Dissociated linkage of cytokine-inducing activity and cytotoxicity to different domains of listeriolysin O from Listeria monocytogenes

Abstract

Listeriolysin O (LLO), a cholesterol-binding cytolysin of Listeria monocytogenes, exhibits cytokine-inducing and cytolytic activities. Because the cytolytic activity was abolished by cholesterol treatment but the cytokine-inducing activity was not, these activities appeared to be linked to different domains of the LLO molecule. In this study, we constructed recombinant full-length LLO (rLLO529) and various truncated derivatives and examined their cytolytic, cholesterol-binding, and gamma interferon (IFN-gamma)-inducing activities. rLLO529 exhibited both IFN-gamma-inducing and cytolytic activities. Four truncated rLLOs possessing different C termini, which did not exert either cytolytic or cholesterol-binding activity, stimulated IFN-gamma production in normal spleen cells. However, a truncated rLLO corresponding to domain 4 (rLLO416-529) did not exhibit IFN-gamma-inducing activity, whereas it did bind to immobilized cholesterol. In addition, though the hemolysis induced by rLLO529 was inhibited by rLLO416-529, such inhibition was not detected upon rLLO529-induced IFN-gamma production. These data indicated that domain 4 was responsible for binding of LLO to membrane cholesterol followed by oligomerization and pore formation by the entire LLO molecule. In contrast, the other part of LLO, corresponding to domain 1-3, was essential for IFN-gamma-inducing activity. These findings implied a novel aspect of the function of LLO as a bacterial modulin.

FIG. 1.

Construction of various rLLOs. The amino acid sequence is numbered from the N terminus. Gray bars represent polypeptide regions of full-length and truncated LLOs. The filled box in the LLO structure represents a region of the undecapeptide (ECTGLAWEWWR; residues 483 to 493) which is common among thiol-activated cytolysins.

FIG. 2.

SDS-PAGE and Western blot analysis for the purity of rLLOs. The purified rLLOs were subjected to SDS-PAGE with a 10 to 20% gradient gel. (A) The gel was stained with Coomassie brilliant blue. (B) The proteins were electroblotted to a PVDF membrane, and Western blotting was performed with anti-His tag antibody. The positions of molecular mass standards are indicated on the left. Lanes: A, rLLO529; B, rLLO493; C, rLLO482; D, rLLO415; E, rLLO59-415; F, rLLO416-529

FIG. 3.

Cholesterol-dependent hemolysis induced by rLLO529 and comparison of hemolytic activity among rLLOs. (A) rLLO529 (1,000 nM) was serially diluted twofold with PBS and treated with cholesterol (10 μg/ml) or left untreated. Four volumes of SRBC suspension was added to make a final concentration of 0.5%, and the suspension was incubated for 30 min at 37°C. Hemolysis was measured by release of hemoglobin in the supernatant. (B) Hemolysis caused by rLLOs was measured as described above. HU, hemolytic units.

FIG. 4.

Binding of rLLOs to immobilized cholesterol. (A) Different amounts of cholesterol were plated in 96-well microtiter plates with PVDF membranes at the bottom. Cholesterol-treated or untreated rLLO529 (2 nM) was added, and the wells were incubated with anti-His tag antibody followed by HRP-conjugated anti-mouse IgG. rLLO529 bound to immobilized cholesterol was detected by the addition of TMB to the wells. (B) Cholesterol (0.5 μg) was plated in the PVDF membrane-containing 96-well plates. Recombinant LLOs were adjusted at 2 nM, and the binding activities to cholesterol were determined. (C) Various concentrations of cholesterol were spotted on TLC plates. TLC plates were incubated with 100 nM rLLOs and then incubated with monoclonal anti-His tag antibody followed by HRP-conjugated anti-mouse IgG. Binding of rLLOs to cholesterol was detected by peroxidase staining.

FIG. 5.

Cytotoxicity of rLLO529 for J774.1 cells. J774.1 cells (10⁵ cells) were cultured overnight in 48-well plate. The supernatant was removed, and various concentrations of rLLO529 were added with or without cholesterol treatment. LDH released in the supernatant was measured 6 h after addition of rLLO529 to the cells.

FIG. 6.

IFN-γ production by normal spleen cells stimulated with rLLO529. (A) Normal spleen cells were stimulated for 24 h with various concentrations of rLLO529 which had been treated with cholesterol or left untreated. The supernatant was collected and the level of IFN-γ was measured by enzyme-linked immunosorbent assay. (B) Normal spleen cells were stimulated with 50 nM cholesterol-treated rLLO529 or LPS (0.1 and 1 ng/ml) in the presence or absence of polymyxin B (PMB; 0.5 μg/ml) and anti-LLO antibody (Ab) (1 μg/ml) for 24 h.

FIG. 7.

IFN-γ-inducing activity of rLLO529 and five truncated LLOs. Normal spleen cells were stimulated with various concentrations of cholesterol-treated rLLOs for 24 h, and the level of IFN-γ production was determined. ND, not done.

FIG. 8.

Inhibitory effect of rLLO416-529 (domain 4) on hemolytic activity but not on IFN-γ-inducing activity of rLLO529. Normal spleen cells were treated with various concentrations of rLLO416-529 for 1 h at 4°C and were stimulated with 50 nM cholesterol-treated rLLO529 for 24 at 37°C. The supernatant was collected, and the IFN-γ in the supernatant was measured. SRBC (1%) were treated with various concentrations of rLLO416-529 for 1 h at 4°C. rLLO529 was added to the SRBC suspension at 50 nM, and the mixture was incubated for 30 min at 37°C. The hemolysis caused by rLLO529 was determined. The inhibitory activity of rLLO416-529 for IFN-γ production and hemolysis was expressed as percent decrease of activity in the presence of rLLO416-529.

FIG. 9.

Inhibitory effect of rLLO416-529 on the binding of rLLO529 to cholesterol. Cholesterol (0.06 μg) was spotted on a TLC sheet. The sheet was treated with 0 (lane 1), 500 (lane 2), 1,000 (lane 3), and 2,000 (lane 4) nM rLLO416-529 for 1 h. After several washes, the sheet was incubated with 50 nM rLLO529 for 15 min. Silica gel was scraped off, and bound rLLO416-529 and rLLO529 were detected by Western blotting using anti-His tag antibody.