α-Tubulin was used as a loading control. (B) Immunoblots for TLR2 and TLR4 with α-tubulin as a control in the presence or absence of PMB. (C) Cells were stained with PE-conjugated anti-TLR2 antibody for 20 min at room temperature in the dark and analyzed by flow cytometry. (D and E) For immunoprecipitation, lysates from RAW 264.7 cells after 2 h of treatment were incubated with 2 μg of mouse anti-SslE antibody overnight at 4°C. Immune complexes were immobilized on protein A/G Plus agarose beads, and the proteins pulled down were detected by immunoblotting using anti-TLR2, anti-TLR1, anti-TLR6, and anti-TLR4 antibodies (D) and anti-MyD88, anti-TRAF6, and anti-TIRAP antibodies (E).

We next performed a pulldown assay using anti-SslE antibody to further establish the receptor specificity of SslE. Immunoblots showed that immobilized anti-SslE antibody specifically pulled down TLR2 and TLR1 (Fig. 1C) but not TLR4 and TLR6 (data not shown) when RAW 264.7 cells were treated with 150 nM SslE for 2 h. Hence, our results clearly indicate that SslE specifically interacts with TLR2/TLR1 heterodimers in RAW 264.7 cells.

We further performed immunoprecipitation to show the involvement of downstream adaptors in SslE-mediated TLR2 signaling. Generally, TLRs signal via two pathways, a MyD88-dependent pathway or a MyD88-independent pathway. The MyD88-dependent pathway employs other downstream adaptors, i.e., IRAK-4, TIRAP, and TRAF6. Western blots of the immunoprecipitated samples with anti-MyD88, anti-TIRAP, and anti-TRAF6 antibodies revealed that immobilized anti-SslE antibody pulled down MyD88, TIRAP, and TRAF6 in SslE-treated RAW 264.7 cells (Fig. 1D), indicating a MyD88-dependent TLR2 signaling pathway for SslE in murine macrophages.

SslE activates both MAP kinases and NF-κB via TLR2 signaling in RAW 264.7 cells.

As we intended to explore the signaling pathway involved in SslE-mediated proinflammation, we examined the levels of phosphorylation of different MAP kinases (JNK1/2, p38, and ERK1/2), the crucial mediators in proinflammation. Western blots after 3 h of treatment with SslE showed significant phosphorylation of p38, JNK1/2, and ERK1/2 compared to that in the control or untreated cells (Fig. 2A). Hence, our data indicated that SslE activates the p38, JNK1/2, and ERK1/2 MAP kinase signaling pathways.

FIG 2.

SslE induces activation of both NF-κB and MAP kinase signaling pathways in RAW 264.7 cells via TLR2. (A) RAW 264.7 cells were incubated with or without SslE (150 nM) for 3 h, and cell lysates were analyzed by Western blotting using anti-phospho-ERK1/2 (p-ERK1/2), anti-phospho-p38 (p-p38), and anti-phospho-JNK1/2 (p-JNK1/2) to study the phosphorylation/activation of ERK1/2, p38, and JNK1/2, respectively. Specific Abs for each unphosphorylated kinase and α-tubulin were used as loading controls. (B) Nuclear (N) and cytosolic (C) lysates were used to investigate the effect of SslE on the translocation of NF-κB from cytosol to nucleus in RAW 264.7 cells. Histone and α-tubulin were used as nuclear and cytosolic loading controls, respectively. (C) Immunofluorescence was done on SslE-treated and untreated cells to detect translocation of both p65 and p50 from cytosol to nucleus using anti-p65 and anti-p50 primary antibodies (1:200) followed by TRITC-conjugated secondary antibody (1:500). Nuclear staining was done with DAPI. (D) Total cell lysates were used to determine the time-dependent degradation of IκBα (0 to 5 h) with α-tubulin as a control. (E) RAW 264.7 cells were transiently transfected with TLR2 siRNA for 24 h and were then treated with SslE. Lysates from SslE-treated, untransfected cells (lane 1), SslE-treated scrambled siRNA-transfected cells (lane 2), and SslE-treated, TLR2 siRNA-transfected cells (lane 3) were analyzed by Western blotting with anti-TLR2 antibody to assess the transfection efficiency. α-Tubulin was used as a loading control. (F and G) Whole-cell lysate (F) and nuclear and cytosolic lysates (G) from SslE-treated, untransfected cells (lane 1), SslE-treated scrambled siRNA-transfected cells (lane 2), SslE-treated, TLR2 siRNA-transfected cells (lane 3), and untreated cells (lane 4) were analyzed by Western blotting using anti-phospho-ERK1/2 (p-ERK1/2), anti-phospho-p38 (p-p38), and anti-phospho-JNK1/2 antibodies (p-JNK1/2) (F) and anti-p65 antibody (G). (H) To determine the inhibition in nuclear translocation of p65 in TLR2 siRNA-transfected cells, immunofluorescence was done on SslE-treated TLR2 siRNA-transfected cells, SslE-treated scrambled siRNA-transfected cells, SslE-treated, and untreated cells using anti-p65 antibody (1:200) followed by TRITC-conjugated secondary antibody (1:500). Nuclear staining was done with DAPI.

We further investigated the activation of NF-κB, an important transcription factor in proinflammation which is generally sequestered by IκB in cytosol, is activated by phosphorylation and proteasomal degradation of IκB, and thereafter is translocated from the cytosol to the nucleus. Immunoblots showed translocation of both NF-κB subunits (p65 and p50) from the cytosol to the nucleus in SslE-treated cells, whereas the subunits were located primarily in the cytoplasm of the control cells (Fig. 2B). This was further confirmed by immunofluorescence, which revealed that p65 and p50 were localized mostly in the nucleus in SslE-treated cells, while in the control cells the levels of p65 and p50 were quite lower in the nucleus (Fig. 2C). Degradation of IκB-α was also checked by immunoblots, which showed degradation over a time period of 3 to 4 h and reappearance of IκB at 5 h poststimulation (Fig. 2D). All these data together demonstrated that SslE could induce the translocation of p50 and p65 from the cytosol to the nucleus and hence mediate activation of NF-κB.

We next determined whether SslE-induced activation of p38, ERK1/2, JNK1/2, and NF-κB is TLR2 mediated. For this, RAW 264.7 cells were transiently transfected with TLR2 small interfering RNA (siRNA) or scrambled siRNA or remained untransfected. Transfection efficiency was confirmed by comparing the expression of TLR2 in TLR2 siRNA-transfected, scrambled siRNA-transfected, and untransfected RAW 264.7 cells by immunoblotting (Fig. 2E). Transfection with TLR2 siRNA resulted in a maximal decrease of TLR2 expression at 24 h after transfection compared to that of the untransfected control cells. In contrast, the cells transfected with the scrambled siRNA showed no change in the basal level of TLR2 expression (Fig. 2E). At 24 h after transfection with TLR2 or scrambled siRNA, cells were treated in the presence or absence of SslE. Lysates were analyzed by Western blotting to determine the levels of phosphorylation of p38, ERK1/2, and JNK1/2. Interestingly, phosphorylation of p38, JNK1/2, and ERK1/2 was highly attenuated in SslE-treated TLR2 knockdown cells compared to that in the SslE-treated untransfected controls. Conversely, the cells transfected with scrambled siRNA and treated with SslE showed strong phosphorylation of p38, JNK1/2, and ERK1/2 (Fig. 2F). Similarly, nuclear translocation of p65 in SslE-treated, TLR2 siRNA-transfected cells was diminished remarkably compared to that in SslE-treated, scrambled siRNA-transfected cells, which showed nearly complete translocation of p65 from the cytosol to the nucleus as confirmed by Western blotting (Fig. 2G) and immunofluorescence (Fig. 2H). Hence, these observations signify that SslE is able to induce activation of MAP kinases and NF-κB in a TLR2-dependent manner.

TLR2 is essential in SslE-induced proinflammatory cytokine secretion, which involves the activation of both NF-κB and the MAP kinase signaling cascade.

In an attempt to establish the relative contributions of NF-κB and MAP kinases to the proinflammatory responses initiated by SslE, cells were treated with specific inhibitors against each pathway. An enzyme-linked immunosorbent assay (ELISA) was done to measure the levels of secreted IL-1α, IL-1β, and TNF-α in RAW 264.7 cells pretreated with SB203580 (a p38 inhibitor), SP600125 (a JNK1/2 inhibitor), U0126 (an ERK1/2 inhibitor), calphostin C (a protein kinase C alpha [PKC-α] inhibitor), and MG-132 (an NF-κB inhibitor). Production of IL-1α was inhibited significantly by MG-132 (80% inhibition) and SP600125 (75% inhibition) and partially by U0126 (54% inhibition). Likewise, production of IL-1β was inhibited mostly by MG-132 (85%) and SP600125 (71%) and partially by U0126 (55%). The secretion of TNF-α was remarkably blocked by MG-132 (80%) and SB203580 (71%) (Fig. 3A).

FIG 3.

SslE-mediated cytokine secretion is blocked by MAP kinase and NF-κB inhibitors and is TLR2 dependent. (A) RAW 264.7 cells were treated with JNK1/2 inhibitor SP600125 (10 μM), p38 inhibitor SB203580 (10 μM), ERK1/2 inhibitor U0126 (10 μM), NF-κB inhibitor MG-132 (2.5 μM), and PKC-α inhibitor calphostin C (3 μM) for 1 h before treatment with SslE (150 nM). Culture supernatants were collected after 8 h of treatment for measuring the levels of secreted TNF-α, IL-1α, and IL-1β by ELISA. (B) RAW 264.7 cells were preincubated with anti-TLR2 MAb 1 h prior to treatment with SslE for 8 h, and ELISA was done with culture supernatants to measure the levels of secreted cytokines. To further confirm the involvement of TLR2, RAW 264.7 cells were transfected with TLR2 siRNA and a control siRNA for 24 h before SslE treatment and ELISA for the cytokines. Results are expressed as mean ± SD from three different experiments. Significant differences between SslE treatment with or without each inhibitor were calculated using the Student t test (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

To illustrate the prerequisite for TLR2 in SslE-induced secretion of proinflammatory cytokines, ELISA was done with the culture supernatant of SslE-treated TLR2 knockdown cells and supernatant of RAW 264.7 cells preincubated with TLR2 MAb before treatment with SslE. Secretion of all the cytokines were significantly inhibited in anti-TLR2 MAb-treated cells (For IL-1α, inhibition was 86%; for IL-1β, inhibition was 82%; and for TNF-α, inhibition was 87%) (Fig. 3B). In addition, secretion of the cytokines was also abrogated in TLR2 knockdown cells. Secretion of IL-1α was reduced to 66%, that of IL-1β was reduced to 64.5% and that of TNF-α was reduced to 63% (Fig. 3B).

All these data together showed that SslE-induced cytokine secretion in mouse macrophages is TLR2 mediated with the involvement of mostly p38, JNK1/2, and NF-κB and partially ERK1/2.

SslE stimulates iNOS expression and NO and reactive oxygen species (ROS) production in a TLR2-dependent manner in RAW 264.7 cells.

Production of NO in macrophages following an infection is an important host defense response against various pathogens. In our study, we investigated the ability of SslE to stimulate the expression of inducible NO synthase (iNOS) and the concomitant production of one of the important proinflammatory mediators such as the short-lived free-radical NO. iNOS mRNA levels were measured by reverse transcription-PCR (RT-PCR) at different time intervals. Expression of iNOS was found to be upregulated on treatment with SslE in RAW 264.7 cells, with almost an optimum expression level reached at 6 h poststimulation (Fig. 4A). To correlate the increased mRNA levels with the production of NO, RAW 264.7 cells were incubated with different doses of SslE, and culture supernatants were collected after 12 h of treatment to measure the levels of accumulated nitrite, a stable oxidized end product of NO, by the Griess reaction. We found a dose-dependent increase in the production of NO with an optimum response at 500 nM SslE, and SslE at a dose of 62.5 nM was still able to trigger the production of NO (Fig. 4B).

FIG 4.

SslE induces iNOS expression and NO and ROS production in RAW 264.7 cells in a TLR2-dependent manner. (A) RAW 264.7 cells were treated with SslE for different time periods (0 to 9 h). Total RNA was extracted from SslE-treated cells, and RT-PCR was done to check the overexpression of iNOS. Expression of GAPDH was measured as a control. (B) RAW 264.7 cells were treated with different concentrations of SslE (0 to 500 nM) for 8 h. Culture supernatants were collected to measure the amount of accumulated nitrite, the stable end product of NO, by Griess assay. (C and D) RAW 264.7 cells before treatment with SslE (150 nM) were preincubated with JNK1/2 inhibitor SP600125 (10 μM), p38 inhibitor SB203580 (10 μM), ERK1/2 inhibitor U0126 (10 μM), NF-κB inhibitor MG-132 (2.5 μM), and PKC-α inhibitor calphostin C (3 μM) for 1 h. (C) Total RNA was extracted after 6 h from SslE-treated or untreated cells, and RT-PCR was done to check the expression of iNOS in the absence or presence of different inhibitors with corresponding GAPDH as a control. (D) Culture supernatants were collected after 12 h of treatment, and production of NO was measured by Griess assay. Results are expressed as mean ± SD from three different experiments. Significant differences between SslE treatment with or without each inhibitor were calculated using the Student t test (*, P < 0.05; **, P < 0.01; ***, P < 0.001). (E) RAW 264.7 cells were transiently transfected with TLR2 siRNA and scrambled siRNA before SslE treatment. The Griess assay was performed with the culture supernatants to measure the level of accumulated nitrite in order to determine the amount of NO produced in presence or absence of TLR2 siRNA. Significant differences between SslE treatment with or without TLR2 siRNA were calculated using the Student t test (*, P < 0.05; **, P < 0.01; ***, P < 0.001). (F) RAW 264.7 cells (TLR2 siRNA transfected, control siRNA transfected, and untransfected) were incubated in presence or absence of SslE for 2 h and were stained using 2′,7′-dichlorofluorescein acetate (DCFDA) for 20 min at 37°C. After staining, cells were washed twice with PBS and viewed under a fluorescence microscope at a magnification of ×40.

Generally, NF-κB is known to be involved in the production of NO in macrophages. Therefore, we tested the involvement of NF-κB in SslE-mediated iNOS expression and NO production in RAW 264.7 cells. Our results show that preincubation of RAW 264.7 cells with MG-132 for 1 h before treatment with SslE attenuated the expression of iNOS (Fig. 4C) and concomitant NO production compared to those for the untreated control (Fig. 4D). We also tested the expression of iNOS and production of NO in the presence of inhibitors for p38, JNK1/2, and ERK1/2. Importantly, no changes could be found in iNOS expression (Fig. 4C) and NO production (Fig. 4D) in the presence of the inhibitors SB203580, SP600125, and U0126.

We further determined whether SslE-induced NF-κB mediated NO production is TLR2 dependent. We found a significant reduction in the amount of nitrite at 12 h poststimulation with SslE in TLR2 siRNA-transfected cells, whereas production of NO was unaltered in scrambled siRNA-transfected cells (Fig. 4E). Thus, our results suggest that TLR2 is essential in SslE-stimulated NO production with the involvement of NF-κB.

We further compared the levels of ROS, one of the critical proinflammatory hallmarks of macrophages, in SslE-treated and untreated RAW 264.7 cells by staining with 2′,7′-dichlorofluorescein diacetate (DCFDA), a nonpolar dye that easily diffuses into cells and is deacetylated by cellular esterases to a nonfluorescent derivative, DCFH. In the presence of ROS, DCFH is oxidized to highly fluorescent 2′,7′-dichlorofluorescein (DCF). SslE treatment of RAW 264.7 cells for 2 h was found to increase the cellular ROS level as detected by increased fluorescence of DCF compared to that for the untreated control (Fig. 4E). To further establish that this effect of SslE is TLR2 dependent, the level of ROS was measured in TLR2 siRNA-transfected cells, which showed decreased DCF staining compared to the untransfected RAW 264.7 cells and RAW 264.7 cells transfected with scrambled control siRNA (Fig. 4E), establishing that SslE induces TLR2-dependent production of ROS in mouse macrophages.

Expression of M1 or proinflammatory chemokines in response to SslE.

Production of chemokines is one of the crucial determinants in macrophage activation and polarization. Proinflammatory or M1 macrophages are conventionally associated with a set of specific chemokines known as M1 chemokines. We therefore looked for the chemokine profile induced by SslE. RAW 264.7 cells were found to overexpress the mRNAs for M1 chemokines such as MIP-1α, MIP-,1β and RANTES after 6 h of treatment with SslE compared to the untreated control (Fig. 5A). On the other hand, no effect of SslE on the mRNA expression of M2 chemokines such as MDC, eotaxin-2, and TARC could be found (Fig. 5A).

FIG 5.

SslE induces secretion of several M1 chemokines in RAW 264.7 cells. (A) Cells were incubated in absence or presence of SslE for 6 h. Total RNA was extracted, cDNA was prepared, and RT-PCR was done using specific primers for RANTES, MIP-1α, MIP-1β, eotaxin-2, MDC and TARC. GAPDH was used as a control. (B) Kinetics of SslE-induced secretion of M1 chemokines determined by ELISA.

The culture supernatants after treatment with SslE were collected at different time points to study the kinetics of chemokine secretion. The chemokines that showed overexpression in RT-PCR were selected to study the kinetics on induction with SslE. Among all the chemokines tested, RANTES showed rapid kinetics of stimulation on induction with SslE, being 3.9 ± 1.25-fold higher as early as 2 h poststimulation than the corresponding control level at 0 h (304 ± 23.4 versus 77 ± 18.7 pg/ml; P < 0.0001) (Fig. 5B). The level was found to be optimum at 6 h postinduction, being 9 ± 2-fold higher than the control level at 0 h (720 ± 33 versus 77 ± 18.7 pg/ml; P = 0.0001). The level of secreted RANTES decreased a little at 8 h and further decreased by 1.3-fold and 1.6-fold at 10 h and 24 h poststimulation, respectively, indicating a sustained release over the time period studied (Fig. 5B). The secretion of MIP-1α was significant at 4 h postinduction, being 5.6 ± 2-fold higher than the respective control level at 0 h (204 ± 27.6 versus 36.3 ± 13.6 pg/ml; P < 0.0001) and reached maximum at 8 h, being 15.3 ± 2.2-fold higher than the corresponding control level at 0 h (551 ± 30 versus 36 ± 13.6 pg/ml; P < 0.0001) (Fig. 5B). Like that of RANTES, the level of MIP-1α decreased slightly after maximum secretion and finally declined by only 1.7-fold at 24 h of exposure. MIP-1β showed kinetics similar to those of MIP-1α, being optimally induced at 8 h posttreatment, to a level 14.7 ± 3.4-fold higher than the control level at 0 h (500 ± 21 versus 34 ± 6.2 pg/ml; P < 0.0001), and thereafter it remained nearly consistent until 24 h of detection (Fig. 5B).

SslE induces upregulation of MHC II and costimulatory molecules CD80 and CD86 on macrophages.

Stimulating surface overexpression of MHC II and other costimulatory molecules on macrophages is one of the essential steps for an immunogen to commence successful immune response. We therefore investigated the relative upregulation in fluorescence intensity for CD80, CD86, and MHC II in SslE-treated RAW 264.7 cells compared to the untreated control. Analysis of relative fluorescence intensity showed that SslE induced optimum expression of MHC class II, 13 ± 2-fold higher than that in the untreated control (P < 0.0001) (Fig. 6A). Expression of CD80 was 5 ± 0.3-fold higher (P < 0.001) and that of CD86 was 2.4 ± 0.12-fold higher (P = 0.05) than those in untreated cells (Fig. 6A). The study of relative fluorescence intensity indicated that SslE induces overexpression of CD80, CD86, and MHC class II on RAW 264.7 cells.

FIG 6.

(A) SslE stimulates increased expression of MHC II and costimulatory molecules on RAW cells. RAW cells were treated (red) or not (gray) with SslE for 6 h, stained with FITC-conjugated anti-CD80, CD86, and MHC II antibodies for 20 min at room temperature in the dark, and analyzed by flow cytometry for the expression of these molecules. (B) Schematic representation of SslE-induced proinflammation and macrophage activation through TLR1/2 with the involvement of NF-κB and MAP kinase signaling pathways. SslE triggers activation of TLR1/2 heterodimers and subsequent recruitment of adaptors MyD88 and TIRAP, which in turn recruits and activates TRAF6 and other downstream molecules. This leads to IκB proteosomal degradation followed by nuclear translocation of p50 and p65 and phosphorylation of p38, ERK1/2, and JNK1/2. The activation of NF-κB and MAP kinase signaling pathways leads to the secretion of various proinflammatory cytokines such as TNF-α, IL-1α, and IL-1β, different M1 chemokines such as MIP-1α, MIP-1β, and RANTES, and also other inflammatory hallmarks such as cellular ROS and NO. SslE also induces surface expression of MHC II and costimulatory molecules CD80 and CD86, leading to activation and M1 polarization of mouse macrophages.

HEK293 cells used to study the receptor specificity of SslE.

We next tried to establish the receptor specificity of SslE in HEK293 cells. First we studied the overexpression of TLR2 on induction with SslE in HEK293-TLR2 cells. Immunoblotting showed increased expression of TLR2 in the presence of SslE in HEK-TLR2 cells (Fig. 7A), whereas no upregulation in the expression of TLR4 could be found in HEK-TLR4/MD2/CD14 (denoted HEK-TLR4) cells on stimulation with SslE (Fig. 7B). LPS was used as a positive control that increased the expression of TLR4 in HEK-TLR4 cells (Fig. 7B). Flow cytometry also showed 3.2 ± 1-fold (P < 0.001) and 9 ± 2.8-fold (P < 0.0001) elevated TLR2 expression in SslE-stimulated HEK-TLR2 cells compared to unstimulated HEK-TLR2 and HEK-vector cells, respectively (Fig. 7C). Immunofluorescence likewise showed strong anti-TLR2 fluorescence on the surface of HEK-TLR2 cells treated with SslE compared to the untreated HEK-TLR2 cells, which showed slightly less intense fluorescence, whereas HEK-vector cells showed no fluorescence (Fig. 7D). These results indicate that SslE induced overexpression of TLR2 in HEK-TLR2 cells. Next, a pulldown experiment was performed to test the interaction between SslE and TLR2. Western blotting showed that immobilized SslE pulled down TLR2 but not the vector control (Fig. 7E) and TLR4 (data not shown). This confirms that SslE specifically interacts with TLR2.

FIG 7.

SslE binds to TLR2 in HEK293 cells and induces NF-κB and MAP kinase activation with cytokine secretion in a TLR2-dependent manner in HEK293 cells. (A, C, and D) HEK293 cells were transiently transfected to generate HEK293-TLR2 and HEK293-TLR4 cells. After 24 h of transfection, cells were treated or not with SslE for 6 h to check overexpression of TLR2 in SslE-stimulated HEK293-TLR2 cells by immunoblotting (A), flow cytometry (C), and immunofluorescence (D). HEK293-vector cells were used as a control. (B) Expression of TLR4 was studied in HEK293-TLR4 cells in the presence or absence of SslE using LPS as a positive control. (E) Immunoprecipitation was done (after 2 h of treatment with SslE) with lysates from HEK-TLR2 and HEK-vector cells to check pulldown of TLR2. (F). Nuclear translocation of p65 was checked by immunofluorescence. (G) Cell lysates were analyzed by Western blotting to study the phosphorylation/activation of ERK1/2 and JNK1/2 (using anti-phospho-ERK1/2 [p-ERK1/2] and anti-phospho-JNK1/2 [p-JNK1/2], respectively) and degradation of IκB-α (using anti-IκB-α). Specific Abs for each unphosphorylated kinase and α-tubulin were used as loading controls. (H) The level of secreted IL-8 was measured by ELISA after 8 h of treatment with SslE.

SslE-induced TLR2-dependent signaling and cytokine secretion in HEK293 cells.

To investigate the functional status of TLR2 in HEK cells and to further validate the involvement of TLR2 and the subsequent signaling pathway in SslE-induced cytokine secretion, we next analyzed the activation of NF-κB and different MAP kinases in SslE-stimulated HEK-TLR2 cells. HEK-TLR4 and HEK-vector cells were used as controls. As mentioned above, IκB degradation is a measure of NF-κB activation, so we determined IκB degradation and found that at 3 h, IκB was almost completely degraded in SslE-induced HEK-TLR2 cells but not in SslE-stimulated HEK-vector or HEK-TLR4 cells, whereas in LPS-stimulated HEK-TLR4 cells, IκB was found to be totally degraded (Fig. 7G). In addition, immunofluorescence further showed that in SslE-stimulated HEK-vector and HEK-TLR4 cells, most of the p65 was localized in the cytosol, whereas a clear translocation of p65 to the nucleus could be seen in HEK-TLR2 cells (Fig. 7F). We also found phosphorylation of JNK1/2 and ERK1/2 in SslE-induced HEK-TLR2 cells and LPS-induced HEK-TLR4 cells (Fig. 7G). Thus, these results indicate that SslE-induced activation of NF-κB and MAP kinase is TLR2 dependent and that the stimulation is solely due to SslE and not due to any contaminating LPS in the SslE preparation.

Finally, SslE was found to trigger secretion of IL-8 in HEK-TLR2 cells. At 8 h poststimulation, secretion of IL-8 was 12 ± 4-fold and 19 ± 6-fold higher in HEK-TLR2 cells than in HEK-TLR4 cells (591 ± 31-fold versus 51 ± 8-fold; P < 0.0001) and HEK-vector cells (591 ± 31-fold versus 31 ± 5.5-fold; P < 0.0001), respectively (Fig. 7H).

DISCUSSION

Neonates depend primarily on the innate immune system to combat infection (20, 21), and TLRs, the family of innate immune signaling receptors, are critical in detecting invading pathogens, leading to proinflammation and subsequent activation of the host's innate immune responses (22, 23) not only in adults but also in neonates (35, 36). Among the 13 TLRs identified to date, the involvement of TLR2 in neonatal sepsis has been widely established. TLR2 was found to be involved in recognition of live Staphylococcus epidermis, a major cause of nosocomial bacteremia, in mouse peritoneal macrophages and whole blood and in subsequent clearance of S. epidermis bacteremia (37). Studies showing increased expression of TLR2 on granulocytes and monocytes of septicemic neonates compared to healthy neonates (38) and higher expression of TLR2 on neutrophils in saliva and blood in septicemic neonates (39) not only provide significant evidence that TLRs are essential in combating infection in neonates but also identify TLR2 as a likely candidate for a sepsis marker. In our present study, we demonstrated that rSslE from a NSEC isolate induces proinflammatory cytokine secretion in a TLR2-dependent manner as indicated by impaired cytokine secretion in the presence of a monoclonal antibody (MAb) against TLR2 and in TLR2 siRNA knockdown mouse macrophages. The ability of SslE to specifically pull down TLR2 but not TLR4 in mouse macrophages further supports the receptor specificity of SslE. Furthermore, the increased TLR2 expression in-SslE stimulated HEK293-TLR2 cells as evidenced by immunoblotting, immunofluorescence, and fluorescence-activated cell sorter (FACS) studies and the ability of SslE to interact only with HEK293-TLR2 cells and not with HEK293-TLR4 cells or HEK293-vector cells (as evidenced by immunoprecipitation) indicated that SslE is a potent agonist for TLR2. A common limitation of expressing recombinant His tag proteins in E. coli is LPS contamination, which we have discussed in our previous studies (19, 40). In the present study, we have taken various measures to prove that the involvement of TLR2 is solely due to SslE and not due to any contaminating LPS if present. First, all the treatments in mouse macrophages were done in the presence of PMB, which nullifies any effect of contaminating LPS if present. Second, the unaltered expression of TLR4 in mouse macrophages and also in HEK-TLR4 in either presence or absence of PMB upon SslE stimulation compared to the control not only indicates that SslE has no effect in TLR4 upregulation but also clearly indicates that our recombinant SslE preparation is LPS free, as LPS is well-known agonist for TLR4. LPS binds to LPS-binding protein (LBP), which delivers LPS to a cell surface receptor complex composed of CD14, MD-2, and TLR4 (41–44). Third, immunoblotting showing similar levels of TLR2 expression in mouse macrophages and HEK-TLR2 cells in either the absence or presence of PMB demonstrates that TLR2 upregulation is due to SslE and not to any contaminating LPS. This is in agreement with previous studies that clearly showed that TLR2 exclusively recognizes bacterial lipoproteins (45–47). Importantly, lipoproteins from group B Streptococcus (GBS), the most important Gram-positive causative agent of neonatal sepsis, also have been found to be crucial in activating TLR2, an essential signaling molecule identified in a mouse model of neonatal GBS sepsis (48). Thus, from the above discussion, it is clear that SslE is a potent agonist for TLR2. As it is known that TLR2 signaling depends on the coexpression of other TLRs, either TLR1 or TLR6 (49–52), we determined the contributions of TLR1 and TLR6 in SslE-induced signaling in RAW 264.7 cells by overexpression and immunoprecipitation studies. Our results show that SslE costimulates TLR1 with TLR2 as a combinatorial receptor to be recognized by macrophages. A number of studies have found the involvement of TLR2/TLR1 heterodimers in cellular responses of host innate immune components through PAMPs of different bacterial species (53, 54).

TLRs employ two kinds of signaling, a MyD88-dependent or a MyD88-independent signaling pathway. Most TLRs use MyD88, with the exception of TLR3. In MyD88-dependent signaling, upon ligand binding, conformational changes occur in the TIR domain of TLRs, leading to the association of MyD88 with TLR via homotypic interaction between their TIR domains. MyD88 further recruits IRAK-4 via interaction between their death domains. IRAK-4 employs and phosphorylates IRAK-1, which then recruits another adaptor, TRAF6, to the MyD88 complex, finally resulting in the phosphorylation and degradation of IκB and activation of the NF-κB signaling pathway (26, 47). Our results showed that incubating cell lysates of SslE-treated RAW cells with anti-SslE antibody pulled down the adaptors TIRAP, MyD88, and TRAF6.

Proinflammatory cytokine secretion can be induced by NF-κB and MAP kinase pathways individually or in combination, depending on the cell type and the stimulus. NF-κB, the primary transcription factor activated in TLR signaling functions as dimers formed by the interactions of two (p50 and p65) of the five Rel family proteins. The dimer is generally sequestered in cytosol by IκB-inhibitory proteins, which in the presence of a stimulus become phosphorylated and degraded, rendering dimeric NF-κB free to enter the nucleus. Hence, levels of p-IκB and IκB degradation are the measures of NF-κB activation (55). We found that SslE induces a time-dependent degradation of IκB in mouse macrophages and also augments the nuclear translocation of both p65 and p50, which became impaired in TLR2 siRNA knockdown cells. Degradation of IκB and nuclear translocation of p65 were also evident in SslE-induced HEK-TLR2 cells. Furthermore, SslE-induced macrophage release of IL-1α, IL-1β, and TNF-α was drastically attenuated in the presence of MG-132, an inhibitor of NF-κB nuclear translocation. All these observations suggest that SslE-induced cytokine secretion encompasses activation of NF-κB and is TLR2 dependent. This is consistent with several earlier studies which have shown the engagement of NF-κB in TLR-dependent secretion of IL-1α, IL-1β, and TNF-α (56–58).

Proinflammation also employs signaling through a family of Ser/Thr kinases, the MAP kinases such as ERKs, JNKs, and p38, which are activated sequentially through a triple phosphorylation cascade (59, 60). Our results demonstrate that p38, JNK1/2, and ERK1/2 were significantly phosphorylated in the presence of SslE, which was hindered in TLR2 knockdown macrophages, suggesting TLR2-dependent activation of the MAP kinase pathway. Moreover, we found cytokine-specific inhibition with different MAP kinase inhibitors. Secretion of IL-1α and IL-1β was obstructed by use of a JNK1/2 inhibitor (SP600125) and an ERK1/2 inhibitor (U0126), and secretion of TNF-α was inhibited by a p38 inhibitor (SB203580). The p38 signaling pathway was found to be critical in TLR2- and TLR4-mediated production of TNF-α in recombinant Brucella cell surface protein 31 (rBCSP31)-induced macrophages (58). In another study, p38 was shown to be necessary in TLR2-dependent Shigella flexneri outer membrane protein-induced cytokine release in mouse macrophages (56). ERK1/2 was found to be involved in TLR2- and TLR4-mediated production of IL-1β in mouse macrophages if they were treated with OM85-BV, an extract mixture derived from 8 strains of the most common bacteria involved in upper respiratory tract infection (57). Hence, it can be concluded that SslE-induced production of cytokines is regulated in a complex manner involving both NF-κB and MAP kinase signal transduction pathways working together. In addition, there exists a distinct signaling pathway for each cytokine induction, reflecting the functional differences among the cytokines.

Macrophages, the essential components of the innate immune system, are able to release various inflammatory cytokines, chemokines, and reactive oxygen and nitrogen species to act against invading pathogens, and hence these proinflammatory macrophages have strong microbicidal activity (25, 31). In our study, we found that RAW 264.7 cells on stimulation with SslE produce NO in a dose-dependent manner. NO, being an antibacterial effector, can inhibit pathogen replication (61) and is involved in inflammation and innate immunity of host immune cells (62, 63). In macrophages, NO is produced by an inducible NO synthase (iNOS or NOS2) in response to different environmental stimuli such as endotoxins or cytokines (64, 65). We found that SslE can upregulate iNOS gene expression which is impeded in TLR2 knockdown cells, indicating the involvement of TLR2 in SslE-induced iNOS expression and concomitant NO production. Numerous studies have shown the involvement of TLRs in iNOS expression and subsequent NO production in macrophages (66). In addition, we found that SslE can stimulate the production of ROS in a TLR2-dependent manner. ROS has been suggested to play vital roles not only in inflammation and macrophage activation but also in the functional regulation of macrophages, including phagocytosis activity, motility, and proliferation of macrophages (32).

In addition to proinflammatory cytokines, NO, and ROS, SslE was also found to upregulate the expression of various proinflammatory chemokines, such as MIP-1α, MIP-1β, and RANTES, which further fortifies its ability to induce proinflammation and macrophage activation. Being secreted from macrophages, these chemokines can attract and activate multiple immune cells, including leukocytes and Th1 cells, to the sites of infection and also perform other functions such as T-cell differentiation (30).

Hence, the ability of SslE to induce the secretion of different proinflammatory cytokines, chemokines, ROS, and NO gave us a hint that SslE may activate and polarize macrophages toward the proinflammatory or M1 phenotype. Furthermore, the increased surface expression levels of MHC II and costimulatory molecules such as CD80 and CD86 validate our hypothesis that SslE promotes macrophage activation. The increased expression of MHC II also implies that SslE may also encourage adaptive immune responses by presenting peptides with MHC II to T lymphocytes. Further studies are needed to establish the role of SslE in T-cell priming and enhancing the Th1 immune response.

In conclusion, the present study identifies the cell surface receptor and the signaling pathway involved in SslE-induced proinflammation in mouse macrophages. Our study shows that SslE signals through TLR2/TLR1 heterodimers with the subsequent involvement of MyD88 to finally trigger the activation of both NF-κB and MAP kinase signaling pathways. Furthermore, our study is the first to reveal that SslE encourages the production of reactive nitrogen and oxygen species, proinflammatory chemokines, and MHC II and other costimulatory molecules (CD80 and CD86) in mouse macrophages. All these results collectively indicate that SslE of NSEC plays a crucial role in activating macrophages and initiating innate immune responses of the host. Hence, the use of SslE in a vaccine against E. coli sepsis should draw serious attention, as SslE has been identified as a broadly conserved immunogen among ExPEC strains that confers almost complete protection against E. coli sepsis in a mouse model, in corroboration with our finding that it can trigger proinflammation and macrophage activation involving the innate immune receptor TLR2.

MATERIALS AND METHODS

Cell culture and treatment.

Mouse macrophages (RAW 264.7 cells) were grown and maintained in RPMI 1640 medium (GIBCO, Rockville, MD) with 10% fetal bovine serum (FBS) (GIBCO) in the presence of 5% CO₂ at 37°C.

For treatment, cells were starved overnight in serum-free medium and incubated with polymyxin B (PMB) at a concentration of 50 μg/ml in serum-free medium at 37°C with 5% CO₂. After 1 h of incubation, purified rSslE was added to the cells for specific time intervals as required for each experiment.

Purification of rSslE.

Recombinant SslE (rSslE) was purified as described in our earlier study (19). Briefly, recombinant clone pTSslE7 was induced with 0.02% l-arabinose for 5 h at 37°C for optimum expression of SslE, which was purified by successive Ni-nitrilotriacetic acid (NTA) and gel filtration (Sephadex G-200) chromatography. Protease activity was confirmed using the pNA oligopeptide substrate assay as described in our earlier study (19).

Raising of antisera.

Antiserum against SslE was raised by immunizing 3- to 4-week-old Swiss albino mice with a purified fraction of SslE eluted from a Sephadex G-200 column. Each mouse was injected subcutaneously with 25 μg of purified protein. After four successive injections, each at a 7-day interval, mice were sacrificed to collect the antiserum, which was stored at −70°C until use.

Preparation of cell lysate and immunoblotting.

The experimental RAW 264.7 cells were washed twice with ice-cold phosphate-buffered saline (PBS)and lysed in lysis buffer containing 50 mM HEPES (pH 7.4), 100 mM NaCl, 2 mM EDTA, 0.5% Nonidet P-40 (NP-40), 10% glycerol, 50 mM β-glycerophosphate, 1 mM sodium fluoride, and 1× protease inhibitor cocktail on ice. All the reagents were purchased from Sigma Chemical Company, St. Louis, MO. The lysate were collected by centrifuging the cells in lysis buffer at a relative centrifugal force (RCF) of 20,817 for 10 min. The proteins in the lysates were estimated using the Bradford assay, and approximately 50 to 60 μg protein was loaded for 12% SDS-PAGE. The separated proteins were transferred to polyvinylidene difluoride (PVDF) membrane (Millipore, USA), blocked with 3% bovine serum albumin (BSA) (Sigma) in Tris-buffered saline–Tween 20 (TBST), and incubated with specific primary antibodies (1:2,000) overnight at 4°C. The membranes were washed and reincubated with horseradish peroxidase (HRP)-conjugated secondary antibody (1:5,000) (Thermo Scientific Pierce, Rockford, IL) for 1 h at room temperature and developed using West Pico Plus chemiluminescence substrate (Thermo Scientific Pierce).

Immunoprecipitation.

RAW 264.7 cells were treated with 150 nM SslE for 2 h for immunoprecipitation. Both the untreated and treated cells were lysed using lysis buffer as described above. A total of 500 μg of proteins from each cell lysate was incubated with 2 μg of anti-SslE antibody raised in mouse overnight at 4°C in an end-to-end rocker with gentle rocking. Immune complexes were precipitated the next day with 80 μl of protein A/G Plus agarose beads (Santa Cruz Biotechnology, USA) for 4 h at 4°C with gentle rocking, followed by centrifugation at an RCF of 3,824 for 2 min. Beads were washed twice with NP-40 lysis buffer, resuspended in 70 μl of 2× sample loading buffer, and boiled for 10 min prior to SDS-PAGE and subsequent immunoblotting.

Nuclear and cytosolic fractionation.

After treatment for 3 h with 150 nM SslE, nuclear and cytosolic fractions from both treated and untreated RAW 264.7 cells were prepared. Briefly, RAW 264.7 cells were scraped in PBS and collected by centrifugation. The pellet was resuspended in low-salt buffer (10 mM HEPES [pH 7.9], 1.5 mM MgCl₂, 10 mM KCl, 0.2 mM phenylmethylsulfonyl fluoride [PMSF], 0.5 mM dithiothreitol [DTT], 0.2 mM EDTA, 1× protease inhibitor cocktail) and incubated on ice for 15 min. Following incubation, NP-40 was added at a final concentration of 0.5%, and the cell suspensions were passed through a 21-gauge needle (BD Diagnostics, Franklin Lakes, NJ, USA) 6 or 7 times. The cell suspensions were spun at an RCF of 20,817 for 10 min at 4°C to separate the nuclear and cytosolic fractions. For the extractions of nuclear proteins, the cell pellet was again resuspended in high-salt buffer (20 mM HEPES [pH 7.9], 0.4 M NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 25% glycerol, 0.5 mM PMSF, 0.5 mM DTT) and incubated on ice for 30 min with vigorous shaking (200 rpm), followed by centrifugation at an RCF of 20,817 for 10 min at 4°C.

Immunofluorescence for detection of NF-κβ nuclear translocation.

Cells were seeded on coverslips for treatment with SslE. Experimental RAW 264.7 cells were washed with ice-cold PBS twice and fixed using 4% paraformaldehyde (Sigma) at 4°C for 20 min. Cells were washed again and permeabilized with 0.1% Triton X-100 (Sigma) in 0.1% sodium citrate (Sigma) solution at 4°C for 20 mins. Cells were then blocked using 5% fetal bovine serum in PBS–Tween 20 (PBST for) 1 h at room temperature and incubated with anti-p50 (Santa Cruz Biotechnology) and anti-p65 (Santa Cruz Biotechnology) primary antibodies (1:200) overnight in a moist chamber at 4°C. After washing, cells were incubated with tetramethyl rhodamine isocyanate (TRITC)-conjugated secondary antibody (1:500) (Thermo Scientific Pierce) for 1 h at room temperature. Nuclear staining was done with DAPI (4′,6′-diamidino-2-phenylindole) (HiMedia, India) for 10 min at room temperature in a moist chamber. The coverslips were mounted on glass slides, and the glass slides were viewed under a confocal microscope (model LSM 510 Meta; Zeiss, Germany) and excited at 543 nm for TRITC. The emission filters used were 558 nm to 601 nm for TRITC.

Blocking test using anti-TLR2 antibody and NF-κB and MAP kinase inhibitors.

To establish the involvement of TLR2 in cytokine, chemokine, and NO production, RAW 264.7 cells were incubated with or without 10 μg/ml monoclonal anti-TLR2 antibody (Abcam, UK) in a 24-well plate for 1 h at 37°C. After incubation, SslE at a concentration of 150 nM was added to the cells. Culture supernatant was collected after desired time intervals to detect cytokine, chemokine, and NO production.

To evaluate the roles of the ERK1/2, JNK1/2, p38, and NF-κB signaling pathways in SslE-mediated production of IL-1α, IL-1β, TNF-α, and NO, specific inhibitors against each pathway were used. RAW 264.7 cells were incubated with 10 μM U0126 for ERK1/2, 10 μM SP600125 for JNK1/2, 10 μM SB203580 for p38, and 2.5 μM MG-132 for NF-κB for 1 h at 37°C in a 24-well plate before treatment with SslE. After 8 h of incubation with SslE, culture supernatant was collected to determine the production of different cytokines and NO.

siRNA knockdown.

The expression of TLR2 was blocked by transfection with siRNA (Santa Cruz Biotechnology) using jetPRIME siRNA transfection reagent (Polyplus) following the manufacturer's protocol. A scrambled siRNA (Santa Cruz Biotechnology) was also used as a control. RAW 264.7 cells were transiently transfected with TLR2 siRNA and scrambled siRNA for 24 h and were then used for subsequent assays.

Cytokine ELISA.

Quantitation of different proinflammatory cytokines, such as IL-1α, IL-1β, and TNF-α, in the culture supernatant of experimental RAW 264.7 cells was done by sandwich ELISA using specific ELISA kits (R&D Systems, Minneapolis, MN) following the manufacturer's instructions. The standard curve for each cytokine was plotted using the standards provided in each kit. The concentrations of the unknowns were detected from the standard curves and expressed in pg/ml of cell supernatant.

Nitrite analysis by Griess assay.

The amount of NO produced by macrophages was detected by measuring the accumulation of its stable end product nitrite by the Griess reaction. Briefly, 50 μl of culture supernatant was mixed with 25 μl of sulfanilamide in 0.5 N HCl in a 96-well plate and incubated at room temperature for 5 min in dark, and then 25 μl of 0.02% naphthylethylenediamine dihydrochloride (NEDD) (Sigma) in 0.5N HCl was added and absorbance was measured at 540 nm using a microplate reader (Bio-Rad). Unknown concentrations of nitrite were calculated based on the standard curve generated with sodium nitrite (Sigma).

RT-PCR analysis for chemokines and iNOS.

Total RNA from experimental RAW 264.7 cells was isolated using TRIzol reagent (Invitrogen). One microgram of RNA from either SslE-treated or untreated RAW 264.7 cells was used for cDNA synthesis using the Revert Aid reverse transcriptase kit (Thermo Scientific). cDNAs were amplified using specific primers for each chemokine and iNOS (Table 1) by PCR in an automated thermal cycler (Bio-Rad, USA) using Green-Taq PCR master mix (Thermo Fisher, USA). GAPDH (glyceraldehyde-3-phosphate dehydrogenase) was amplified as a control. PCR conditions were as follows: for chemokines, initial denaturation at 95°C for 5 min, 30 cycles of denaturation at 95°C for 30 s, annealing at the indicated temperatures (Table 1) for 30 s, and extension at 72°C for 1 min, and a final extension at 72°C for 7 min; for iNOS, initial denaturation at 95°C for 5 min, 30 cycles of denaturation at 95°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 1 min, and a final extension at 72°C for 7 min; and for GAPDH, initial denaturation at 95°C for 5 min, 30 cycles of denaturation at 95°C for 30 s, annealing at 50°C for 30 s, and extension at 72°C for 30 s, and a final extension for 7 min at 72°C.

TABLE 1.

Primers used

Name	Sequence (5′→3′)	Concn (μM)	Annealing temp (°C)	Product size (bp)	Reference
MIP-1α	5′CCTTGCTGTTCTTCTCTGTACC3′	0.5	66	559	56
	5′ACAGTGTGACCAACTGGGAGG3′
MIP-1β	5′GTTCTCAGCACCAATGGGCTCT3′	0.5	68	429	56
	5′TCTCCATGGGAGACACGCGTC3′
RANTES	5′TGCCTCACCATATGGCTCGGA3′	0.5	66	219	56
	5′CCTCTATCCTAGCTCATCTCCA3′
TARC	5′CATGAAGACCTTCACCTCAGC3′	0.5	64	256	56
	5′GTCTGCACAGATGAGCTTGC3′
MDC	5′CACCTGACGAGGACACATAAC3′	0.5	64	266	56
	5′TGCCTGGGATCGGCACAGAT3′
Eotaxin-2	5′GCTTTGAACTCTGAGCTGTGC3′	0.5	64	374	56
	5′ACGGCGTCTCTGGACAGCAA3′
iNOS	5′CAGCCCAACAATACAAGATGACCC3′	0.4	55	525	66
	5′CAGTTCCGAGCGTCAAAGACCTGC3′
GAPDH	5′GACCCCACTAACATCAAAT3′	0.8	50	189	This study
	5′TGAGTTGTCATATTTCTCGT3′

Chemokine ELISA.

The chemokines that showed overexpression in SslE-treated RAW 264.7 cells, such as MIP-1α, MIP-1β, and RANTES, by RT-PCR analysis were selected to study the kinetics by ELISA. RAW 264.7 cells were treated with 150 nM SslE, and cell supernatants were collected after different time intervals (0, 2, 4, 6, 8, 10, 12, and 24 h) to study the kinetics of the secreted chemokines using specific ELISA kits (R&D Systems, Minneapolis, MN) the following manufacturer's protocol.

Detection of cellular ROS by DCFDA staining.

The in situ ROS level was measured by 2′,7′-dichlorofluorescein diacetate (DCFDA) staining. DCFDA, a fluorogenic nonpolar dye, becomes deacetylated after its diffusion into cells by cellular esterases to a nonfluorescent compound. This compound is oxidized by cellular ROS to a highly fluorescent compound, 2′,7′-dichlorofluorescein (DCF). SslE-treated (150 nM for 2 h) and untreated RAW cells were incubated with DCFDA for 20 min at 37°C, washed with PBS twice, and viewed under a fluorescence microscope (Olympus-IX71) at a magnification of ×40.

Flow cytometry.

SslE-treated and untreated cells were washed with 1× PBS and stained with phycoerythrin (PE)-conjugated anti-mouse CD11b MAb (BD Biosciences Pharmingen) for 20 min at room temperature in the dark, followed by fluorescein isothiocyanate (FITC)-conjugated anti-mouse CD80 or CD86 (BD Biosciences Pharmingen) or MHC II (eBioscience). To detect surface expression of TLR2, cells were stained with PE-conjugated anti-mouse TLR2 (eBioscience) after treatment with SslE for 6 h. Stained cells were analyzed on a FACSCalibur instrument using CellQuest software.

Transient transfection of HEK293 cells.

Transient transfection was carried out using jetPRIME DNA transfection reagent (Polyplus) following the manufacturer's protocol. HEK293 cells were cotransfected with the TLR4 construct along with pDUO2-hMD2/CD14 (InvivoGen, France) to generate transiently transfected HEK293-TLR4/CD14/MD2 cells needed for activation of TLR4 signaling. pcDNA3-TLR2-CFP (Addgene plasmid 13015) and pcDNA3-TLR4-YFP (Addgene 13018) were gifts from Doug Golenbock.

Statistical analysis.

All experiments were performed in triplicate, and results obtained from cytokine ELISA, chemokine ELISA, Griess assay, and FACS were expressed as mean ± standard deviation (SD). Statistical comparison between two groups was done by Student's t test using GraphPad Prism software, version 5.0, and statistical significance for all tests was indicated as follows: *, P < 0.05; **, P < 0.01; and ***, P < 0.001.