Staphylococcus aureus induces Gasdermin A-dependent keratinocyte pyroptosis

Abstract

Staphylococcus aureus is a common colonizer of human skin, which, despite its ubiquitous nature, has a high virulence potential. Tolerating microbes in health but responding effectively to pathogens presents a challenge to the barrier tissues. Here, we examined the interaction of S. aureus with skin keratinocytes to study this early step of pathogenesis and pathogen discrimination. During infection, the S. aureus protease Staphopain A (ScpA) cleaves inert Gasdermin A (GSDMA). This releases an active N-terminal fragment similar to that formed by host protease regulators of other gasdermins family members. The resulting cell death by pyroptosis allows keratinocytes to deprive invasive S. aureus of an intracellular niche. These data support a model of GSDMA as an autonomous sensor of pathogenicity, in contrast to the conventional regulation of other gasdermins, which have dedicated host cell pathways. Gasdermins abundant in other tissues may have similar functions in host defense for the threat assessment of a microbe.

Fig. 1. S. aureus induces ScpA-dependent cell death of keratinocytes.

Primary human keratinocytes were pretreated with 10 μM cytochalasin D where indicated, then infected with wild-type S. aureus WBG10049 (WT) (multiplicity of infection (MOI) = 50), staphopain A knockout (ΔscpAB), and scpAB complemented ΔscpAB (ΔscpAB+ pscpAB) for 6 h, and a lysis measured by lactate dehydrogenase (LDH) release, and b staphopain A activity measured using the specific substrate Mca-EAYVHDAPV-K(Dnp)-NH₂ (sub111). n = 4. c Immunofluorescence microscopy images of keratinocyte permeabilization to propidium iodide after 5 h S. aureus (WT), or ΔscpAB infection (MOI = 25). Scale bar, 50 μm. Image representative of three independent experiments. d Keratinocytes were infected (MOI = 50) with S. aureus (red), ΔscpAB (teal), and ΔscpAB+ pscpAB (black), and cell death was assessed at 1, 3, and 6 h post-infection by LDH release. n = 4 e, f Keratinocytes were infected with a panel of S. aureus clinical isolates (MOI = 50) for 6 h. ScpA activity was measured with sub111, and lysis was measured by LDH release. WBG10049 and ΔscpAB controls are shown in red and teal, respectively. n = 4. g–j Keratinocytes were infected (MOI = 50), for g, i 3 h or h, i 6 h, then g, h intracellular, n = 4, or i, j total, n = 6, S. aureus quantified by plating CFU. a, b, d, e, g–j Data are representative of three independent experiments and presented as mean ± SD, one-way ANOVA with Dunnett’s multiple comparison test was used to determine statistical significance.

Fig. 2. S. aureus induces GSDMA-dependent cell death.

a Lysates of Flag-GSDM-transfected HEK293T cells incubated with ScpA for 1 h and analyzed by immunoblot. Blot is representative of three independent experiments. b RNAseq measurements of expression of each gasdermin in healthy human skin, expressed as normalized transcripts per million (nTPM). HeLa cells transfected with GSDMA for 24 h were infected with S. aureus (MOI = 50), ΔscpAB, and pscpAB complemented ΔscpAB. After 4 h, GSDMA cleavage was detected by c immunoblot, and after 6 h d lysis was measured by LDH release (n = 4). Blot is representative of three independent experiments. e Mouse keratinocytes from C57BL/6 J WT (dark purple) or Gsdma1–3 KO (light purple) mice were infected for 6 h with WT S. aureus (MOI = 25), or ΔscpAB and lysis was measured by LDH release. n = 4. f Cytotoxicity of human GSDMA transfected with or without staphopain A (ScpA) or SpeB into HeLa cells after 24 h measured by LDH release. n = 4. g Human keratinocyte lysis measured by LDH release after treatment with ScpA dilutions packaged within polyethylenimine (PEI) liposomes, or ScpA alone after 2 h. n = 4. d–g Data are representative of three independent experiments and are presented as mean ± SD, one-way ANOVA with Dunnett’s multiple comparison test was used to determine statistical significance.

Fig. 3. Staphopain A cleaves and activates GSDMA.

a SDS–PAGE of recombinant human GSDMA cleavage by purified ScpA in the presence or absence of E-64 after 1 h. Blot is representative of three independent experiments. b Binding of GSDMA with or without ScpA to liposomes consisting of a mixture of phosphatidylcholine (PC), phosphatidylethanolamine (PE) and cardiolipin (CL) analyzed by SDS–PAGE. The gold arrow identifies GSDMA-N. Blot is representative of three independent experiments. c Summary of experimental results in (b). d Liposome leakage monitored by sulforhodamine B (SRB) fluorescence on incubation with GSDMA with or without ScpA. Detergent added after 21 min (dotted line). Data are representative of three independent experiments and are presented as mean ± SD of biological replicates, n = 4. e SDS–PAGE of recombinant human GSDMA cleavage by ScpA over time in the presence or absence of liposomes made from PC, PE, and CL. The gold arrow identifies GSDMA-N. Summary of experimental results shown above SDS–PAGE. Blot is representative of three independent experiments. f Cleavage site on GSDMA identified by Edman sequencing. c Created in BioRender. LaRock, D. (2025) https://BioRender.com/0w1s8az (e) Created in BioRender. LaRock, D. (2025) https://BioRender.com/zgdj89u.

Fig. 4. GSDMA protects mice in S. aureus skin infection.

a Human GSDMA gene arrangement compared to mouse Gsdma1–3. Expression of mouse Gsdma in keratinocytes is indicated. b Lysates of HEK293T cells transfected with the indicated Flag-GSDMs were incubated with purified ScpA for 1 h and analyzed for immunoblotting for Flag-GSDMs. Data are representative of three independent repeats. Wildtype (C57BL/6 J–dark purple) and GSDMA-deficient (gsdma123^−/−–light purple) mice after epicutaneous application of 10⁸ CFUs of S. aureus and c transepithelial water loss (TEWL) monitored, and skin imaged at day 7. Results represent mean ± SD, n = 6 for each point. d At day 2 and 7 of infection, skin was examined by Hematoxylin and Eosin stain. Scale bar is 500 μM. e Colony-forming units (CFUs) of S. aureus at the site of infection were measured by plating 2 days post-infection (n = 9, WT, n = 8, ΔscpAB) and 7 days post-infection (n = 17). Results represent mean ± SD with individual biological replicates shown; one-way ANOVA with Dunnett’s multiple comparison test was used to determine statistical significance.