Tissue resident cells differentiate S. aureus from S. epidermidis via IL-1β following barrier disruption in healthy human skin

Abstract

The Staphylococcus sp. are a dominant part of the human skin microbiome and present across the body. Staphylococcus epidermidis is a ubiquitous skin commensal, while S. aureus is thought to colonize at least 30% of the population. S. aureus are not only colonizers but a leading cause of skin and soft tissue infections and a critical healthcare concern. To understand how healthy human skin may differentiate commensal bacteria, such as S. epidermidis, from the potential pathogen methicillin-resistant S. aureus (MRSA), we use ex vivo human skin models that allow us to study this host-bacterial interaction in the most clinically relevant environment. Our work highlights the role of the outer stratum corneum as a protective physical barrier against invasion by colonizing Staphylococci. We show how the structural cells of the skin can internalize and respond to different Staphylococci with increasing sensitivity. In intact human skin, a discriminatory IL-1β response was identified, while disruption of the protective stratum corneum triggered an increased and more diverse immune response. We identified and localized tissue resident Langerhans cells (LCs) as a potential source of IL-1β and go on to show a dose-dependent response of MUTZ-LCs to S. aureus but not S. epidermidis. This suggests an important role of LCs in sensing and discriminating between bacteria in healthy human skin, particularly in intact skin and provides a detailed snapshot of how human skin differentiates between friend and potential foe. With the rise in antibiotic resistance, understanding the innate immune response of healthy skin may help us find ways to enhance or manipulate these natural defenses to prevent invasive infection.

Fig 1. Inflammatory response to colonizing Staphylococcus sp. in human skin.

(A) CFU counts of S. epidermidis 1457 and S. aureus USA300 48 h after epicutaneous colonization of ex vivo human skin. Data show means ± SD. Dots represent individual data points, colours differentiate skin donors (n = 6). Dotted line depicts inoculum. Significance determined by one-way ANOVA with Tukey’s correction for multiple comparisons. Significance is denoted by *** p ≤ 0.001 or **** p ≤ 0.0001. (B) IL-8 ELISA, as x-fold change over uninfected controls. Dots represent individual data points, colours differentiate skin donors (n = 6). Significance determined by unpaired t-test with Welch’s correction. ns = not significant (C) Representative immunofluorescence staining of cytokeratin-10 (orange), f-actin (white), nuclei (blue) in skin model topically colonized with S. aureus USA300 GFP (green) for 48 h. Scale bars = 50μm. N = 3. (D) Insert of area in (C) showing intracellular S. aureus USA300 GFP.

Fig 2. Intracellular Staphylococcus sp. induce an IL-8 response in keratinocytes and fibroblasts, but not corneocytes.

(A) Intracellular CFU counts of S. epidermidis 1457 (filled circles) and S. aureus USA300 (empty circles) 3 h p.i. of tape-stripped corneocytes, human keratinocytes (HEKa) and fibroblasts (HDFn). Bars represent means ± SD, dots show individual data points (corneocytes and fibroblasts n = 3; keratinocytes n = 5; 3 technical replicates). Dotted line = initial inoculum. Significance was determined by a two-way ANOVA with Tukey’s correction for multiple comparisons. Significance is denoted by * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001 or **** p ≤ 0.0001. (B) IL-8 response measured by ELISA in cell culture supernatants of S. epidermidis 1457 or S. aureus USA300 infected keratinocytes and fibroblasts. Dot plot shows individual data points (fibroblasts n = 3; keratinocytes n = 5; 3 technical replicates each). Significance was determined by a two-way ANOVA with Tukey’s correction for multiple comparisons. Significance is denoted by ** p ≤ 0.01 or **** p ≤ 0.0001. Only significant differences are annotated for clarity. (C) Microscopic detection of intracellular S. aureus GFP and S. epidermidis GFP in cell monolayers of human keratinocytes (HEKa) or fibroblasts (HDFn); uninfected CTR vs 3h infection. Immunostaining of f-actin (red) and nuclei (blue) in HEKa and HDFn. Scale bars = 50μm.

Fig 3. Stratum corneum disruption results in increased inflammatory response to S. aureus colonization.

(A) CFU counts of S. epidermidis 1457 and S. aureus USA300 48 h after epicutaneous colonization of biopsies with disrupted stratum corneum (s.c.). Bars = means ± SD, dots show individual data points (n = 7, skin donors indicated by colors). Dotted line depicts inoculum. Significance determined by one-way ANOVA with Tukey’s correction for multiple comparisons. Significance denoted by **** p ≤ 0.0001. (B) IL-8 response displayed as x-fold change over uninfected controls measured by ELISA in homogenates of skin biopsies with disrupted s.c. inoculated with S. epidermidis 1457 or S. aureus USA300. Dot plot shows individual data points (n = 7, skin donors indicated by colors). Significance determined by unpaired t test with Welch’s correction. Significance denoted by * p ≤ 0.05. (C-D) Cytokine expression displayed as x-fold change from uninfected controls in colonized skin explants with an intact s.c. (C) or disrupted s.c. (D) measured by LEGENDplex bead-based immunoassay. The box plots represent interquartile range (IQR) with the median shown as black line. Dots show individual data points (n = 6, 2–3 individual skin donors). Statistical significance determined by two-way ANOVA with Šídák’s correction for multiple comparisons and denoted by ** p ≤ 0. 01 and **** p ≤ 0. 0001. n.d. = not detected. Only significant differences are annotated for clarity. (E) Selected cytokines on a radar plot shown as mean x-fold changes over uninfected controls. (F) Spearman correlation between CFUs and IL-1β, IL-17A and IL-12p70 measured in human skin biopsies 48 h p.i. Significance for each cytokine compared to the CFU/ml denoted by ** p ≤ 0. 01 and **** p ≤ 0. 0001.

Fig 4. Skin resident Langerhans cells are located throughout the epidermis and MUTZ-LCs can differentiate S. aureus from S. epidermidis.

(A) Orthogonal projection of an epidermal sheet close to the basal membrane. Immunofluorescent staining of f-actin (red), nuclei (blue) and langerin (green). Scale bar = 20μm. Arrow indicates Langerhans cell (green) (B) Orthogonal projection of an epidermal sheet close to the stratum corneum. Immunofluorescent staining of f-actin (red), nuclei (blue) and langerin (green). Scale bar = 20μm. Arrows indicate overlap of Langerhans cells projections with keratinocytes cell walls. Scale bar = 20μm. (C) XZ projection of an epidermal sheet showing a Langerhans cell body at the base of the epidermis with projections extending to the skin surface. F-actin (red), nuclei (blue) and langerin (green). Scale bar = 20μm. (D) Maximum projection of MUTZ-LCs. F-actin (red), nuclei (blue) and langerin (green). Scale bar = 25μm. (E) IL-1β measured by ELISA in MUTZ-LCs supernatants at 24h p.i. after 3 h infection with S. epidermidis 1457 or S. aureus USA300. Bars represent means ± SD, dots show individual data points (n = 3–6 experimental repeats, with 2 technical replicates). Statistical significance determined by two-way ANOVA with Tukey’s correction for multiple comparisons. Statistical significance denoted by ** p ≤ 0.01 or **** p ≤ 0.0001. Only significant differences are annotated for clarity. (F) IL-1β measured by ELISA in keratinocyte (HEKa) supernatants at 24h p.i. after 3 h infection with S. epidermidis 1457 or S. aureus USA300. Bars represent means ± SD, dots show individual data points (n = 3 experimental repeats, with 2 technical replicates). Statistical significance determined by two-way ANOVA with Tukey’s correction for multiple comparisons. Statistical significance denoted by **** p ≤ 0.0001. Only significant differences are annotated for clarity. (G) IL-1β measured by ELISA in fibroblast (HDFn) supernatants at 24h p.i. after 3 h infection with S. epidermidis 1457 or S. aureus USA300. Bars represent means ± SD, dots show individual data points (n = 3 experimental repeats). Detection limit of ELISA shown by dotted line.