OUP user menu

Virulence factor p60 of Listeria monocytogenes modulates innate immunity by inducing tumor necrosis factor α

Hiroshi Sashinami, Dong-Liang Hu, Sheng-Jun Li, Toshihito Mitsui, Ken-Ichi Hakamada, Yoh Ishiguro, Shinsaku Fukuda, Akio Nakane
DOI: http://dx.doi.org/10.1111/j.1574-695X.2010.00666.x 100-107 First published online: 1 June 2010

Abstract

We investigated the effect of p60, a virulence factor of Listeria monocytogenes, on host immune response in vitro and in vivo. Administration of p60 before a sublethal infection with L. monocytogenes enhanced innate host resistance in naïve mice. Mouse macrophage RAW264.7 cells produced tumor necrosis factor (TNF)-α in response to stimulation with recombinant p60. Toll-like receptor 4 may be involved in TNF-α production from RAW264.7 cells and enhanced host resistance induced by p60 administration. Our findings demonstrated that p60 modulates innate immune responses against L. monocytogenes infection.

Keywords
  • Listeria monocytogenes
  • p60
  • TLR4
  • TNF-α

Introduction

Listeria monocytogenes is a gram-positive, intracellular-growing bacterium that causes systemic infection in immunocompromised hosts, such as newborns, the elderly and pregnant women. Listeria monocytogenes is transmitted by intake of contaminated food products, and listeriosis is manifested as gastroenteritis, meningitis, encephalitis, vertical transmission and sepsis. To establish pathogenicity to its host, L. monocytogenes attaches, invades and replicates in various types of host cells. A number of listerial virulence factors are involved in the intracellular life of L. monocytogenes (Cossart & Lecuit, 1998). p60 is a 60-kDa extracellular protein produced by L. monocytogenes. p60 acts as a murein hydrolase required in the last step of cell division. p60 is mainly produced as a secretory form; in the stationary growth phase, a large amount of p60 accumulates in the culture supernatant (Wuenscher et al., 1993). An L. monocytogenes mutant that impaired synthesis of p60 led to abnormal cell division and formed short filaments during the logarithmic phase of growth (Pilgrim et al., 2003). This p60-deletion mutant also showed attenuated virulence in mice and lost the ability to invade fibroblasts (Pilgrim et al., 2003). These results suggested that p60 is involved in cell viability of L. monocytogenes and also contributes to the pathogenicity of L. monocytogenes in vivo.

Host cells express various pattern recognition receptors for sensing and reacting to diverse microbial pathogens. One such pattern recognition receptor is the Toll-like receptor (TLR) family, which was identified as a homolog of Toll receptor in Drosophila. The TLR family senses various types of pathogen-associated molecules, including lipids, peptidoglycans, lipoproteins, flagellins and nucleotides. Listerial peptidoglycan is recognized by TLR2 (Takeda et al., 2003) and flagellin is recognized by TLR5 (Hayashi et al., 2001). Recent research has shown that listeriolysin O is a ligand for TLR4 (Park et al., 2004). Recognition of pathogen-associated molecular patterns by TLRs stimulates cytokine production through signal transduction pathways that are dependent or independent of myeloid differentiation factor 88 (MyD88), an adaptor molecule involved in signal transduction of all TLRs except TLR3 (Akira et al., 2001). In the present study, we demonstrate that p60 mediates host innate immune responses by activation of nuclear factor κB (NF-κB) and induces proinflammatory cytokine production.

Materials and methods

Animals

Mice (C57BL/6, TLR4-lacking C3H/HeJ and C3H/HeN as the control strain of C3H/HeJ strain) were purchased from CLEA Japan Inc. (Tokyo, Japan). Mice were used at 6–8 weeks old. Animals were cared for under specific pathogen-free conditions in the Institute for Animal Experimentation, Hirosaki University Graduate School of Medicine. All animal experiments were conducted in accordance with the Animal Research Ethics Committee, Hirosaki University Graduate School of Medicine and followed the Guidelines for Animal Experimentation of Hirosaki University.

Bacterial infection

Listeria monocytogenes 1b 1684 (Sashinami et al., 2005) and Salmonella enterica serovar Typhimurium χ3306 (Yamamoto et al., 2001) cells were prepared as described previously. The concentration of washed cells was adjusted spectrophotometrically at 550 nm and cells were maintained at −80 °C until use. Six mice per group were infected with L. monocytogenes intravenously or infected with S. Typhimurium intraperitoneally at the indicated doses, and spleens and livers were obtained at various time points. Organs were homogenized in phosphate-buffered saline (PBS). The numbers of viable L. monocytogenes and S. Typhimurium in the organs of infected animals were counted by plating serial 10-fold dilutions of organ homogenates on trypticase soy agar (BD Diagnosis Systems, Sparks, MD) and Luria–Bertani agar (Invitrogen, Carlsbad, CA), respectively. Colonies were routinely counted 24 h later. For C57BL/6 mice, the 50% lethal dose of intravenously infecting L. monocytogenes was 5 × 105 CFU, whereas that of intraperitoneally infecting S. Typhimurium was <50 CFU (Sashinami et al., 2003).

Cells

Mouse macrophage cell line RAW264.7 was purchased from Dainippon Sumitomo Pharmaceutical Co. Ltd (Osaka, Japan). Cells were cultured in Dulbecco's modified Eagle medium (Nissui Pharmaceutical Co., Tokyo, Japan) supplemented with 10% fetal calf serum (FCS; JRH Biosciences, Lenexa, KS), and 3%l-glutamine (Wako Pure Chemical Industries, Osaka, Japan). Bone marrow-derived macrophages (BMDMs) were prepared as follows: femoral bones from mice were obtained aseptically, then both ends of the bones were cut off and bone marrow was flushed out with RPMI 1640 medium (Nissui Pharmaceutical Co.). Bone marrow cells were washed three times and cultured in RPMI 1640 containing 10% FCS supplemented with 5 µg mL−1 of macrophage colony-stimulating factor (Sigma-Aldrich Japan, Tokyo, Japan) for 3 days. Adherent cells were then washed and cultured for 7 days in fresh medium. Cells were detached with 5 mM EDTA in PBS, and used for further assays.

Construction of recombinant protein

Genomic DNA was isolated from L. monocytogenes by standard procedures (Omoe et al., 2002). For construction of recombinant p60 expression plasmid, PCR primers were designed to amplify a full length of the iap gene encoding p60 (forward: CCCGAATTCATGAAAAAAGCAACTATCGCGGC, reverse: CCCCAGCTGTTATACGCGACCGAAGCCAAC; restriction enzyme recognition sites underlined). The iap gene was amplified by PCR using Pyrobest DNA polymerase (Takara, Shiga, Japan). The DNA fragment was digested with EcoRI and SalI and subcloned into an expression vector, pGEX-6p-1 (GE Healthcare Bio-Sciences, Tokyo, Japan), to express glutathione S-transferase (GST) fusion protein. Protein expression, purification and removal of GST were performed as described previously (Omoe et al., 2002).

Preparation of anti-p60 antibody

To generate anti-p60 antibody, rabbits were hyperimmunized with recombinant p60 as described previously (Omoe et al., 2002). Sera were pooled and the immunoglobulin fraction was obtained by precipitation with a saturated ammonium sulfate solution (76.7%), followed by extensive dialysis against a PBS column. The concentration of immunoglobulin was determined using the Bradford assay.

Monitoring of contamination with endotoxin

To confirm that recombinant p60 and anti-p60 antibody were not contaminated with endotoxin, a Limulus amoebocyte lysate assay was carried out. Recombinant p60 protein contained endotoxin at <3 pg per 100 µg of p60 and anti-p60 antibody contained endotoxin at <30 pg per 5 mg of antibody. For heat inactivation, recombinant protein was boiled at 100 °C for 30 min. To inhibit the effect of lipopolysaccharide on cytokine induction, culture medium supplemented with 50 µg mL−1 of polymyxin B (PMB; Sigma-Aldrich Japan) was used. Lipopolysaccharide from Escherichia coli (Sigma-Aldrich Japan) was used as a control.

Determination of cytokine production

Production of tumor necrosis factor (TNF)-α was determined by enzyme-linked immunosorbent assay (ELISA). Determination of TNF-α was carried out as described previously (Sashinami et al., 2006).

Electrophoretic mobility shift assay (EMSA)

Nuclear extracts from samples were prepared using NE-PER™ nuclear and cytoplasmic extraction reagent (Pierce, Rockford, IL) and EMSA was carried out using a gel shift assay system (Promega Co., Madison, WI). Briefly, consensus NF-κB oligonucleotide was phosphorylated with [γ-32P]ATP (PerkinElmer, Boston, MA) by T4 polynucleotide kinase and nuclear extract was reacted with labeled oligonucleotide for 20 min. After electrophoresis of the complexes, the gel was dried and exposed to X-ray film at −70 °C overnight. The expected band for NF-κB was confirmed based on a shift of bands by addition of anti-NF-κB antibody binding.

Gene silencing

Expression of TLR4 and MyD88 in RAW264.7 cells was suppressed using small interference RNA (siRNA). siRNAs were purchased from Qiagen (Tokyo, Japan) and gene silencing was carried out using HiPerFect reagent (Qiagen) according to the manufacturer's instructions. For confirmation of the effect of siRNA, expression of TLR4 and MyD88 was assessed by reverse transcriptase-PCR.

Statistical analysis

Data are expressed as means±SD, and Student's t-test was used to determine the significance of differences in bacterial counts of the specimens in the control and experimental groups. Each experiment was repeated at least twice.

Results

Administration of recombinant p60 enhances host resistance against a primary infection with L. monocytogenes

We first assessed the effect of administration of p60 on innate immunity. Recombinant p60 was administered intravenously 6 h before infection, and the numbers of viable L. monocytogenes were determined at various time points. Mice injected with p60 showed lower bacterial numbers in the spleen (Fig. 1a) and liver (Fig. 1b) compared with PBS-injected controls, indicating that administration of p60 enhanced host resistance against L. monocytogenes infection in the early phase. We investigated whether the effect of p60 on host resistance against bacterial infection was specific to L. monocytogenes. Administration with p60 suppressed the growth of S. Typhimurium in the organs of infected mice (Fig. 1c and d). We then assessed whether antibody against p60 could neutralize the effect of p60 in vivo (Fig. 1e). Anti-p60 antibody was administered 24 h before p60 administration. Listeria monocytogenes infection was carried out 6 h after p60 injection and bacterial numbers in the organs of mice on day 3 after infection were determined. The enhanced bacterial elimination from the organs was inhibited by anti-p60 antibody (Fig. 1e).

Figure 1

Treatment with recombinant p60 enhances host resistance against bacterial infection. Mice were administered with 100 µg recombinant p60 (square) or PBS as a control (circle) intravenously 6 h before infection. (a, b) Mice were infected with 5 × 105 CFU Listeria monocytogenes intravenously. The numbers of viable L. monocytogenes in the spleens (a) and livers (b) at various time points were determined. (c, d) Mice were infected with 100 CFU of Salmonella enterica serovar Typhimurium intraperitoneally 6 h after administration with p60. The numbers of viable S. Typhimurium in the spleens (c) and livers (d) at various time points were determined. (e) Anti-p60 antibody canceled the effect of p60 on host resistance against L. monocytogenes infection in vivo. Mice were administered with 1 mg rabbit anti-p60 antibody intravenously, and then with 100 µg recombinant p60 or PBS as a control intravenously 6 h before infection. Mice were infected with 5 × 105 CFU L. monocytogenes intravenously. The numbers of viable bacteria in the spleens (filled) and livers (open) were determined on day 3 after infection. *Significantly different from the PBS-treated control (P<0.01).

Recombinant p60 induces production of proinflammatory cytokines in mouse macrophages

We assessed the effect of recombinant p60 on macrophage responses in vitro. Mouse macrophage RAW264.7 cells were stimulated with recombinant p60 and titers of TNF-α in the culture supernatants were determined. Significant production of TNF-α was induced by treatment with recombinant p60 (Fig. 2a). Stimulation with p60 also induced NF-κB activation (Fig. 2b) These results showed that recombinant p60 induced TNF-α production in mouse macrophages depending on the NF-κB pathway.

Figure 2

Effect of recombinant p60 on cytokine production and signal transduction of mouse macrophages. (a) RAW264.7 cells were prepared to 1 × 106 mL−1 in 24-well plates and stimulated with various doses of recombinant p60 for 24 h. Cells and culture supernatants were the collected and titers of TNF-α in culture supernatants were determined by ELISA. ND, below the level of detection. (b) NF-κB activation of nuclear protein extract from p60-stimulated cells. RAW264.7 cells were prepared to 1 × 106 mL−1 in 24-well plates and stimulated with 100 µg per well of recombinant p60 for 12 h. The stimulated cells were lysed and protein was extracted. NF-κB activation was detected by EMSA.

Cytokine induction by p60 is not due to lipopolysaccharide contamination

Given the potential problem of lipopolysaccharide contamination, we performed experiments to rule out the possibility of such contamination. Results showed that with heat inactivation recombinant p60 lost the ability to induce TNF-α (Fig. 3a). PMB treatment did not reduce the titer of TNF-α induced by p60 treatment (Fig. 3b). The Limulus amoebocyte lysate assay showed that 100 µg of recombinant p60 contained <3 pg of lipopolysaccharide. Additionally, 3 pg of lipopolysaccharide did not induce TNF-α production from RAW264.7 cells (Fig. 3b). These results strongly suggest that cytokine production and signal transduction by recombinant p60 are not due to lipopolysaccharide contamination.

Figure 3

Effect of heat-inactivated p60 and treatment with PMB on cytokine induction. (a) RAW264.7 cells were prepared to 1 × 106 mL−1 in 24-well plates and stimulated with 100 µg mL−1 recombinant p60 (filled) or heat-inactivated recombinant p60 (open) for 24 h. Culture supernatants were then collected and titers of TNF-α were determined by ELISA. (b) RAW264.7 cells were stimulated with recombinant p60 or lipopolysaccharide (LPS) in culture medium with or without 50 µg mL−1 PMB for 24 h. Culture supernatants were then collected and titers of TNF-α were determined by ELISA. ND, below the level of detection.

TLR4 is involved in p60-induced cytokine production by mouse macrophages

We next investigated whether the TLR family is involved in p60-activated signal transduction in mouse macrophages. Gene silencing of TLR2, TLR3, TLR4 and TLR7 with siRNA successfully suppressed production of TNF-α induced by each TLR agonist from RAW264.7 cells (data not shown). Therefore, we investigated the effect of gene silencing of TLRs on p60-induced cytokine production in mouse macrophages. TNF-α production in p60-stimulated cells was suppressed when the expression of TLR4 was silenced (Fig. 4a). Knockdown of TLR2, TLR3 or TLR7 produced no change in TNF-α production (data not shown). We then investigated whether the TLR4-dependent signal transduction pathway is involved in p60-stimulated cytokine production. Suppression of MyD88 expression inhibited TNF-α production in p60-stimulated macrophages (Fig. 4b), suggesting that TLR4 is involved in cytokine production in mouse macrophages by stimulation with p60. To confirm that the inducible potential of recombinant p60 for TNF-α production was TLR4 dependent, we investigated cytokine production by stimulation with recombinant p60 in BMDMs derived from TLR4-deficient C3H/HeJ mice. TNF-α was produced from BMDMs derived from C3H/HeN as the control strain, whereas BMDMs derived from C3H/HeJ showed significantly diminished TNF-α production (Fig. 4c). This result suggested that cytokine production by macrophages was TLR4 dependent. To confirm that the effect of p60 administration was dependent on TLR4, we assessed the effect of recombinant p60 on host resistance against L. monocytogenes infection. Bacterial growth in the spleen and liver of mice administered with recombinant p60 was suppressed comparing with PBS-treated control mice at 72 h after infection (Fig. 4d). However, the numbers of L. monocytogenes in the organs of TLR4-deficient C3H/HeJ mice showed no significant difference between the p60- and PBS-treated groups (Fig. 4e). These results suggested that recombinant p60 enhanced host innate resistance against infection with L. monocytogenes, depending on TLR4.

Figure 4

p60-induced host resistance against bacterial infection is TLR4 dependent. (a, b) Effect of gene silencing on cytokine production from mouse macrophages stimulated with p60. RAW264.7 cells were prepared to 1 × 106 mL−1 in 24-well plates and expression of TLR4 (a) or MyD88 (b) was suppressed by siRNA treatment. Cells were then stimulated with 100 µg per well of recombinant p60 for 24 h. The culture supernatants were collected and the titers of TNF-α in culture supernatants from the PBS-treated (filled) and p60-treated (open) cells were determined by ELISA. Results were obtained from three experiments with nine wells in each group. (c) Production of TNF-α in BMDMs of TLR4-deficient mice by recombinant p60. BMDMs from C3H/HeN mice and C3H/HeJ mice were prepared to 1 × 105 per 100 µL in 96-well plates and stimulated with 10 µg per well of recombinant p60 for 48 h. Culture supernatants were then collected and the titers of TNF-α produced in the PBS-treated (filled) and p60-treated (open) cells were determined by ELISA. (d, e) Host resistance against Listeria monocytogenes infection by pretreatment of p60 in C3H/HeJ mice. Six C3H/HeN (d) and C3H/HeJ (e) mice were injected with 100 µg of recombinant p60 intravenously and 5 × 105 CFU of L. monocytogenes was injected at 6 h after p60 administration. Spleens and livers from infected mice were obtained at 72 h after infection, and bacterial numbers in the organs from PBS-treated control (filled) and p60-treated (open) mice were determined. Asterisks indicate that the value is significantly different between the two groups: *P<0.01, **P<0.05.

Discussion

The innate immune response is essential for host resistance against infection with L. monocytogenes (Unanue, 1997). Interferon (IFN)-γ induced by natural killer cells and the resultant activation of resident macrophages are important for innate resistance against L. monocytogenes infection (Tripp et al., 1993). Previous studies using cytokine-deficient mice revealed that both IFN-γ and TNF-α are essential for host resistance against L. monocytogenes (Huang et al., 1993; Rothe et al., 1993). Bacterial components of L. monocytogenes reportedly induce cytokine production that contributes to activation of the innate immune response. For example, heat-killed L. monocytogenes induces IFN-α and IFN-γ production from human peripheral blood mononuclear cells (Nakane & Minagawa, 1981). Our results showed that pretreatment with p60 enhanced host resistance against L. monocytogenes infection (Fig. 1a and b) and that anti-p60 antibody canceled the effect of p60 (Fig. 1e). In addition, administration with p60 enhanced host resistance against S. Typhimurium (Fig. 1c and d). Production of TNF-α, a critical factor for host resistance against infection with L. monocytogenes (Rothe et al., 1993) and S. Typhimurium (Nauciel & Espinasse-Maes, 1992; Gulig et al., 1997), was induced by recombinant p60 (Fig. 2a). TLR4 is essential for host resistance against S. Typhimurium infection (Bernheiden et al., 2001). The present results showed that p60 induced proinflammatory cytokines by NF-κB activation from mouse macrophages in vitro (Fig. 2b).

Host immune responses are triggered by receptor–ligand interactions and downstream signaling. In response to microbial molecules, innate immune responses are activated through TLRs. TLR2 recognizes bacterial peptidoglycan, lipoteichoic acid and lipoproteins (Takeda et al., 2003). Lipopolysaccharide is recognized by TLR4 (Poltorak et al., 1998). Signals triggered by TLRs except for TLR3 are transduced through MyD88 (Takeda & Akira, 2004). MyD88 is an adaptor molecule that triggers the inflammatory response. MyD88 leads to the activation of mitogen-activated protein kinases and activates a transcription factor, NF-κB, to regulate the expression of inflammatory cytokine genes. Generally, TLR2 and TLR5 mediate cytokine induction by infection with L. monocytogenes (Hayashi et al., 2001; Seki et al., 2002). Our results showed that signal transduction through TLR4, but not TLR2, was involved in the production of TNF-α induced by recombinant p60 (Fig. 4a). TNF-α production by p60 stimulation was MyD88-dependent (Fig. 4b) and macrophages derived from TLR4-deficient C3H/HeJ mice failed to produce TNF-α in response to p60 stimulation (Fig. 4c). Our study also showed that recombinant p60 enhanced bacterial clearance at the early phase of L. monocytogenes infection in a TLR4-dependent manner (Fig. 4d and e). These results suggest that p60 enhances host innate immunity via upregulation of TNF-α, depending on TLR4.

Contaminated endotoxin of recombinant p60 may induce proinflammatory cytokines. We confirmed that the endotoxin below the level of detection by the Limulus amoebocyte lysate assay and the heat-treated recombinant p60 was not able to induce cytokine production in macrophages (Fig. 3a). In addition, PMB treatment did not alter the titer of TNF-α induced by p60 and the dose of endotoxin that might be present in recombinant p60 (<3 pg) could not induce TNF-α from RAW264.7 cells (Fig. 3b). From these results, we considered that recombinant p60 was not contaminated with lipopolysaccharide (which influences the effect of p60 on induction of cytokines). Generally, TLR4 recognizes lipids, such as lipopolysaccharide. However, recent studies showed that some bacterial and viral proteins are recognized by TLR4. Escherichia coli adhesin protein FimH is a ligand of TLR4 (Mossman et al., 2008), and pneumolysin from Streptococcus pneumoniae is recognized by TLR4 (Malley et al., 2003). Mouse RNA virus (Burzyn et al., 2004), hepatitis C virus (Machida et al., 2006), hepatitis B virus (Isogawa et al., 2005) and Kaposi sarcoma herpesvirus (Lagos et al., 2008) are also recognized by TLR4. Park (2004) also showed that listeriolysin O and other gram-positive cytolysins are TLR4 ligands. These studies suggest that protein could be a ligand for TLR4.

We have demonstrated here that recombinant p60 induces proinflammatory cytokines and modulates host immune responses. Previous studies showed that p60 also has potent antigenicity (second to listeriolysin O) (Pamer et al., 1991; Pamer, 1994). p60-specific CD8+ T cells induced by L. monocytogenes infection protected the host from L. monocytogenes infection (Geginat et al., 1998). It has been reported that p60 induces a strong antigen-specific response to CD4+ and CD8+ T cells (Geginat et al., 1999), and that p60 activates natural killer cells and induces an IFN-γ-dependent immune response (Humann et al., 2007). These studies suggest that p60 promotes both innate and adaptive immunity. We focused on the effect of p60 on innate immunity as the response through CD4+ and CD8+ T cells was not involved. Further studies on the effect of p60 on adaptive immunity through epitope recognition are required.

Acknowledgements

This study was supported by a Grant-in-Aid from the Zoonoses Control Project of the Ministry of Agriculture, Forestry and Fisheries of Japan and Grant for Hirosaki University Institutional Research.

Footnotes

  • Editor: Patrick Brennan

References

View Abstract