Identification of a crucial residue required for Staphylococcus aureus LukAB cytotoxicity and receptor recognition

Abstract

The bicomponent leukotoxins produced by Staphylococcus aureus kill host immune cells through osmotic lysis by forming β-barrel pores in the host plasma membrane. The current model for bicomponent pore formation proposes that octameric pores, comprised of two separate secreted polypeptides (S and F subunits), are assembled from water-soluble monomers in the extracellular milieu and multimerize on target cell membranes. However, it has yet to be determined if all staphylococcal bicomponent leukotoxin family members exhibit these properties. In this study, we report that leukocidin A/B (LukAB), the most divergent member of the leukotoxin family, exists as a heterodimer in solution rather than two separate monomeric subunits. Notably, this property was found to be associated with enhanced toxin activity. LukAB also differs from the other bicomponent leukotoxins in that the S subunit (LukA) contains 33- and 10-amino-acid extensions at the N and C termini, respectively. Truncation mutagenesis revealed that deletion of the N terminus resulted in a modest increase in LukAB cytotoxicity, whereas the deletion of the C terminus rendered the toxin inactive. Within the C terminus of LukA, we identified a glutamic acid at position 323 that is critical for LukAB cytotoxicity. Furthermore, we discovered that this residue is conserved and required for the interaction between LukAB and its cellular target CD11b. Altogether, these findings provide an in-depth analysis of how LukAB targets neutrophils and identify novel targets suitable for the rational design of anti-LukAB inhibitors.

FIG 1

LukAB exists as a heterodimer in solution and in S. aureus culture filtrate. (A) Purification of His-tagged leukotoxin S subunits (LukA, LukE, LukS-PV, and HlgC) in the presence of F subunits (LukB, LukD, LukF-PV, and HlgB) by metal affinity chromatography. Total protein staining shows input and purified fractions. (B) Intoxication of PMN-HL60 cells with E. coli purified LukA and LukB as individual or combined subunits at the indicated concentrations for 1 h. Cell viability was determined by measuring cellular metabolism with CellTiter. (C) Intoxication of PMN-HL60 cells with LukAB as a copurified heterodimer from S. aureus at the indicated concentrations for 1 h. Cell viability was monitored with CellTiter. (D) Immunoblot of purified LukAB incubated with 0, 1, 2, or 4 mM glutaraldehyde where LukA or LukB was detected with anti-His (LukA) or anti-LukB antibodies. (E) Immunoblot of S. aureus culture filtrate incubated with 0, 1, 2, or 4 mM glutaraldehyde where LukA or LukB was detected with anti-LukA or anti-LukB antibodies. For panels B and C, data are represented as the averages of triplicate samples ± standard deviations from at least two independent experiments.

FIG 2

The distinct LukA N- and C-terminal extensions affect the cytolytic activity of LukAB. (A) Amino acid sequence alignment of the leukotoxin S-subunit N and C termini using DNAStar MegAlign software. (B) Purification of the His-tagged wild-type (WT) LukA, LukA C-terminal deletion mutant (Δ10C), or LukA N-terminal deletion mutant (Δ33N) and copurification of LukB by metal affinity chromatography from S. aureus. Total protein staining of 2 μg of the purified products and immunoblotting with anti-His and anti-LukB to detect the LukA and LukB subunits, respectively, are shown. (C) Immunoblot of purified Δ10C LukAB or Δ33N LukAB incubated with 0, 1, 2, or 4 mM glutaraldehyde where LukB was detected with an anti-LukB antibody. (D and E) Intoxication of PMN-HL60 cells with the indicated concentrations of the WT, Δ10C, or Δ33N LukAB proteins for 1 h. Cell viability was measured with CellTiter (D), and pore formation was evaluated with the fluorescent dye ethidium bromide (EtBr) (E). For panels D and E, data are represented as the averages of triplicate samples ± standard deviations from at least two independent experiments. RFU, relative fluorescence units.

FIG 3

The glutamic acid residue at position 323 of the LukA C terminus is essential for LukAB activity. (A) Copurification of the His-tagged single-amino-acid LukA C-terminal mutants with LukB from S. aureus culture filtrates by metal affinity chromatography. Total protein staining of 2 μg of the purified products with purified WT and Δ10C LukAB shown for comparison (top) and corresponding cytotoxicity data after a 1-hour intoxication of PMN-HL60 cells with 2.5 μg/ml of the indicated proteins. (B) Immunoblot of purified E323A LukAB incubated with 0, 1, 2, or 4 mM glutaraldehyde where LukB was detected with an anti-LukB antibody. (C) Intoxication of PMN-HL60 cells with high concentrations of WT, Δ10C, or E323A LukAB. Cell viability was measured with CellTiter. (D) Pore formation by WT, Δ10C, or E323A LukAB on PMN-HL60 cells following a 1-hour intoxication with 10 μg/ml of toxin, as determined by EtBr incorporation. (E) Sequence alignment of the 10 amino acids composing the C-terminal region of LukA from representative S. aureus strains in the protein database (NCBI) was performed using the NCBI BLAST protein. The glutamic acid at position 323 is highlighted in bold, and residues with observed polymorphisms are marked with asterisks. Data are represented as the averages of triplicate samples ± standard deviations from at least two independent experiments. *** indicates P < 0.0001 by one-way analysis of variance.

FIG 4

The E323 residue in the C terminus of LukA is essential for LukAB-mediated killing of PMNs by extracellular and phagocytosed USA300. (A) Immunoblotting to detect LukA, LukB, LukS-PV, and Hla levels in WT, ΔlukAB, and ΔlukAB strains chromosomally complemented with WT lukAB or lukAB with the indicated C-terminal mutations using toxin-specific antibodies. (B) Infection of human PMNs for 1 h with various MOIs of the indicated USA300 strains under nonphagocytosing conditions. PMN membrane damage was evaluated with the fluorescent dye SYTOX green. (C) Infection of human PMNs for 1 h with an MOI of 10 of the indicated opsonized USA300 strains under phagocytosing conditions. PMN membrane damage was evaluated with the fluorescent dye SYTOX green. (D) Growth rebound of the indicated opsonized USA300 strains during infection with human PMNs at an MOI of 10 under phagocytosing conditions. Bacterial CFU were determined at 60, 120, or 180 min postsynchronization and were normalized to input at time zero, which was set at 100%. For panels B to D, results represent the means from PMNs isolated from 6 donors ± standard errors of the means from at least two independent experiments. * indicates P < 0.05, ** indicates P < 0.01, and *** indicates P < 0.0001 by one-way analysis of variance.

FIG 5

The interaction between LukAB and its receptor CD11b requires the E323 residue in the C terminus of LukA. (A) Circular dichroism (CD) analyses indicate that the wild-type (black), E323A (dark gray), and Δ10C (gray) proteins are stable and properly folded, with similar melting temperatures, 45.8°C, 47.0°C, and 43.0°C, respectively. Temperature melts were conducted from 4 to 95°C at a wavelength of 208 nm. (B) Competition dot blot assay where purified recombinant human CD11b I domain was membrane bound and then incubated with 5 μg/ml fluorescently labeled LukAB (Alexa 488-LukAB) in the presence of a 10-fold excess (50 μg/ml) of unlabeled WT LukAB, the 10C or E323A LukAB mutant, or PVL. Alexa 488-LukAB binding was quantified by densitometry. Data represent the averages of triplicate samples ± standard errors of the means from at least two independent experiments. (C and D) Measurement of the interaction of LukAB with human CD11b I domain (C) or human Mac-1 (D) by SPR. Representative sensorgrams of two experiments performed in triplicate are shown. RU, relative units.