Investigating the crosstalk between structural and immune components of human skin to understand pathogen sensing in cutaneous health and inflammation

Abstract

The epidermal environment of local cytokine, immune and microbial events are often mediated by keratinocyte-centric responses and dysregulation can lead to inflammatory skin disease. High dimensional technologies (microarrays and RNA-sequencing) allow interrogation of the global transcriptome in complex multi-cell tissues, but the transcriptomic changes from specific cellular subsets are difficult to isolate from the bulk signal. Computational approaches utilising unsupervised and hypothesis-driven methods resolved bulk skin gene expression into key immune
and keratinocyte-related compartments underscoring inflammatory skin disease. Using an in-silico algorithm called CIBERSORT, bulk skin data were resolved into 19 cell signatures of purified populations from publicly available datasets, including 11 from keratinocytes stimulated with disease-associated cytokines.
Microarray data from whole skin biopsy samples were deconvoluted to observe shifting keratinocyte responses to local TH-specific inflammation. Atopic samples, regardless of biopsy lesion status, showed a TH2 immunophenotype, termed KC2. Psoriatic lesions and chronic atopic
lesions, in contrast, resolved an IL-17 (KC17) induced signature. Atopic dermatitis and psoriasis
treatment responses further validated the provenance of these KC2 and KC17 fractions.
The cutaneous microbiome is also important for maintenance of cutaneous health. Models of stratified keratinocyte-only epidermis were challenged with pathogenic S. aureus or commensal S.
epidermidis. Bulk data from models indicated a strong transcriptomic response to S. epidermis.
Applied as a reference signature for deconvolution from biopsy samples from atopic dermatitis,
correlations of this fraction to KC2 and KC17 fractions, and barrier expression, strongly suggested a
role for an altered commensal microbiome in the pathogenesis of atopic dermatitis.
Keratinocyte responses to Staphylococcal spp. challenge were further investigated at the singlecell level using DropSeq. This revealed highly stratified epidermal responses to challenge. Pathogenchallenged
models indicated an inflammatory spinous population, which could also be resolved from psoriatic lesions and was ameliorated by etanercept treatment. Commensal and pathogen challenge of the model system induced distinct basal population of keratinocytes, compared together and to sterile controls. Basal keratinocytes from commensal-challenged models indicated
a subtle alteration to a remodelling phenotype, which was found to be intensified by pathogenchallenge
in addition to inflammatory MMP and IL-1 alarmin expression.
The potential of this experimental pipeline for modelling keratinocyte transcriptomic responses
coupled with an in-silico deconvolution approach allows for elucidation of key pathogenic
mechanisms and cellular-driven disease processes. This work underscores the importance of keratinocyte-centric responses to the maintenance of cutaneous health and their role during inflammation and dysbiosis.

More information

Identifiers

ORCID for Kalum Clayton: orcid.org/0000-0002-1143-3931

Catalogue record

Export record