Prevalent glutamyl-endopeptidases in the commensal skin microbiome have itch-relevant activity

Atopic dermatitis (AD) is frequently accompanied by pruritus, which has predominantly been attributed to skin colonization by Staphylococcus aureus, particularly through cleavage of protease-activated receptor 1 (PAR1) by the glutamyl endopeptidase (GEP) V8 protease. Whether related GEPs from other skin-associated staphylococci contribute to this process remains unclear. Here, we analyzed 273 staphylococcal isolates from skin swabs of 10 AD patients with pruritus, dominated by S. aureus and Staphylococcus epidermidis. Genome mining using a custom hidden Markov model identified 678 candidate GEPs, which were clustered and resolved into five structurally distinct protease architectures. Representative proteases (V8, Esp, SplB, Csp, and Hsp) were expressed and characterized. Esp displayed GEP activity and PAR1 tethered-ligand cleavage comparable to V8, generating noncanonical cleavage products, while Csp cleaved PAR1 with reduced but substantial efficiency. All representative proteases significantly disrupted barrier integrity in an epithelial barrier model. Analysis of isolate genomes and publicly available Staphylococcus genomes showed that V8 and Esp are highly conserved and largely species-restricted, whereas Csp is more broadly distributed across species. These findings identify Esp and Csp as functional GEP virulence factors in S. epidermidis and S. capitis, capable of activating itch signaling and compromising barrier function. Our study suggests that GEP-mediated pruritus and barrier dysfunction in AD may arise not only from pathogens like S. aureus, but also from commensals or opportunistic pathogens such as S. epidermidis and S. capitis. ImportanceStaphylococcus aureus colonization on the skin is closely associated with itch in atopic dermatitis (AD) through a secreted protease that cleaves PAR1 on sensory neurons. However, AD patients can experience pruritus without S. aureus colonization. This raises the question of whether other skin-colonizing staphylococci contribute to this process. We identify glutamyl endopeptidase homologs, including Esp from Staphylococcus epidermidis and Csp from Staphylococcus capitis, that cleave the same itch receptor and disrupt epithelial barrier integrity. These serine proteases are distributed across staphylococcal species that colonize human skin. Therefore, itch and barrier dysfunction in AD may not be restricted to S. aureus but instead arise from the proteolytic activity of multiple staphylococcal species. This functional redundancy means that other staphylococcal species can sustain GEP-driven itch and barrier dysfunction even in the absence of S. aureus, suggesting that therapeutic strategies targeting only S. aureus may overlook these alternative drivers of disease.