Absorption of dietary iron is largely regulated by the liver hormone hepcidin, which is released under conditions of iron overload and inflammation. Although hepcidin-dependent regulation of iron uptake and circulation is well-characterized, recent studies have suggested that the skin may play an important role in iron homeostasis, including transferrin receptor-mediated epidermal iron uptake and direct hepcidin production by keratinocytes. In this study, we characterized direct keratinocyte responses to conditions of high and low iron. We observed potent iron storage capacity by keratinocytes in vitro and in vivo and the effects of iron on epidermal differentiation and gene expression associated with inflammation and barrier function. In mice, systemic iron was observed to be coupled to epidermal iron content. Furthermore, topical inflammation, as opposed to systemic inflammation, resulted in a primary iron-deficiency phenotype associated with low liver hepcidin. These studies suggest a role for keratinocytes and epidermal iron storage as regulators of iron homeostasis with direct contribution by the cutaneous inflammatory state.

Staphylococcus epidermidis is a common microbe on human skin and has beneficial functions in the skin microbiome. However, under conditions of allergic inflammation, the abundance of S. epidermidis increases, establishing potential danger to the epidermis. To understand how this commensal may injure the host, we investigate phenol-soluble modulin (PSM) peptides produced by S. epidermidis that are similar to peptides produced by Staphylococcus aureus. Synthetic S. epidermidis PSMs induce expression of host defense genes and are cytotoxic to human keratinocytes. Deletion mutants of S. epidermidis lacking these gene products support these observations and further show that PSMs require the action of the EcpA bacterial protease to induce inflammation when applied on mouse skin with an intact stratum corneum. The expression of PSMd from S. epidermidis is also found to correlate with disease severity in patients with atopic dermatitis. These observations show how S. epidermidis PSMs can promote skin inflammation.

Our group has challenged the concept that S. epidermidis (SE) is only a beneficial commensal microbe by showing that when skin colonization of SE increases, it promotes skin inflammation through production of the protease EcpA. Analysis of the SE genome has revealed that it encodes 4 genes homologous to the α-type phenol soluble modulin (PSM) toxic peptide made by S. aureus (SA), thus suggesting another mechanism by which SE can harm the skin. Principle component analysis of RNA-Seq data from primary keratinocytes (nHEKs) treated with synthetic SE PSM peptides PSMα, PSMδ, PSMε and δ-toxin (hld) revealed that the transcriptional response of keratinocytes to most SE PSMs was similar to SA PSMα3 but the response to SE PSMα was like the untreated controls. Induction of genes in the IL-17 pathway including CXCL1, CXCL2, CXCL5, CXCL8 (IL8), CCL20, TRAF6, IL-1B, and TNFa were observed and validated by qPCR. Next, isogenic mutants were constructed in SE for individual PSM genes and IL-8 release was measured by ELISA from nHEKs exposed to sterile-filtered supernatants from the wild-type (WT) or mutant isolates. PSMδ and δ-toxin (hld) deletions decreased release of IL-8 by greater than 50% while a double knockout prevented the majority of IL-8 cytokine release from nHEKs above baseline (IL-8 pg/mL: WT=1752+/-428; Δpsmδ=642+/-100; Δpsmε=1413+/-122; Δhld=724+/-163; ΔpsmδΔhld=397+/-44). To elucidate the mechanism for this response we next examined nHEK LDH release and the role of activation of cell surface GPCR receptors. IL-8 production was inhibited when nHEKs were treated with the GPCR inhibitor pertussis toxin prior to SE PSMδ (IL-8 pg/mL: Control=2004+/-453; Pertussis toxin:984+/-81). LDH release also occurred in a dose-dependent manner after exposure to synthetic SE PSMs. These data reveal a novel role for SE in driving inflammation via both cytolytic and GPCR stimulation pathways and provide further evidence that SE acts as a pathogen if enabled to produce PSMs.

Iron is an essential nutrient that functions as a cofactor for cellular enzymatic redox reactions. Hosts and pathogens compete over iron for survival. Previous studies have implicated the skin as a unique site for iron homeostasis as the major contributing source to iron excretion from the body: Primary human keratinocytes store iron as ferritin as a function of exogenous iron addition, with modulation of their differentiation. Wildtype mice subjected to systemic iron bolus or iron chelation demonstrate changes in total epidermal elemental iron content. We hypothesized that epidermal iron stores influence both systemic iron levels and could affect cutaneous pathogen survival. To test this, C57Bl6J mice were treated with one-time systemic iron bolus (1mg/kg intraperitoneal) or treated with the iron chelator desferrioxamine (300 mg/kg intraperitoneal) and then iron stores in the epidermis and microbial survival on the skin was evaluated. Inductively coupled plasma mass spectrometry (ICP-MS) performed on dispase-split mouse skin identified significant changes in epidermal iron content by these interventions (p < 0.0001). After topical challenge with C.albicans, animals subjected to iron overload had decreased C. albicans CFUs (p = 0.0014) consistent with the clinical observation of C.albicans-associated angular cheilitis in patients with iron deficiency. However, no difference was observed in S. aureus survival on the skin after systemic iron manipulation. These findings suggest that epidermal turnover may be a regulated mechanism of iron excretion that matches epidermal behavior to systemic iron levels, with physiologic implications in some cutaneous infections.

Human skin is home to a variety of commensal bacteria, including many species of coagulase-negative staphylococci (CoNS). While it is well established that the microbiota as a whole maintains skin homeostasis and excludes pathogens (i.e., colonization resistance), relatively little is known about the unique contributions of individual CoNS species to these interactions. Staphylococcus hominis is the second most frequently isolated CoNS from healthy skin, and there is emerging evidence to suggest that it may play an important role in excluding pathogens, including Staphylococcus aureus , from colonizing or infecting the skin.

The bacteria belonging to the genus Staphylococcus are a group of organisms that readily inhabit human skin and the upper respiratory tract. Staphylococcus aureus (SA) is a leading cause of soft tissue infections and bacteremia (Miller and Cho, 2011Miller L.S. Cho J.S. Immunity against Staphylococcus aureus cutaneous infections.Nat Rev Immunol. 2011; 11: 505-518Crossref PubMed Scopus (270) Google Scholar) but is closely related to many other species of Staphylococcus that are skin commensals and rarely cause disease. These species densely populate follicular structures that may serve as a reservoir for long-term colonization (Nakatsuji et al., 2013Nakatsuji T. Chiang H.I. Jiang S.B. Nagarajan H. Zengler K. Gallo R.L. The microbiome extends to subepidermal compartments of normal skin.Nat Commun. 2013; 4: 1431Crossref PubMed Scopus (283) Google Scholar). Furthermore, there is growing evidence that various other microorganisms live in hair follicles (HFs) and interact with skin immune cells (Chen et al., 2018Chen Y.E. Fischbach M.A. Belkaid Y. Skin microbiota-host interactions.Nature. 2018; 553: 427-436Crossref PubMed Scopus (265) Google Scholar). For example, Cutibacterium acnes produces short-chain fatty acids by fermenting the lipids in the sebum, and these substances have an influence on host immune function (Sanford et al., 2019Sanford J.A. O'Neill A.M. Zouboulis C.C. Gallo R.L. Short-chain fatty acids from Cutibacterium acnes activate both a canonical and epigenetic inflammatory response in human sebocytes.J Immunol. 2019; 202: 1767-1776Crossref PubMed Scopus (33) Google Scholar). Thus, it is important to further understand the mechanisms by which both pathogenic and commensal microbes establish colonization of the HF. In this study, we sought to understand how Staphylococcus can enter the HF. Because sebum is secreted from sebaceous glands and fills in the infundibulum of HFs, we hypothesized that sebum functions as a hydrophobic barrier, whereas lipases may enable bacteria to enter the HF independently of the penetration of the stratum corneum. We focused on triacylglycerol lipases that degrade triglycerides, which are the main components (60%) of sebum (Picardo et al., 2009Picardo M. Ottaviani M. Camera E. Mastrofrancesco A. Sebaceous gland lipids.Dermatoendocrinol. 2009; 1: 68-71Crossref PubMed Google Scholar), and studied the response to the deletion mutants of the SA known lipases. SA produces two secreted lipases termed gehA (SAL-1) and gehB (SAL-2), and the expressions of these genes are under the control of the accessory gene regulator (agr) quorum sensing system (Horswill and Gordon, 2020Horswill A.R. Gordon C.P. Structure-activity relationship studies of small molecule modulators of the staphylococcal accessory gene regulator.J Med Chem. 2020; 63: 2705-2730Crossref PubMed Scopus (19) Google Scholar). We compared the capacity of a methicillin-resistant SA USA300 LAC wild-type (SA WT) parent strain, a double lipase mutant SA (SA Δlipases) strain, a SA agr mutant strain, and a mutant SA strain of all 10 known secreted proteases (proteases mutant SA [SA Δproteases]) to induce skin damage and inflammation. Topically applied SA WT induced inflammation, erythema, and crust within 48 hours; SA Δlipases and SA Δproteases strains induced less inflammation, and the agr mutant was the least capable of promoting inflammation (Figure 1a). Measurement of Il6 mRNA and transepidermal water loss of the skin revealed that murine Il6 mRNA in the whole skin was decreased by 56% in the SA Δlipases strain compared with that in the SA WT strain control (Figure 1b). Transepidermal water loss was decreased by 47% in the SA Δlipases strain compared with that in the SA WT strain (Figure 1c). Overall, these data reveal how SA lipases play a role in skin inflammation and damage and revealed that these likely cooperate with other previously discovered agr-regulated skin-damaging factors such as SA proteases (Kolar et al., 2013Kolar S.L. Ibarra J.A. Rivera F.E. Mootz J.M. Davenport J.E. Stevens S.M. et al.Extracellular proteases are key mediators of Staphylococcus aureus virulence via the global modulation of virulence-determinant stability.Microbiologyopen. 2013; 2: 18-34Crossref PubMed Scopus (111) Google Scholar). Next, to determine whether SA lipases allow for penetration of microbes into the HF, the SA WT and mutant strains were applied topically to shaved back skin at 1 × 106 colony-forming unit/cm2 for 2 hours. The 16S ribosomal RNA abundance was quantified in sequential 20 μm horizontal sections of frozen skin. This analysis revealed that the SA Δlipases strain as well as the SA agr mutant and SA Δproteases strains all lacked the ability to penetrate past 100 μm of the skin surface (Figure 1d). This depth of penetration corresponds to the depth of the infundibulum that is typically within approximately 100 μm of the surface. Furthermore, DNA for the SA WT and SA Δproteases strains appeared greater within the 60–100 μm depth than for the SA Δlipases and SA agr mutant strains. This suggested that SA lipases play a role in the initial penetration of bacteria into the lipid-rich infundibulum. To confirm this, immunostaining of murine skin sections was done with an antibody specific to SA, and it revealed that only SA WT and SA Δproteases strains were frequently observed to infiltrate into the infundibulum, but mutants lacking lipases or an active agr system (lacking the capacity to secrete lipases and proteases) could not (Figure 1e). Furthermore, by counting individual staining within multiple follicles and applying the agr and total protease mutant compared with specific SA mutants to either lipase gehA or lipase gehB, we quantified what fraction of total bacterial staining could be seen within the infundibulum. This analysis revealed that SAL-2was the primary SA lipase driving initial HF penetration (Figure 1f and g). Overall, these data suggest that the specific SA lipase gene, gehB, is essential for penetration into the upper HF of mice. Lipases are widely expressed throughout the genus Staphylococcus. In the case of SA, SAL-1 (encoded by gehA) primarily reacts with short-chain glycerides, whereas SAL-2 (encoded by gehB) hydrolyzes both short-chain and long-chain triglycerides (Cadieux et al., 2014Cadieux B. Vijayakumaran V. Bernards M.A. McGavin M.J. Heinrichs D.E. Role of lipase from community-associated methicillin-resistant Staphylococcus aureus strain USA300 in hydrolyzing triglycerides into growth-inhibitory free fatty acids.J Bacteriol. 2014; 196: 4044-4056Crossref PubMed Scopus (50) Google Scholar). Two equivalent lipases are reported in S. epidermidis: SEL-1 (encoded by gehC) and SEL-2 (encoded by gehD) (Longshaw et al., 2000Longshaw C.M. Farrell A.M. Wright J.D. Holland K.T. Identification of a second lipase gene, gehD, in Staphylococcus epidermidis: comparison of sequence with those of other staphylococcal lipases.Microbiology (Reading). 2000; 146: 1419-1427Crossref PubMed Scopus (56) Google Scholar). Other staphylococcal species are known to secrete lipases, and some of their genomes are sequenced. To reveal the evolutionary differences of major lipases among the genus Staphylococcus, we performed a database search and analyzed their similarity using National Center for Biotechnology Information Basic Local Alignment Search Tool (www.ncbi.nlm.nih.gov/BLAST), followed by the analysis conducted in Molecular Evolutionary Genetics Analysis X software (www.megasoftware.net). As expected, our data suggested that lipases are highly preserved among those species and that all the staphylococcal lipases are produced as pre-proenzymes with a signal peptide in the preregion and are secreted as proenzymes needing a specific cleavage for mature configuration. The amino acid positions of the catalytic triad (serine-histidine-aspartate motif) are quite similar among the mature lipases (Figure 2a). Generally, lipases are members of a large group of enzymes possessing the α and β hydrolase folds, with the preserved arrangement of the catalytic residue (histidine-serine-aspartate or cysteine) located in a tight loop after the β5 strand (Ollis et al., 1992Ollis D.L. Cheah E. Cygler M. Dijkstra B. Frolow F. Franken S.M. et al.The alpha/beta hydrolase fold.Protein Eng. 1992; 5: 197-211Crossref PubMed Scopus (1821) Google Scholar). In addition, the calculation indicated that on the basis of the similarity, they are categorized into two groups: one including SAL-1 and the other including SAL-2 (Figure 2b), indicating that all staphylococcal species encode two lipases. This high identity of lipases in the genus Staphylococcus implies important roles in metabolism and adaptation to the environment (Nguyen et al., 2018Nguyen M.T. Luqman A. Bitschar K. Hertlein T. Dick J. Ohlsen K. et al.Staphylococcal (phospho)lipases promote biofilm formation and host cell invasion.Int J Med Microbiol. 2018; 308: 653-663Crossref PubMed Scopus (23) Google Scholar) and suggests that our observations in SA may apply to other species of Staphylococcus in the HF. These observations add important new insight into the previously known contributions of SA proteases to skin damage (Nakatsuji et al., 2016Nakatsuji T. Chen T.H. Two A.M. Chun K.A. Narala S. Geha R.S. et al.Staphylococcus aureus exploits epidermal barrier defects in atopic dermatitis to trigger cytokine expression.J Invest Dermatol. 2016; 136: 2192-2200Abstract Full Text Full Text PDF PubMed Scopus (183) Google Scholar; Williams et al., 2019Williams M.R. Costa S.K. Zaramela L.S. Khalil S. Todd D.A. Winter H.L. et al.Quorum sensing between bacterial species on the skin protects against epidermal injury in atopic dermatitis.Sci Transl Med. 2019; 11eaat8329Crossref PubMed Scopus (89) Google Scholar) and suggest that the expression of lipases by Staphylococcus enables bacteria within this genus to establish residence in the HF. No datasets were generated during this study. All mouse procedures were approved by the University of California San Diego Institutional Animal Care and Use Program (Protocol Number: S09074). Kouki Nakamura: http://orcid.org/0000-0002-2296-6004 Michael R. Williams: http://orcid.org/0000-0002-1523-6353 Jakub M. Kwiecinski: http://orcid.org/0000-0001-9472-2896 Alexander R. Horswill: http://orcid.org/0000-0002-5568-0096 Richard L. Gallo: http://orcid.org/0000-0002-1401-7861 RLG is a cofounder, scientific advisor, and consultant; has equity in MatriSys Biosciences; and is a consultant, receives income from, and has equity in Sente Inc. The remaining authors state no conflict of interest. KN was supported by grants from Uehara Memorial Foundation , Japan. RLG, MRW, and ARH are supported by the National Institute of Health grant R01AI153185. MRW is supported by the National Institute of General Medical Sciences training grant 5K12GM068524-18. RLG is supported by the National Institute of Health grants R01AR074302, R01AR076082, R37AI052453, and U01AI52038. Conceptualization: KN, MRW, RLG; Formal Analysis: KN, MRW; Investigation: KN, MRW, JMK; Supervision: RLG; Writing – Original Draft Preparation: KN, MRW, RLG; Writing – Review and Editing: KN, MRW, JMK, ARH, RLG

Bacterial biofilms are a major factor in delayed wound healing. Many staphylococcal species can form biofilms that have the potential to harm the host. DNA sequencing and qPCR of 16S RNA from bacteria isolated from human hair follicles by laser capture microdissection has shown that Staphylococci are highly abundant in the normal human follicles and co-exist with C. acnes. However, despite the high density of Staphylococci in follicles, biofilms seldom occur on healthy skin. We hypothesized that interactions between species in the commensal skin microbiome may regulate biofilm formation. To test this, two different strains of S. epidermidis and S. aureus were exposed to the conditioned media from several species of Cutibacteria including 2 ATCC defined strains of C. acnes. Biofilm formation was observed by crystal violet staining of S. epidermidis and S. aureus grown on plastic, and all Cutibacteria species and strains tested inhibited biofilm formation by 80% (P<0.002) at concentrations that were not bacteriostatic. To identify the mechanism of action, conditioned media from C. acnes was subjected to biochemical analysis. The active product was stable at 100°C, resistant to proteinase K and lysozyme, and volatile after lyophilization. These properties suggested that the active agents could be short chain fatty acids (SCFAs). C. acnes produces several SCFAs including acetate, propionate, isobutyrate, and isovalerate. Exposure of S. epidermis and S. aureus to pure SCFAs at the concentrations made by C. acnes confirmed that acetate, propionate, isobutyrate, and isovalerate inhibited Staphylococci biofilm formation. Similar to the C. acnes conditioned media, these concentrations of SCFAs were not antibacterial. Our data suggest that a beneficial consequence of diversity in the skin microbiome is that SCFAs from Cutibacteria inhibit biofilm formation by Staphylococci. Recognition of this function can be useful to wound care where low microbial diversity may permit biofilm development and delayed wound healing.

Inflammatory bowel diseases (IBD) are associated with several skin inflammatory diseases but the mechanism responsible for communication between organs is unknown. Local immune activation occurs in part by the action of induced hyaluronidase in the extracellular matrix (ECM) and subsequent recognition of soluble HA fragments. We hypothesized that such HA fragments may also act to enable organ crosstalk between skin and gut. To test this, the intestine was examined in mice expressing hyaluronidase in the skin (K14/Hyal1 mice). K14/Hyal1 mice do not show spontaneous skin inflammation. These mice were then compared with littermate controls or mice with incisional skin wounds (Wd) that induce the endogenous dermal hyaluronidase Cemip. Both groups showed HA digestion in the dermis. Remarkably, all skin specific interventions enhanced DSS-induced inflammation in the colon as seen by greater weight loss (p<0.0001) and lower survival rates (Control 100% survived, K14/Hyal1: 20%, Wd; 80%). Transcriptional profiling by single cell RNA Seq revealed that expression of hyaluronidase in skin promoted large changes in the abundance of colon fibroblast subsets; cluster 5 of 9 increased from 1.21% to 43.2%, and clusters 0, 2 and 7 decreased from 29.8 to 4.16%, 18.9 to 0% and 4.5 to 0%, respectively). Genes altered in these subsets were validated by whole tissue RNA Seq and qRT-PCR, and after DSS challenge showed activation of genes related to reactive adipogenesis in colon. FACS analysis also showed increase in PDGFRa and Thy-1 positive cell populations (K14/Hyal1: p=0.0001, Wd; p=0.07) as well as a shift in resident colon RORgt Tregs (vs K14/Hyal1: p=0.009, Wd; p=0.013). The fecal microbiome also significantly changed in K14-Hyal1 mice compared to cohoused littermate controls(vs K14/Hyal1: p=0.05). Taken together, these data show how disruption of skin ECM HA may explain associations between skin and gut disorders.

BACKGROUND:Staphylococcus aureus and Staphylococcus epidermidis are the most abundant bacteria found on the skin of patients with atopic dermatitis (AD). S aureus is known to exacerbate AD, whereas S epidermidis has been considered a beneficial commensal organism. OBJECTIVE:In this study, we hypothesized that S epidermidis could promote skin damage in AD by the production of a protease that damages the epidermal barrier. METHODS:The protease activity of S epidermidis isolates was compared with that of other staphylococcal species. The capacity of S epidermidis to degrade the barrier and induce inflammation was examined by using human keratinocyte tissue culture and mouse models. Skin swabs from atopic and healthy adult subjects were analyzed for the presence of S epidermidis genomic DNA and mRNA. RESULTS:S epidermidis strains were observed to produce strong cysteine protease activity when grown at high density. The enzyme responsible for this activity was identified as EcpA, a cysteine protease under quorum sensing control. EcpA was shown to degrade desmoglein-1 and LL-37 in vitro, disrupt the physical barrier, and induce skin inflammation in mice. The abundance of S epidermidis and expression of ecpA mRNA were increased on the skin of some patients with AD, and this correlated with disease severity. Another commensal skin bacterial species, Staphylococcus hominis, can inhibit EcpA production by S epidermidis. CONCLUSION:S epidermidis has commonly been regarded as a beneficial skin microbe, whereas S aureus has been considered deleterious. This study suggests that the overabundance of S epidermidis found on some atopic patients can act similarly to S aureus and damage the skin by expression of a cysteine protease.

The microbiome represents a vast resource for drug discovery, as its members engage in constant conflict to outcompete one another by deploying diverse strategies for survival. Cutibacterium acnes is one of the most common bacterial species on human skin and can promote the common disease acne vulgaris. By employing a combined strategy of functional screening, genetics, and proteomics we discovered a strain of Staphylococcus capitis (S. capitis E12) that selectively inhibited growth of C. acnes with potency greater than antibiotics commonly used in the treatment of acne. Antimicrobial peptides secreted from S. capitis E12 were identified as four distinct phenol-soluble modulins acting synergistically. These peptides were not toxic to human keratinocytes and the S. capitis extract did not kill other commensal skin bacteria but was effective against C. acnes on pig skin and on mice. Overall, these data show how a member of the human skin microbiome can be useful as a biotherapy for acne vulgaris.

Netherton syndrome (NS) is a monogenic skin disease resulting from loss of function of lymphoepithelial Kazal-type-related protease inhibitor (LEKTI-1). In this study we examine if bacteria residing on the skin are influenced by the loss of LEKTI-1 and if interaction between this human gene and resident bacteria contributes to skin disease. Shotgun sequencing of the skin microbiome demonstrates that lesional skin of NS subjects is dominated by Staphylococcus aureus (S. aureus) and Staphylococcus epidermidis (S. epidermidis). Isolates of either species from NS subjects are able to induce skin inflammation and barrier damage on mice. These microbes promote skin inflammation in the setting of LEKTI-1 deficiency due to excess proteolytic activity promoted by S. aureus phenol-soluble modulin α as well as increased bacterial proteases staphopain A and B from S. aureus or EcpA from S. epidermidis. These findings demonstrate the critical need for maintaining homeostasis of host and microbial proteases to prevent a human skin disease.

Biofilm formation by bacterial pathogens is associated with numerous human diseases and can confer resistance to both antibiotics and host defenses. Many strains of Staphylococcus epidermidis are capable of forming biofilms and are important human pathogens. Since S. epidermidis coexists with abundant Cutibacteria acnes on healthy human skin and does not typically form a biofilm in this environment, we hypothesized that C. acnes may influence biofilm formation of S. epidermidis. Culture supernatants from C. acnes and other species of Cutibacteria inhibited S. epidermidis but did not inhibit biofilms by Pseudomonas aeruginosa or Bacillus subtilis, and inhibited biofilms by S. aureus to a lesser extent. Biofilm inhibitory activity exhibited chemical properties of short chain fatty acids known to be produced from C. acnes. The addition of the pure short chain fatty acids propionic, isobutyric or isovaleric acid to S. epidermidis inhibited biofilm formation and, similarly to C. acnes supernatant, reduced polysaccharide synthesis by S. epidermidis. Both short chain fatty acids and C. acnes culture supernatant also increased sensitivity of S. epidermidis to antibiotic killing under biofilm-forming conditions. These observations suggest the presence of C. acnes in a diverse microbial community with S. epidermidis can be beneficial to the host and demonstrates that short chain fatty acids may be useful to limit formation of a biofilm by S. epidermidis.

Quorum sensing peptides produced by commensal skin staphylococci protect against skin injury and inflammation induced by Staphylococcus aureus toxins.

Netherton syndrome (NS) results from loss of function mutations in SPINK5, a gene encoding the serine protease inhibitor LEKTI. We recently showed that some bacteria in the skin microbiome induce keratinocyte serine proteases. In this study we hypothesized that a failure of LEKTI to inhibit keratinocyte protease activity induced by bacteria will contribute to the pathogenesis of NS. Shotgun metagenomic sequencing analysis on samples from 10 NS patients and 8 healthy controls revealed all NS patients had a unique skin dysbiosis including increased relative abundance of staphylococci that encode genes that promote protease activity. Furthermore, culture of live bacteria from each subject at multiple time points showed increased S. aureus on both non-lesional and lesional skin compared to healthy controls that had no growth (2.68e3CFU/cm2, 1.35e5CFU/cm2, p<0.001). mRNA expression of PSMα, a S. aureus toxin that induces keratinocyte serine protease activity, was increased in non-lesional and lesional samples from NS compared to healthy controls (1.59e3-fold, 2.14e5-fold, p<0.0001). Expression strongly correlated with disease severity. Direct measurement of PSMα mRNA abundance and application of these bacteria to normal human keratinocytes showed S. aureus from NS subjects could both produce proteases and induce keratinocyte serine protease activity. Finally, SPINK5 siRNA knockdown in keratinocytes showed that PSMα3 synthetic peptide induced serine protease activity independently of SPINK5 expression. These findings suggest a role for the microbiome in triggering the pathogenesis of NS, potentially explaining flares in disease and the transient beneficial response of patients to antibiotic treatment. Overall, NS is shown here to be a human genetic disease promoted by inter-kingdom interactions between host and bacterial genes from the skin microbiome. Recognition of the importance of this mechanism advances understanding and therapeutic strategies for this disorder.

S. aureus (SA) and S. epidermidis (SE) are the most abundant bacteria found on the lesional skin of patients with atopic dermatitis (AD). SA is known to exacerbate AD but SE is commonly considered a beneficial commensal organism. In this study we hypothesized that some SE strains may cause skin damage by mechanisms similar to SA, such as through the production of proteases. Coagulase-negative staphylococcus (CoNS) strains from healthy or atopic human skin were screened for the proteolytic activity. Among the tested CoNS, multiple strains of SE were the most proteolytically active and the level of activity varied greatly from one SE strain to another. Using diverse protease inhibitors, we showed that SE proteolytic activity mainly came from a cysteine protease. This protease was identified to be EcpA since the protease activity of SE1457 was eliminated by targeted deletion of the gene (ΔEcpA). Next, to test the effect of EcpA on skin, we applied 106 CFU/cm2 of the WT or ΔEcpA strains to murine skin for 48h. While the ΔEcpA strain did not significantly induce skin damage, the WT strain disrupted the epidermal barrier (increase of the transepidermal water loss: 48.7 vs 13.2 g/h/m2, p<0.001) and induced skin inflammation (epidermal thickening, dermal immune cell infiltration, increase of IL6 mRNA fold change: 68.3 vs 2.1, p<0.01). Moreover, only the WT strain was able to penetrate the skin as shown by gram staining. To finish, we showed that some CoNS, in particular some S. hominis strains, were able to inhibit EcpA production by inactivating the SE Agr quorum sensing system. Overall, these findings suggest that some SE strains can exacerbate AD through the secretion of EcpA that induces epidermal barrier disruption. Furthermore, others CoNS strains, such as S. hominis, have a protective effect against EcpA production.

Cutibacterium acnes (C. acnes, formerly Propionibacterium acnes), is normally well tolerated by human skin. Different strain-types of C. acnes are enriched in acne, and it has been hypothesized that strain-types may promote disease. To study the host immune response against strain-types and the genetic basis for such differences, we generated a library of >1000 C. acnes isolates from swabs of lesional and non-lesional sites of acne patients and healthy volunteers. C. acnes strain-level identity was determined by high resolution sequence typing and the inflammatory potential of each isolate was assessed by high-throughput screening of IL-8 secretion from SEB-1 sebocytes and normal human keratinocytes (NHEK) exposed to sterile bacterial supernatants. Consistent with prior findings, a higher frequency of C. acnes phylotype IA1 strains were recovered from acne lesional sites (specifically C1 and C2 sequence types (ST)). However, non-lesional sites from acne patients were found to have K1 and K2 STs, similar to healthy skin. Furthermore, and contrary to predictions, many of the K1/K2 STs from healthy skin and non-lesional acne promoted greater IL-8 production compared to lesional C1/C2 STs (p=0.012). Moreover, clonal isolates from the same ST promoted different cytokine responses, with high and low cytokine-inducing strains identified within K1, C1, C2 and H1 STs. The capacity of identical STs to promote different inflammatory responses was validated by intradermal injection of sterile supernatants into mice and subsequent qRT-PCR for IL-8, CXCL1 and IL-1β. The functional differences within these strains could not be attributed to protease or lipase activity, short chain fatty acid or porphyrin production. Instead, whole genome sequencing revealed the presence of a linear plasmid only in the pro-inflammatory strains. Several putative virulence genes are present in this plasmid. We conclude that a virulence plasmid is most highly associated with the proinflammatory activity of C. acnes.

Colonization of the skin of patients by Staphylococcus aureus is considered a risk for skin infection and an exacerbating factor in atopic dermatitis.1Williams M.R. Nakatsuji T. Gallo R.L. Staphylococcus aureus: master manipulator of the skin.Cell Host Microbe. 2017; 22: 579-581Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar Because of the negative effects associated with S aureus colonization, clinicians commonly seek methods to eradicate colonization by S aureus. A diluted bleach solution (sodium hypochlorite [NaOCl] 0.005%) is often used for this purpose, with conflicting reports of its efficacy to decrease the severity of inflammation or eradicate S aureus on the skin.2Fritz S.A. Camins B.C. Eisenstein K.A. Fritz J.M. Epplin E.K. Burnham C.A. et al.Effectiveness of measures to eradicate Staphylococcus aureus carriage in patients with community-associated skin and soft-tissue infections: a randomized trial.Infect Control Hosp Epidemiol. 2011; 32: 872-880Crossref PubMed Scopus (104) Google Scholar, 3Huang J.T. Abrams M. Tlougan B. Rademaker A. Paller A.S. Treatment of Staphylococcus aureus colonization in atopic dermatitis decreases disease severity.Pediatrics. 2009; 123: e808-e814Crossref PubMed Scopus (374) Google Scholar, 4Chopra R. Vakharia P.P. Sacotte R. Silverberg J.I. Efficacy of bleach baths in reducing severity of atopic dermatitis: a systematic review and meta-analysis.Ann Allergy Asthma Immunol. 2017; 119: 435-440Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar In this study we sought to directly test the assumption that NaOCl at concentrations recommended for bleach baths is an effective antibacterial agent. Three laboratory strains of S aureus (USA300, Newman, and Sanger 252) were initially selected for this study. Bacterial colony growth was initiated by plating 2 × 103 colony-forming units (CFUs) of each strain on 100-cm2Fritz S.A. Camins B.C. Eisenstein K.A. Fritz J.M. Epplin E.K. Burnham C.A. et al.Effectiveness of measures to eradicate Staphylococcus aureus carriage in patients with community-associated skin and soft-tissue infections: a randomized trial.Infect Control Hosp Epidemiol. 2011; 32: 872-880Crossref PubMed Scopus (104) Google Scholar Tryptic Soy Broth (TSB) agar plates and then submerging each plate in various concentrations of bleach for 15 minutes at 37°C to test survival on a nutrient-rich surface. Twenty-four hours after this treatment, direct colony counting showed no significant difference in the survival of S aureus exposed to between 0% and 0.01% NaOCl in water (Fig 1, A). A bactericidal effect of dilute bleach against these strains of S aureus was only evident at concentrations of greater than 0.03%, a concentration of bleach that is cytotoxic to human cells and much greater than should be used clinically. The unexpected observation that the concentrations of NaOCl used in bleach baths were not antibacterial against S aureus prompted us to further explore the role of other variables on bacterial survival in defined laboratory culture conditions. Two Staphylococcus epidermidis strains (1475 and ATCC12228) representing another abundant bacterial species found on the skin that were also biofilm-forming (1457) and non–biofilm-forming (ATCC12228) strains were next tested on agar, as done for S aureus. These S epidermidis strains were also not killed by clinically relevant concentrations of NaOCl in water (Fig 1, B). Next, to test whether culture system or source of household bleach influenced these results, S aureus USA300 was grown in TSB broth at 37°C for 24 hours. The source of household bleach had no effect. Similar to growth on agar, bacterial survival was not inhibited at the clinically used concentration (0.005%) of NaOCl (Fig 1, C). Furthermore, because the bacterial growth phase can determine sensitivity to antibiotic agents, with bacteria in a growth phase (log-phase growth) often showing greater sensitivity than stationary phase bacteria,5Eng R.H. Padberg F.T. Smith S.M. Tan E.N. Cherubin C.E. Bactericidal effects of antibiotics on slowly growing and nongrowing bacteria.Antimicrob Agents Chemother. 1991; 35: 1824-1828Crossref PubMed Scopus (266) Google Scholar we also tested the sensitivity of S aureus USA300 in log-phase growth compared with bacteria at the stationary phase. No difference in sensitivity to bleach was observed under these conditions (Fig 1, D). Taken together, we conclude that the concentration of NaOCl recommended for clinical use in bleach baths does not inhibit the survival or growth of S aureus or S epidermidis under laboratory conditions. S aureus growing on agar or in nutrient-rich broth does not accurately model conditions on the skin. The epidermis has a complex 3-dimensional structure composed of skin folds and skin appendages, such as sebaceous glands, eccrine glands, and hair follicles. The composition of the epidermal surface can also influence the capacity of bleach baths to act as antimicrobial agents. To examine this, 1 × 106 CFUs of S aureus USA300 were applied to explants of pig skin for 15 minutes at room temperature, and the skin was then submerged in a range of NaOCl concentrations for 15 minutes to simulate immersion in a bleach bath. Immediately after this treatment, surviving CFUs were measured. Similar to the results in defined cultures, 0.005% NaOCl had no significant bactericidal effect on S aureus compared with water alone (Fig 1, E). Therefore these results suggest that a bleach bath has no antibacterial action against S aureus on skin. In our final experiment, we assessed whether NaOCl might have a beneficial therapeutic effect against S aureus by influencing expression of virulence functions of bacteria rather than directly killing them. The accessory gene regulator (agr) quorum–sensing system plays a central role in regulation of S aureus virulence by controlling the expression of toxins that can cause epidermal damage and skin inflammation.6Nakagawa S. Matsumoto M. Katayama Y. Oguma R. Wakabayashi S. Nygaard T. et al.Staphylococcus aureus virulent PSMalpha peptides induce keratinocyte alarmin release to orchestrate IL-17-dependent skin inflammation.Cell Host Microbe. 2017; 22: 667-677.e5Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 7Liu H. Archer N.K. Dillen C.A. Wang Y. Ashbaugh A.G. Ortines R.V. et al.Staphylococcus aureus epicutaneous exposure drives skin inflammation via IL-36-mediated T cell responses.Cell Host Microbe. 2017; 22: 653-666.e5Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar To test the action of NaOCl on agr activity, an agr–yellow fluorescent protein reporter strain of S aureus was examined during exposure to bleach for 24 hours in TSB at 37°C. A bleach bath solution of 0.005% showed no significant effect on agr activity compared with water (Fig 1, F). These results show that the S aureus agr quorum–sensing system is also not inhibited during bleach bath treatment. Bleach baths have been reported by clinicians and patients to be associated with improvement of inflammation in patients with atopic dermatitis3Huang J.T. Abrams M. Tlougan B. Rademaker A. Paller A.S. Treatment of Staphylococcus aureus colonization in atopic dermatitis decreases disease severity.Pediatrics. 2009; 123: e808-e814Crossref PubMed Scopus (374) Google Scholar and reported to reduce colonization that could result in deep tissue infections.2Fritz S.A. Camins B.C. Eisenstein K.A. Fritz J.M. Epplin E.K. Burnham C.A. et al.Effectiveness of measures to eradicate Staphylococcus aureus carriage in patients with community-associated skin and soft-tissue infections: a randomized trial.Infect Control Hosp Epidemiol. 2011; 32: 872-880Crossref PubMed Scopus (104) Google Scholar It has been a common assumption that the recommendation of inclusion of one-quarter to one-half cup of 6% household bleach in a bathtub full of water (40 gallons) is an effective method to reduce bacterial load on the skin and that clinical effects might be related to this presumed antibacterial activity. However, our study clearly demonstrates that bleach baths have no direct bactericidal activity against S aureus or S epidermidis. To obtain an antibacterial effect, as previously reported,8Eriksson S. van der Plas M.J.A. Morgelin M. Sonesson A. Antibacterial and antibiofilm effects of sodium hypochlorite against Staphylococcus aureus isolates derived from patients with atopic dermatitis.Br J Dermatol. 2017; 177: 513-521Crossref PubMed Scopus (35) Google Scholar bleach concentrations must be at least 0.03%, which is greater than is safe for the skin or that has been proposed for clinical use. Therefore a benefit from use of a bleach bath cannot be attributed to direct killing or inhibition of S aureus. These observations are not intended to directly support or refute the potential benefits of bleach bath therapy but only to dispel the false conclusion that it is a form of antimicrobial therapy. Indeed, because a recent meta-analysis concluded that water baths alone can significantly decrease the severity of atopic dermatitis,4Chopra R. Vakharia P.P. Sacotte R. Silverberg J.I. Efficacy of bleach baths in reducing severity of atopic dermatitis: a systematic review and meta-analysis.Ann Allergy Asthma Immunol. 2017; 119: 435-440Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar beneficial effects can be due to effects other than those acting directly on the skin microbiome. With improved disease, the bacterial burden might indirectly decrease because of enhanced innate antimicrobial functions. The laboratory strains tested here were selected as reflective of clinical S aureus strains on atopic dermatitis but might not represent the sensitivity of all strains. It is also conceivable that bleach baths might be directly anti-inflammatory on the skin. A previous study has reported that NaOCl can oxidize cysteine residues of the inhibitor of nuclear factor κB, inhibit nuclear factor κB activity, and improve skin disease in mice.9Leung T.H. Zhang L.F. Wang J. Ning S. Knox S.J. Kim S.K. Topical hypochlorite ameliorates NF-kappaB-mediated skin diseases in mice.J Clin Invest. 2013; 123: 5361-5370Crossref PubMed Scopus (76) Google Scholar However, the observations of our study show that a bleach bath does not directly inhibit S aureus. This information should be considered in the interpretation of future studies of topical bleach baths. CorrigendaJournal of Allergy and Clinical ImmunologyVol. 144Issue 5PreviewWith regard to the article in the May 2019 issue entitled “Dilute bleach baths used for treatment of atopic dermatitis are not antimicrobial in vitro” (J Allergy Clin Immunol 2019;143:1946-8), 2 errors have been brought to the Editors' attention. The authors wish to note that a typo was present in the grant acknowledgement. The grant number should be AR06781, instead of AR064781. Second, the disclosure statement for author R. L. Gallo should be modified to read: R. L. Gallo is a co-founder, scientific advisor, consultant, and has equity in MatriSys Biosciences and is a consultant, receives income, and has equity in Sente. Full-Text PDF

Epidermal proteases regulate the epidermal barrier, but abnormal proteolytic activity is also a contributor to disease. In this study we examined the potential contribution of proteases from commensal skin bacteria in maintaining or disrupting epidermal homeostasis. A library of coagulase-negative staphylococcus (CoNS) strains from healthy human skin and subjects with atopic dermatitis were screened for the proteolytic activity against collagen and elastin substrates. Among the CoNS that we tested, multiple strains of Staphylococcus epidermidis (SE) were the most proteolytically active species although the level of activity varied greatly from one SE strain to another. Using diverse protease inhibitors, we showed that SE proteolytic activity mainly came from a cysteine protease. This protease was identified to be EcpA since the protease activity of SE1457 in culture was eliminated by targeted deletion (ΔEcpA). Next, to test the effect of EcpA on skin barrier function, we applied 106 CFU/cm2 of the WT or the ΔEcpA mutant to murine skin for 48h. The WT strain induced skin inflammation and disruption of the epidermal barrier as shown by an increase of the transepidermal water loss (TEWL), while treatment with the ΔEcpA strain did not (TEWL: 70.45±14.56 g/h/m2 with the WT strain versus 13.88±8.44 g/h/m2 with the ΔEcpA strain, p<0.001, n=4). Similarly, treatment of mouse skin with a protein extract from the WT strain increased TEWL compared to the protein extract from ΔEcpA (30.84±8.28 g/h/m2 versus 9.17±3.19 g/h/m2, p<0.005, n=4). Overall these findings suggest that EcpA is a potent enzyme capable of altering the skin barrier. Ongoing investigations are evaluating if EcpA allows SE to penetrate the skin and can be considered as a virulence factor in specific contexts. The involvement of EcpA in the pathogenesis of some skin diseases with epidermal barrier defects such as atopic dermatitis, netherthon syndrome or rosacea is being explored.

Iron is an essential nutrient for all living organisms and functions as a cofactor for cellular enzymatic processes. Multiple organ systems are affected by inappropriate levels of circulating iron levels, including the bone marrow, central nervous system, and skin. Cutaneous conditions associated with high and low iron include telogen effluvium, angular cheilitis, koilonychia, ichtyosis-like changes, atrophy, and changes in pigmentation. Iron homeostasis mechanisms exist to maintain total body iron stores by coupling epithelial iron losses through the skin and gut to gut absorption of dietary iron. Further, total and circulating iron stores are tightly regulated by modulation of gut iron absorption, splenic recycling of erythrocyte iron, and release of body iron stores. However, there are no known mechanisms of regulated iron excretion in mammals. Previous studies have assessed and described cutaneous responses to iron in humans and murine models of iron overload. In this study, we evaluated epidermal responses to systemic iron supplementation and chelation. Primary human keratinocytes store iron as ferritin as a function of exogenous iron addition. Furthermore, keratinocyte proliferation and the expression of several epidermal differentiation markers are modulated by iron. Wildtype mice subjected to systemic iron bolus or iron chelation demonstrate changes in total epidermal elemental iron content. By immunofluorescence, ferritin light chain was found to specifically localize to the surface epidermis and hair follicle of mouse skin and was reduced by iron-chelation. Finally, the levels of the epidermal differentiation markers filaggrin and loricrin were found to be dependent on iron in vivo. These findings suggest that epidermal turnover may be a regulated mechanism of iron excretion that matches epidermal behavior to systemic iron levels.

Staphylococcus aureus (SA) is associated with increased severity of atopic dermatitis (AD) and promotes skin barrier disruption and Th2/Th17 skin inflammation in murine models. To investigate the mechanisms responsible for this phenotype, we tested the response of human keratinocytes and murine skin models to SA with targeted gene deletions in specific toxins. SA phenol-soluble modulin α (PSMα) was responsible for induction of trypsin activity and IL6 in keratinocytes. Interestingly both PSMα and SA secreted proteases were crucial for inducing barrier damage in mice (increased transepidermal water loss (TEWL)) and trypsin activity) and inflammation (erythema, elevated Il-6, Il17a/f, and Il4). These SA toxins are under control of the agr quorum-sensing system and we thus hypothesized that this might be a target for other members of the skin microbiome to influence SA. Using a clinically isolated S. hominis (SH) strain and a SA agr p3 YFP reporter strain, we demonstrated that a secreted factor (<3kDa) from SH, identified as a novel autoinducing peptide (AIP), could prevent transcription of SA toxins by inhibiting agr activation in SA (baseline SA agr activity:1794.39±150.27(RFU) vs 188.32±24.45(RFU) with SH). Inhibition of SA agr activity did not inhibit SA growth but prevented the response of keratinocytes to SA (baseline SA trypsin activity:79.46±1.64(ΔOD435nm/h) vs 17.74±4.68(ΔOD435nm/h) with SH). Inhibition of agr activity also potently prevented skin inflammation and barrier damage in mice (baseline TEWL with SA: 90.06±8.63(g/m2h) vs 28.53±23.21(g/m2h) after SH). Additional analysis has defined multiple species and strains of skin microbes that can turn off SA agr activity including SH, S. epidermidis, and S. warneri. Overall these findings show the benefits of a diverse skin microbiome and provide support for a new therapeutic rationale of applying specific commensal microbes to AD patients.