Unlatching of the stem domains in the Staphylococcus aureus pore-forming leukocidin LukAB influences toxin oligomerization

Abstract

Staphylococcus aureus (S. aureus) is a serious global pathogen that causes a diverse range of invasive diseases. S. aureus utilizes a family of pore-forming toxins, known as bi-component leukocidins, to evade the host immune response and promote infection. Among these is LukAB (leukocidin A/leukocidin B), a toxin that assembles into an octameric β-barrel pore in the target cell membrane, resulting in host cell death. The established cellular receptor for LukAB is CD11b of the Mac-1 complex. Here, we show that hydrogen voltage-gated channel 1 is also required for the cytotoxicity of all major LukAB variants. We demonstrate that while each receptor is sufficient to recruit LukAB to the plasma membrane, both receptors are required for maximal lytic activity. Why LukAB requires two receptors, and how each of these receptors contributes to pore-formation remains unknown. To begin to resolve this, we performed an alanine scanning mutagenesis screen to identify mutations that allow LukAB to maintain cytotoxicity without CD11b. We discovered 30 mutations primarily localized in the stem domains of LukA and LukB that enable LukAB to exhibit full cytotoxicity in the absence of CD11b. Using crosslinking, electron microscopy, and hydroxyl radical protein footprinting, we show these mutations increase the solvent accessibility of the stem domain, priming LukAB for oligomerization. Together, our data support a model in which CD11b binding unlatches the membrane penetrating stem domains of LukAB, and this change in flexibility promotes toxin oligomerization.

Keywords: Staphylococcus aureus (S. aureus); bacterial toxin; high-throughput screening; oligomerization; receptor.

Figure 1

HVCN1 is required for LukAB-mediated cell death.A, a schematic of LukAB depicting the rim, cap and stem domains (PDB 5k59 (24)). The cap domain is colored in beige, the rim domain in teal, and the stem domain in magenta. LukA and LukB are labeled. B, intoxication of THP-1 transduced with LentiCRISPRv2 expressing non-targeting (NT) sgRNA with LukAB allelic variants. Cell death measured by LDH release. Error bars represent SEM; N = 4. C, intoxication of THP-1 transduced with LentiCRISPRv2 expressing HVCN1 sgRNA with LukAB allelic variants. Cell death measured by LDH release. Error bars represent SEM; N = 4. D, intoxication of CHO cells stably transduced with CD18, CD18/CD11b, CD18/HVCN1, or CD18/CD11b HVCN1 with CC8 LukAB. Cell death measured by LDH release. Error bars represent SEM. Statistical significance was determined by two-way ANOVA with Dunnett’s multiple comparisons test to CD18 CHO (∗∗∗∗p ≤ 0.001; ∗∗p = 0.0021; ∗p = 0.0363; no stars = not significant). Data is representative of six to eight independent experiments using six different protein preps for an N = 15 to 30. E, binding of CC8 LukAB to CHO cells stably transduced with CD18 shown as a control, and CD18/CD11b measured by Δ Median Fluorescence Intensity (ΔMFI) with PBS signal subtracted. Bound LukAB was detected with an anti-His PE-conjugated antibody by flow cytometry. Error bars represent SEM. Data is representative of six independent experiments using two different protein preps for an N = 12. F, binding of CC8 LukAB to CHO cells stably transduced with CD18 shown as a control, and CD18/HVCN1 measured by ΔMFI with PBS signal subtracted. Bound LukAB was detected with an anti-His PE-conjugated antibody by flow cytometry. Error bars represent SEM. Data is representative of six independent experiments using two different protein preps for an N = 12.

Figure 2

Alanine scanning screen identifies mutations that allow CC8 LukAB to forgo dependency on CD11b.A, intoxication of CD18/HVCN1 CHO with LukAB-containing lysates. Each dot represents lysate from a different point mutant. WT LukAB, controls marked with dashed lines and squares outlined in teal, brown, and gray, respectively. LukAB-containing lysates that kill ≥ 70% of cells are shaded in red. Cell death measured by LDH release. Error bars represent SEM; N = 3. Also refer to Extended Data Figure S2. B, mapping of the top 30 residues (variants highlighted in red in Fig. 2A) on the LukAB structure that, when substituted for alanine, allow for enhanced CD11b-independent activity. Left, LukAB soluble dimer state (PDB 5k59 (24)); right, a dimer taken from the octameric state (PDB 4tw1 (22)). Light green ribbon = LukA, dark green ribbon = LukA stem domain. Light pink ribbon = LukB, magenta ribbon = LukB stem domain. Top 30 residues are highlighted with red spheres. C, heat map representation of the intoxication of CD18/HVCN1 CHO with crude purified LukAB variants that gained CD11b-independent cytolytic activity. Cell death measured by LDH release. Heat map shows the mean of N = 3 independent experiments. D, mapping of the mutated residues of select variants chosen for S. aureus expression, purification, and characterization on the dimeric structure (PDB 5k59 (24)). Light green ribbon = LukA, light pink ribbon = LukB. Purple sphere = G149 LukA, pink sphere = Y145 LukA, navy blue sphere = Y165 LukA, green sphere = V113 LukB, light blue sphere = K120 LukB. E, intoxication of CD18/HVCN1 CHO with purified WT and select LukAB variants. Cell death measured by LDH release. Error bars represent SEM; N = 3. EV, empty vector; Unt, untransformed.

Figure 3

Mutating residues at the stem/cap interface gives LukAB a cytotoxic advantage in the absence of CD11b.A, intoxication of CD18/CD11b CHO with purified WT and select LukAB variants. Cell death measured by LDH release. Error bars represent SEM; N = 4. B, intoxication of THP-1 scramble shRNA with purified WT and select LukAB variants. Cell death measured by LDH release. Error bars represent SEM; N = 4. C, intoxication of THP-1 CD11b shRNA with purified WT and select LukAB variants. Cell death measured by LDH release. Error bars represent SEM; N = 4. D, intoxication of primary human PMNs with purified WT and select LukAB variants. Cell death measured by CellTiter. Error bars represent SEM; N = 11 independent donors.

Figure 4

Disrupting the interface at the LukAB stem domain promotes LukAB to oligomerize more readily.A, Western blots of WT, G149A, K120A, and V113A LukAB un-crosslinked with no additive, crosslinked with no additive, crosslinked with 20% MPD, or crosslinked with 40% MPD. LukAB was detected with anti-His antibody. Schematic on the right represents predicted LukAB oligomeric states. Images are representative of N = 3 independent blots. B, negative stain electron micrographs of WT and V113A LukAB in 0% MPD, 20% MPD or 40% MPD. Scale bar is 50 nm. Examples of LukAB oligomers are boxed in yellow. C–F, 2D class averages of WT LukAB in 40% MPD (C), V113A LukAB in 0% MPD (D), V113A LukAB in 20% MPD (E), and V113A LukAB in 40% MPD (F), labeled with the number of particles in each class. Scale bar is 100 Å. G. Overlay of the LukAB octamer (PDB 4tw1 (22)) with negative stain 2D class averages for WT and V113A shows that circular-shaped negative stain classes are consistent in size and shape with a LukAB octamer.

Figure 5

Oxidative footprinting reveals differentially exposed epitopes.A, schematic describing the workflow for LukAB oxidative footprinting. LukAB (gray surface model, PDB 5k59 (24)) was incubated with hydrogen peroxide to oxidize surface-exposed residues. The figure is modeling a predicted example; if the stem domain was folded in a stable dimer state, we hypothesize the stem/cap interface will be shielded from solvent (gray) while residues exposed to solvent become oxidized (LukA, green; LukB, pink). LukAB was digested with trypsin, and LC/MS/MS was carried out to determine which peptides are oxidized, probing for differences in solvent accessibility between variants. B, bar graph representing the percentage of LukA peptides oxidized for WT, V113A, and K120A. X-axis shows starting residue position followed by single letter amino acid codes. Error bars represent the mean with SD. Statistical significance determined by two-way ANOVA with Dunnett’s multiple comparisons test to WT (∗∗p ≤ 0.01; ∗p ≤ 0.05; no stars = ns); N = 3 technical replicates. V113A: K122-K131 p = 0.0363, V139-K150 p = 0.0102, F151-R159 p = 0.0168, N206-R215 p = 0.0499, Y233-R240 p = 0.0470, S241-K255 p = 0.0095; K120A: K122-K131 p = 0.0239, F151-R159 p = 0.0282. Also refer to Extended Data Figure S4. C, bar graph representing the percentage of LukB peptides oxidized for WT, V113A and K120A. X-axis shows starting residue position followed by single letter amino acid codes. Pink bar represents peptide from K120A that is in a similar region but not digested the same as WT and V113A due to the alanine mutation (See Fig. S8). Error bars represent the mean with SD. Statistical significance determined by two-way ANOVA with Dunnett’s multiple comparisons test to WT (∗∗∗p ≤ 0.001; ∗∗p ≤ 0.01; ∗p ≤ 0.05; no stars = ns); N = 3 technical replicates. V113A: T121-R129 p = 0.0025, M219-K240 p = 0.0417, S245-K252 p = 0.0335; K120A: A57-R66 p = 0.0009, G130-K138 p = 0.0077, G165-R184 p = 0.0003, S245-K252 p = 0.0354. Also refer to Extended Data Figure S4. D, mapping of statistically significant peptides found in V113A that differ in oxidation levels compared to WT on the LukAB dimer (left) and octamer (right) for visualization (PDB 5k59 (24), 4tw1 (22)). Green sphere shows V113A residue location. Grey = no significant difference. Yellow represents an increase in peptide oxidation with p ≤ 0.05; Orange represents an increase in peptide oxidation with p ≤ 0.01; Light blue represents a decrease in peptide oxidation with p ≤ 0.05. Stem domains highlighted with dashed circle. E, mapping of statistically significant peptides found in K120A that differ in oxidation levels compared to WT on the LukAB dimer (top) and octamer (bottom) for visualization (PDB 5k59 (24), 4tw1 (22)). Blue sphere shows K120 residue location. Grey = no significant difference. Yellow represents an increase in peptide oxidation with p ≤ 0.05; Orange represents an increase in peptide oxidation with p ≤ 0.01; Red represents an increase in peptide oxidation with p ≤ 0.001; Light blue represents a decrease in peptide oxidation with p ≤ 0.05; Navy blue represents a decrease in peptide oxidation with p ≤ 0.001; Magenta represents peptide which has >30% oxidation in K120A, yet this exact peptide is not seen in WT due to digest differences (See Fig. S8). Stem domains highlighted with dashed circle.

Figure 6

Proposed model of the CC8 LukAB pore-forming process. The CC8 LukAB heterodimer is represented as green and pink ovals, with their corresponding stem domains as lines. Integrin CD18/CD11b is represented as a light blue and dark blue complex, while proton channel HVCN1 is represented in purple. (1) Secreted LukAB dimer engages CD11b, which can trigger a conformational change in the stem domain, allowing it to be more flexible. (2) This flexibility stimulates LukAB to oligomerize into an octamer. Whether LukAB forms a pre-pore or membrane-inserted pore here is unclear, since what stage the stem domains perforate the membrane remains unknown. With an alanine mutation in the stem domain of LukA or LukB (yellow star), the LukAB stem domain is more flexible, which primes LukAB to oligomerize and allows for the bypass of CD11b (yellow arrow), only requiring HVCN1 to function. (3) HVCN1 is an essential receptor, yet how it contributes to LukAB activity is unknown. (4) Following CD11b and HVCN1 engagement, LukAB forms an octameric β-barrel pore that penetrates the target cell membrane, resulting in cell lysis.