Skin programming of inflammatory responses to Staphylococcus aureus is compartmentalised within epidermal keratinocytes differentiation status.

During acute cutaneous inflammation in diseases such as atopic eczema there are alterations in the microbiome as well as histological and ultrastructural changes to the stratified epidermis, but the precise interaction between the keratinocyte proliferation and differentiation status and the skin microbiome has not been fully explored. We hypothesised that the skin microbiome contributes to regulation of keratinocyte differentiation and can modify antimicrobial responses. Therefore, we examined the effect of exposure to topical commensal or pathogenic Staphylococcal challenge on skin models. To further explore the cell types regulating inhibition specifically targeting SA, single-cell DropSeq analysis of unchallenged naïve, SE challenged, and SA challenged epidermis models was undertaken. Transcriptomic analysis distinguished cells from basal, spinous, and granular layers which could then be interrogated individually in relation to model exposure. In contrast to SE, SA specifically induced a sub-population of spinous cells which highly expressed transcripts related to epidermal inflammation and antimicrobial response such as IL36G, IVL, IL1RN, KLK7, PI3, MMP1, S100A7, S100A8, S100A9, SERPINB1, SERPINB2, SERPINB3, and SLPI. Furthermore, SA, but not SE, specifically induced a basal population which highly expressed IL-1-alpha and IL-1-beta. Confluent monolayer primary keratinocyte cultures were used to seed and establish reconstituted human epideris models after 13-15 days of growth within cell culture inserts at the air-liquid interface. Approximate absolute numbers of 10^6 CFU of bacteria were used per model for the challenge protocol. Models were challeged with either S. aureus (ATCC 29213 or NCTC-8325-4), S. epidermidis (ATCC 12228) or S. capitis (ATCC 27840). The challenge protocol consisted of an intial three hour incubation, then treated by PBS washing and further incubation of 21 hours. Subsequently models underwent trypsinisation to prepare for single-cell sequencing by DropSeq. Models were challenged over three batches (0110, 1811, 2011), with n=3-5 models challenged per condition that were pooled to reduced model variation. Per batch, challenge conditions were then multiplexed using cell hashtag antibodies for control (GTCAACTCTTTAGCG), S. aureus (TGATGGCCTATTGGG) and S. epidermidis (TTCCGCCTCTCTTTG).