Crosstalk of Keratinocytes with Commensals and Neutrophils shapes Staphylococcus aureus Skin Colonization

Abstract

Our skin is constantly exposed to a large number of pathogens while at the same time undergoing selective colonization by commensal microorganisms such as Coagulase-negative Staphylococci. Staphylococcus aureus, however, is a facultative pathogen that is usually absent from healthy skin but frequently colonizes the inflamed skin of atopic dermatitis (AD) patients where it further promotes inflammation. Thereby, increasing S. aureus colonization was shown to correlate with a loss of microbiome diversity indicating a role for skin commensals to shape pathogen colonization. Keratinocytes, as the most abundant and outermost cell type in the epidermis, need to discriminate commensals from pathogens and orchestrate subsequent immune reactions in response to colonizing microbes. However, the mechanisms how individual commensals cooperate with keratinocytes and the immune system of the skin to prevent pathogen colonization are barely understood. Therefore, this work aimed at investigating the functional effects of two skin commensals, S. epidermidis and S. lugdunensis, on S. aureus skin colonization. Using an in vitro adhesion and invasion assay with primary human keratinocytes and an epicutaneous mouse skin colonization model we show that pretreatment with S. epidermidis or its secreted factors significantly reduces S. aureus skin colonization. However, in this work we also demonstrate that this protection is dependent on the integrity of the epithelial barrier and is completely lost during skin inflammation. Using the same models, we further demonstrate that the S. lugdunensis-derived cyclic peptide antibiotic, lugdunin, which was previously shown to inhibit S. aureus epithelial colonization, induces a similar protective effect in human keratinocytes and mouse skin. Additionally, lugdunin can amplify the S. epidermidis-induced effect. Further analysis revealed that, beyond its bactericidal activity, lugdunin also possesses TLR/MyD88-dependent immune-modulatory activities which lead to expression and release of LL-37 and CXCL8/MIP-2 in human keratinocytes and mouse skin as well as to the recruitment of monocytes and neutrophils in mouse skin. Ultimately, synergistic antimicrobial activity in combination with skin-derived AMPs indicates that lugdunin is a multi-functional peptide providing host protection against S. aureus by multiple mechanisms. Since S. aureus frequently colonizes the inflamed skin of AD patients, this work also aimed at understanding how inflammation contributes to the initial skin colonization event with S. aureus. We found that recruited neutrophils in response to tape-stripping enhance S. aureus colonization. These findings were confirmed using an in vitro co-culture model with keratinocytes and neutrophils. Further analysis of the mechanism demonstrated that neutrophil extracellular traps (NETs) influence keratinocytes in a way that favors S. aureus colonization. Finally, we show that S. epidermidis can reduce neutrophil recruitment induced by S. aureus, which might partly explain how microbiota contribute to pathogen skin protection during homeostasis. At the same time this work suggests that during skin inflammation the release of NETs by infiltrating neutrophils superimposes the microbiota-mediated skin protection and might thus result in enhanced S. aureus colonization. In conclusion, this work describes how two members of our skin microbiota, S. epidermidis and S. lugdunensis, can prevent S. aureus skin colonization. Thus, it delineates new ways how commensals protect our skin from pathogens. In addition, the work presented here also illustrates that S. epidermidis contributes to increased S. aureus colonization during skin inflammation and that the recruitment of immune cells can have conflicting effects. Consequently, the signals of the skin microbiome need to be interpreted in the corresponding microenvironment.