If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Dermatology, Venereology and Allergology, University Medical Center Göttingen, Göttingen, GermanyCharles Institute for Translational Dermatology, University College Dublin, Dublin, IrelandDepartment of Dermatology and Surgery, University of California, San Francisco, California, USA
Department of Dermatology and Surgery, University of California, San Francisco, California, USADepartment of Dermatology, University of Münster, Münster, Germany
Charles Institute for Translational Dermatology, University College Dublin, Dublin, IrelandDepartment of Dermatology and Surgery, University of California, San Francisco, California, USADepartment of Dermatology, University of San Diego, USA
Rosacea is a common chronic inflammatory skin disease of unknown etiology. Our knowledge about an involvement of the adaptive immune system is very limited. We performed detailed transcriptome analysis, quantitative real-time reverse-transcriptase–PCR, and quantitative immunohistochemistry on facial biopsies of rosacea patients, classified according to their clinical subtype. As controls, we used samples from patients with facial lupus erythematosus and healthy controls. Our study shows significant activation of the immune system in all subtypes of rosacea, characterizing erythematotelangiectatic rosacea (ETR) already as a disease with significant influx of proinflammatory cells. The T-cell response is dominated by Th1/Th17-polarized immune cells, as demonstrated by significant upregulation of IFN-γ or IL-17, for example. Chemokine expression patterns support a Th1/Th17 polarization profile of the T-cell response. Macrophages and mast cells are increased in all three subtypes of rosacea, whereas neutrophils reach a maximum in papulopustular rosacea. Our studies also provide evidence for the activation of plasma cells with significant antibody production already in ETR, followed by a crescendo pattern toward phymatous rosacea. In sum, Th1/Th17 polarized inflammation and macrophage infiltration are an underestimated hallmark in all subtypes of rosacea. Therapies directly targeting the Th1/Th17 pathway are promising candidates in the future treatment of this skin disease.
Rosacea is a common, chronic inflammatory skin disease with poorly understood etiology. On the basis of its symptoms such as facial erythema, chronic inflammation, and fibrosis (
), complex pathomechanisms involving dysregulation in the immune, vascular, and nervous systems, as well as barrier function impairments, are suggested (
The knowledge about rosacea immunology is still very limited, and most studies focused on the involvement of the vascular or innate immune system. Environmental rosacea trigger factors (e.g., UV, temperature changes, spicy food, exercise, “stress”, demodex) or an altered skin microbiome activates (directly or indirectly) pattern recognition receptors such as Toll-like receptors (TLRs), which stimulate the release of antimicrobial peptides (among them the cathelicidin LL-37) or cytokines, for example (
). In rosacea, an increased expression of TLR2, LL-37, and kallikrein 5 (the protease that cleaves the cathelicidin precursor protein into LL-37) has been observed (
Reduction in serine protease activity correlates with improved rosacea severity in a small, randomized pilot study of a topical serine protease inhibitor.
). In addition, cells of the innate immune system, among them macrophages, mast cells, and neutrophils, were described as part of the cellular infiltrate in rosacea (
Little is known about the adaptive immune system in rosacea. It was shown recently that in other inflammatory skin diseases such as acne, psoriasis, or atopic dermatitis, the adaptive immune system is anticipated to contribute to the inflammatory processes (
). Only very few studies exist, which investigate a role of the adaptive immune system in rosacea and show that the cellular infiltrate in rosacea also contains T cells and B cells (
). With regard to T cells, CD4+ T cells are reported to be more frequently found than CD8+ T cells in papulopustular rosacea (PPR) and in nonpustular rosacea (
). However, these studies are only based on histological findings mainly in pustular rosacea, and they do not investigate and distinguish between different cutaneous rosacea subtypes simultaneously, and they do not correlate histological findings with cellular pathways.
Therefore, the aim of this study was to characterize the inflammatory infiltrate of the three cutaneous rosacea subtypes comprehensively. Moreover, we performed whole transcriptome expression analysis and confirmed the results with quantitative real-time reverse-transcriptase–PCR (qRT–PCR) and quantitative immunohistochemistry of rosacea patient samples, as compared with healthy skin and facial lupus erythematosus (LE).
Results
CD4+ T helper cells dominate the immune cell infiltrate in erythematotelangiectatic rosacea (ETR) and PPR
Our study reveals strong levels of T-cell involvement in all three rosacea subtypes (Figure 1 and Supplementary Figure S1 online), reaching a maximum of activation in PPR (PPR>phymatous rosacea (PhR)>ETR). At the transcriptome level, strong upregulation of CD3G, CD3D, and CD247 (=CD3Z), which transcribe the CD3 subunits of the αβ T-cell receptor (
), can be observed for all rosacea stages. In addition, this correlates well with increased expression of the T-cell activation– and proliferation–associated genes LCK, MICB, VAV1, and LAG3 (
). We also observe identical patterns in genes encoding for costimulatory molecules for T-cell activation (such as CD80, CD86, TNFSF14), as well as the proinflammatory cytokine IL-1β (
). For confirmation of our transcriptome data, we performed a qRT–PCR study on a total of 68 immunological markers for each rosacea subtype. We globally obtained a good agreement between transcriptome and qRT–PCR results (Supplementary Table S1 online). In support, quantitative immunohistochemistry (IHC) verifies a significant increase of CD4+ cells in all subtypes, with highest CD4 counts in PPR. Interestingly, CD4+ cells in PPR were mainly distributed around hair follicles, whereas CD4+ cells were found to be predominantly localized in the perivascular region in the other subtypes of rosacea. Although expression of CD4 was only moderately elevated in all rosacea stages (30–40% higher than controls) based on the Affymetrix analysis, the results were always significant (false discovery rate<0.002 in all three stages). Immunohistochemical staining for CD8 showed spare cells in all three subtypes with a trend toward higher expression in PPR, but without reaching statistical significance. At the transcriptome level, CD8A was significantly elevated in all rosacea forms (PPR>PhR>ETR), whereas the increase of CD8B did not reach significance.
Figure 1Rosacea shows activation and proliferation of T cells, which is mainly owing to CD4+ T helper cells, but not CD8+ cytotoxic T cells. The table (a) depicts genes that are involved in T-cell activation and proliferation and significantly modulated in at least one of the rosacea subtypes (full list of gene sets in Supplementary Figure S1 online). Fold modulations in bold face type depict false discovery rate (FDR)<0.05; fold modulations in italic type depict FDR>0.05. The upper graphs depict representative pictures (b,c,e–g) and morphometric analysis (d) of CD4+ cells in each rosacea subgroup (erythematotelangiectatic rosacea/rosacea type I (RI), papulopustular rosacea/rosacea type II (RII), phymatous rosacea/rosacea type III (RIII)), lupus erythematosus (LE) patients, and healthy skin (HS), all comprising n≥5 patients with five representative areas per slide (eff.pos.area = effective positive area). The lower graphs and pictures display the results for CD8+ cells (h–m). All images are × 200 magnification (bar=100 μm).
Immune response in rosacea displays Th1/Th17 polarization
Detailed transcriptome analysis of the induced T-cell response genes revealed significantly higher gene expression levels for the Th1-signature cytokines IFN-γ and tumor necrosis factor-α (Figure 2). This finding is consistent with higher messenger RNA expression levels for the Th1-related transcription factors signal transducer and activator of transcription 1 and 4 (
). We confirmed these results by IHC and found a largely higher staining for IFN-γ, especially in PPR.
Figure 2The T helper cell subset is polarized toward a Th1/Th17 immune response. The upper table (a) shows Th1 cell–related genes significantly modulated in at least one of the rosacea subtypes. Fold modulations in bold face type depict false discovery rate (FDR)<0.05; fold modulations in italic type depict FDR>0.05. Representative pictures of IFN-γ staining by immunohistochemistry (IHC) are depicted below (b–g), with results shown in each rosacea subgroup (erythematotelangiectatic rosacea/ rosacea type I (RI), papulopustular rosacea/rosacea type II (RII), phymatous rosacea/rosacea type III (RIII)), healthy skin (HS), murine spleen, and controls without primary antibody. The lower table (h) displays significant alterations of Th17 cell–related gene expression only and corresponding IHC for IL-17A (i–n). All images are × 200 magnification (bar=100 μm); inserts taken from the same slide located in the lower right corner are × 1,000 magnification (bar=20 μm). More detailed gene lists including nonaltered Th2 and regulatory T-cell gene sets, as well as IHC for noninvolved IL-4 in rosacea, are displayed in Supplementary Figure S2 and S3 online.
Besides the Th1-polarized immune response in rosacea, we also found induction of Th17-associated genes (Figure 2). The genes coding for Th17-signature cytokines IL17A and IL22 were significantly elevated (again PPR>PhR>ETR). The same was true for the genes IL6, tumor necrosis factor, IL20, and CCL20, which induce IL-17/IL-22 production, and for the Th17-associated transcription factor signal transducer and activator of transcription 3 (
). These results were verified by significant IL-17 immunostaining in all subtypes, with highest staining in PPR patients. Neither transcriptome nor IHC analysis verified an involvement of Th2-associated genes of the cytokine IL-4 (Supplementary Figure S2 and S3 online). Finally, no upregulation of genes were found associated with activation of regulatory T cells (unaltered expression of FOXP3, IL10, CCR4, and CCR8, depicted in Supplementary Figure S3 online).
Recruitment and activation of macrophages and neutrophils differ between rosacea subtypes
To define an involvement of macrophages in rosacea, we analyzed messenger RNA expression levels of CD68, CD14, CD163, and MSR1. The finding of significant upregulation of these macrophage markers in PPR was supported by identical upregulation of ITGB2 and ITGAM messenger RNA (Figure 3), genes that encode proteins that form the macrophage-1 antigen (or integrin αMβ2). Several other important activators of macrophages (and other inflammatory immune cells) were also significantly upregulated at the transcriptome level, once more following the pattern PPR>PhR>ETR: TLR1, TLR4, CHI3L1/YKL-40, ARG1 (
). Immunohistochemical staining for CD68 confirmed our transcriptome data, demonstrating highly significant increases of CD68+ cells in all subtypes (PPR: P<0.001, ETR and PhR: P<0.05). In particular in PPR, CD68+ cells were vastly organized as granulomatous areas and interfollicularly (Figure 3), whereas ETR macrophages showed a more perivascular diffuse pattern similar to CD4+ cells.
Figure 3Recruitment and activation of macrophages and neutrophils differ between rosacea stages. The upper table (a) depicts monocyte/macrophage-related genes significantly modulated in at least one of the rosacea subtypes, which are involved in activation and proliferation (full list of gene sets in Supplementary Figure S4 online). Fold modulations in bold face type depict false discovery rate (FDR)<0.05; fold modulations in italic type depict FDR>0.05. The graphs depict representative pictures (b,c,e–g) and morphometric analysis (d) of CD68+ cells in each rosacea subgroup (erythematotelangiectatic rosacea/rosacea type I (RI), papulopustular rosacea/rosacea type II (RII), phymatous rosacea/rosacea type II (RIII)), lupus erythematosus (LE) patients, and healthy skin (HS), all comprising n≥5 patients with five representative areas per slide (eff.pos.area = effective positive area). The lower table (h) displays neutrophil-associated genes with significant modulation.
Next, we analyzed RNA levels and distribution of neutrophils in all three subtypes of rosacea. Transcriptome analysis of CXC chemokines (Figure 3) shows significant upregulation of CXCL8 messenger RNA (formerly known as IL-8 or neutrophil chemotactic factor), one of the most potent chemotactic molecules for neutrophils. Similarly, other chemotactic chemokines such as CXCL1, CXCL2, CXCL5, and CXCL6 were found to be upregulated (
). Although neutrophils were found in pustular regions of PPR and PhR patients by immunohistochemistry, we found no evidence for an involvement of neutrophils in ETR.
Using immunohistochemistry (CD1a) and transcriptome analysis, no significant cell enhancement or RNA upregulation was observed for Langerhans cells in all three rosacea subtypes (Supplementary Figure S5 online). Distribution and transcriptome analysis of mast cells has recently been published (
Involvement of B cells and plasma cells in the regulation of immune responses in rosacea
Transcriptome analysis revealed significantly increased RNA expression levels of the B-cell marker MS4A1 (CD20) in PPR and PhR (Figure 4, Supplementary Figure S6 online). Similar results were obtained for the PAX5 gene, which encodes the transcription factor B-cell lineage–specific activator protein, and for the expression of CD24, which functions on activated B cells as a T-cell costimulator for CD4+ T-cell clonal expansion (
). We also found up to 170-fold upregulated genes involved in different parts of the light and heavy chains for immunoglobulins. Interestingly, all these genes reached the highest expressions in PhR (Figure 4). Because rearrangement and combinatorial procedures of immunoglobulins are especially complex in molecule assembly, these results need to be evaluated carefully. With IHC, we found highly significant increases in CD79a+ cells in all rosacea subtypes, with highest cell numbers in PhR. Of note, CD79a+ cells were often localized perivascularly or perifollicularly, or grouped as pseudofollicles in the papillar dermis. CD20 immunostaining showed an identical increase of positive cells in ETR and PPR, but dropped to almost baseline expression in PhR. As CD79a positivity is retained in plasma cells, whereas this late stage of B-cell differentiation shows no CD20 expression, it may be concluded that most of the CD79a+ cells in PhR are in fact immunoglobulin-producing plasma cells (
). Analysis of several genes roughly differentiating antigen-dependent (e.g., CD28) and antigen-independent (e.g., IL6, IL7) B-cell activation did not show a distinct pattern of activation (Figure 4).
Figure 4B cells and plasma cells are activated in the regulation of the immune response. The table (a) depicts B cell–related genes significantly modulated in at least one of the rosacea subtypes, which are involved in activation and proliferation (full list of gene sets in Supplementary Figure S5 online). Fold modulations in bold face type depict false discovery rate (FDR)<0.05; fold modulations in italic type depict FDR>0.05. The upper graphs depict representative pictures (b,c,e–g) and morphometric analysis (d) of CD79a+ cells in each rosacea subgroup (erythematotelangiectatic rosacea/rosacea type I (RI), papulopustular rosacea/rosacea type II (RII), phymatous rosacea/rosacea type III (RIII)), lupus erythematosus (LE) patients, and healthy skin (HS), all comprising n≥5 patients with five representative areas per slide (eff.pos.area = effective positive area). The lower pictures (h,I,k–m) and graph (j) display the results for CD20+ cells. All images are × 200 magnification (bar=100 μm).
Chemokine expression patterns support the Th1/Th17 polarization of the T-cell response
As the family of chemokines appears redundant and their ligand–receptor relationships are promiscuous for many cell types, interpretation of immune cell and immune system polarization behavior by chemokines needs to be done with caution (
). Moreover, as ligands of the Th1 cell–associated receptor CXCR3 and the Th2 cell–associated CCR3 are reciprocally natural antagonists, these ligands can form mutually exclusive microenvironments that are in favor of either Th1 or Th2 cell differentiation (
). Among all chemokines, we found the highest expression levels for the CXCR3 ligands CXCL9, CXCL10, CXCL11, and CXCL13 (which are also induced by IFN-γ). In contrast, CCR3 ligands such as CCL3, CCL5, CCL11, and CCL13 were only modestly upregulated or unchanged compared with healthy controls (Figure 5).
Figure 5Chemokine expression patterns support the Th1/Th17 polarization of the T-cell response. In all, 50 most upregulated and downregulated cytokines and chemokines, respectively, as well as their receptors, are depicted, sorted by expression in rosacea type II (RII). Fold modulations in bold face type depict false discovery rate (FDR)<0.05; fold modulations in italic type depict FDR>0.05.
C-C chemokine receptor 3 antagonism by the beta-chemokine macrophage inflammatory protein 4, a property strongly enhanced by an amino-terminal alanine-methionine swap.
). Activation of macrophages and neutrophil granulocytes is also reflected by the increased expression profile of the receptors CCR1, CCR2, CCR5, CCR7 (macrophages and monocytes), and CXCR2 (neutrophils), and their corresponding ligands CCL2, CCL3, CCL4, CCL5, CCL8, CCL19, and CXCL1, CXCL2, CXCL5, CXCL6, respectively (
). In contrast, CCL27 and IL-37 are the most significantly downregulated transcripts among all cytokines and chemokines, emphasizing again the importance of Th1/Th17 involvement in the pathophysiology of rosacea. CCL27 is involved in cutaneous homing of Th2 cells in atopic dermatitis, it is downregulated in psoriasis plaques, and IFN-γ has been demonstrated to be the responsible component for downregulation of CCL27 in keratinocytes (
). IL-37 (IL-1F7) suppresses immune responses, and its protective activity affects a broad spectrum of Th1/Th17 proinflammatory insults, placing this cytokine as a new member in the portfolio of anti-inflammatory cytokines such as IL-10 or TGF-β (
). Thus, the considerable downregulation of the anti-inflammatory IL-37 may result in increased inflammation in rosacea. In summary, although a vast majority of altered cytokines and chemokines reach their maximum of upregulation and downregulation in PPR patients, significant changes in the RNA expression levels for cytokines/chemokines can be observed in all three cutaneous subtypes.
Discussion
Although of importance, a detailed characterization of the inflammatory infiltrate and the associated cytokine/chemokine profiles in the three skin subtypes of rosacea has not been performed yet. Here, we comprehensively analyzed the inflammatory infiltrate of the three rosacea subtypes (ETR, PPR, and PhR) in comparison with healthy skin. In addition, we characterized the inflammation hallmarks of rosacea and LE, a facial chronic inflammatory autoimmune disease of different origin. Our combined transcriptome and immunohistochemistry analysis demonstrates Th1/Th17 dominance in all three subtypes (Figures 1 and 2), as well as the presence of macrophages (Figure 3) and mast cells in all subtypes (
). In contrast, B cells or neutrophils can only be found in PPR and PhR with a specific staining pattern. In contrast, activation of Langerhans cell–associated genes or cell numbers could not be observed. Together, these observations suggest the involvement of the innate and adaptive immune system in the pathophysiology of rosacea.
Recently, the involvements of a dysfunctional innate immune system and a dysregulated neuro-immune communication network controlling vascular responses have been discovered as important contributors in the pathophysiology of rosacea. In early rosacea (ETR), a malicious vasodilation to normally innocuous physical stimuli (sunlight, temperature changes, spices) and associated imbalanced neuro-immune circuits may be important for the characteristic flushing responses of rosacea patients (
). A detailed analysis of the inflammatory mediators and cells that are involved in innate and adaptive immunity in rosacea patients, however, is so far lacking.
We show that the adaptive immune system is broadly and significantly activated, which interestingly starts as early as in ETR, and increases further in PPR. We identified the increased T-cell population as consisting of predominantly CD4+ T helper cells, and our transcriptome data revealed an upregulation of Th1- and Th17-polarizing gene sets. We confirmed these findings by immunohistochemistry for the signature cytokines IFN-γ and IL-17A. Of note, the highest increase for all genes involved was observed in PPR, followed by PhR and ETR (both with roughly comparable changes). We observed an identical pattern of activation in macrophages and neutrophils (PPR>PhR>ETR). Together with our data that mast cells and mast cell–associated genes are increased in all three subtypes of rosacea, our studies indicate a critical role of Th1/Th17 and mast cells, as well as for macrophages in rosacea pathophysiology.
Previous studies on T cells in rosacea have been very limited in number or subtypes, and were exclusively histology based. One recent study compared T-cell subsets in nonpustular rosacea vs. LE patients with immunohistochemistry (
). They found a T cell–rich lymphocytic infiltrate in rosacea with a ratio of CD4+/CD8+ T cells of 2.80. This is in line with our detailed findings of a CD4+ T cell–driven inflammatory response and their associated genes. Two other histological studies also describe an increase of CD4+ over CD8+ cells exclusively in PPR patients, and in ocular rosacea, respectively (
). Our study, however, suggests a contribution of CD4+ cells already in ETR, and also in PhR, although at different levels. In support, CD4+-associated genes are already significantly upregulated in ETR.
Of note, Th17 pathways were recently unveiled as being importantly involved in acne vulgaris (
have demonstrated that Propionibacterium acnes has a central role in driving the Th17/Th1 response in acne vulagris. Hypothetically, microorganisms may also be involved in polarizing the Th1/Th17 immune response in rosacea, although the role of microorganisms has been a matter of considerable debate in recent years. Several studies link the microbes Demodex folliculorum, Bacillus oleronius, Helicobacter pylori, Staphylococcus epidermidis, Chlamydia pneumoniae, and Bartonella quintana to rosacea pathophysiology (
). Underlying mechanisms may be increased by bacterial loads and/or hyperreactivity of the patients’ immune system. Nevertheless, the causative role of microorganisms (and especially Demodex spp. and/ or their associated microbiota) remains unclear in rosacea as of now. Although reduction of facial Demodex colonization has been achieved in multiple clinical trials on rosacea, a higher reduction of Demodex colonization did not directly translate into better clinical outcome in these patients (
It has been well recognized over the past years that dysregulation of the innate immune response contributes to rosacea pathophysiology significantly. The antimicrobial pro-protein cathelicidin and its fragment LL-37 were found to be elevated and abnormally processed (
). As the kallikrein (KLK) serine proteases 5 and 7 are major activators of antimicrobial peptides such as cathelicidin, the abundance of KLK5 in rosacea is responsible for processing the excess cathelicidins into aberrant forms, which fuels the course of the disease (
Calcipotriol increases hCAP18 mRNA expression but inhibits extracellular LL37 peptide production in IL-17/IL-22-stimulated normal human epidermal keratinocytes.
), and therefore Th17 cytokines may also lead to or enhance the dysregulation of LL-37 expression.
Besides polarization of Th1/Th17-dominated immune responses, we also demonstrated upregulation of macrophage- and neutrophil-associated gene sets, as well as the influx of macrophages and their perifollicular and perivascular localization by IHC. This adds up well to our recently published observations on upregulation of mast cells in all rosacea subtypes, reaching a maximum in PPR (
). We did not observe any significant activation or expansion of Langerhans cells by transcriptome data, quantitative PCR, and IHC in all rosacea substages.
With regard to B cells, we found an unexpectedly massive upregulation of immunoglobulin-associated gene sets in all subgroup analyses, and IHC confirmed a significant increase in plasma cells, but not of CD20+ B cells (Figure 4). Plasma cells and antibody production showed the highest involvement in PhR at the gene level, and strikingly already ETR patients revealed a significant increase in plasma cell numbers. Although the increase of plasma cells has been acknowledged in earlier studies, only limited research has focused on their pathophysiological role in rosacea (
). Thus, the role of plasma cells in rosacea pathophysiology remains highly speculative at this time. Interestingly, the bacterium B. oleronius can be isolated from a D. folliculorum mite extracted from a PPR patient (
). It was demonstrated that two highly immunogenic proteins from B. oleronius were capable of mounting an inflammatory response in PPR patients. Of note, the same reactivity to B. oleronius was already found in ETR patients, and also confirmed in patients with ocular rosacea (
Correlation between serum reactivity to Demodex-associated Bacillus oleronius proteins, and altered sebum levels and Demodex populations in erythematotelangiectatic rosacea patients.
). These promising findings warrant further verification, and proof of clinical improvement in patients with decreased numbers of B. oleronius is still lacking. Alternatively, autoimmune processes associated with molecular mimicry of antigens from microbiota and skin cells cannot be excluded, and should be more considered in the future.
In conclusion, our study showed significant involvement of the immune system in all subtypes of rosacea, characterizing even ETR as a chronic disease with significant inflammatory activity. The T-cell response is dominated by a Th1/Th17-polarized immune response, as we verified by correlating enhanced expression profiles and staining for signature cytokines including IFN-γ and IL-17. Chemokine expression patterns also support the Th1/Th17 polarization of the T-cell response, and recruitment, as well as activation, of macrophages or neutrophils, which is already present in ETR and reaches a maximum in PPR. Our current hypothesis of the complex interplay of the different skin cells in rosacea is depicted in Figure 6. We also provide evidence for a vast activation of antibody-producing plasma cells, beginning already in ETR, followed by a crescendo pattern toward PhR. Thus, an altered adaptive, as well as innate, immunity is the hallmark in the development of rosacea. Therapies targeting the Th1/Th17 pathway are promising candidates for an optimized treatment of this frequent, chronic skin disease.
Figure 6Hypothesis on involvement of the immune system in rosacea. This cartoon depicts our hypothesis how the complex network of soluble and cellular components of the innate and acquired immune system may interact for the generation of the Th1/Th17-polarized immune response in rosacea.
For Affymetrix microarray studies and qRT–PCR-analysis, we studied a group of 19 rosacea patients, of whom 7 were diagnosed with ETR, 6 with PPR, and 6 with PhR based on the classification system of the American Rosacea Society (
). Skin biopsies were taken for all disease subtypes from the nasolabial fold of the patients with the aim to minimize differences based on biopsy localization. In addition, skin biopsies of 10 healthy individuals were investigated. Patients with systemic LE displaying a malar rash were identically biopsied from the nasolabial fold for comparison. Skin biopsies were obtained in local anesthesia after the patients gave their written consent. Permission for human studies was given by the Ethical Committee of the University of Münster, Germany, in accordance with the ethical standards of the 1964 Declaration of Helsinki Principles.
Additional materials and methods for this manuscript can be found in the supplement.
ACKNOWLEDGMENTS
We thank C Mess, C Menigot, S Corvaisier, and S Wise for excellent technical assistance. Furthermore, we highly appreciate the contribution of all patients in this study. This work was supported by grants of the German Research Foundation to TB (BU2991/1-1), MSu (SU803/1-1), MSt (IZKF, STE1014), and by the Dublin Skin and Cancer Hospital Charity (Ireland) to MSt.
Correlation between serum reactivity to Demodex-associated Bacillus oleronius proteins, and altered sebum levels and Demodex populations in erythematotelangiectatic rosacea patients.
C-C chemokine receptor 3 antagonism by the beta-chemokine macrophage inflammatory protein 4, a property strongly enhanced by an amino-terminal alanine-methionine swap.
Calcipotriol increases hCAP18 mRNA expression but inhibits extracellular LL37 peptide production in IL-17/IL-22-stimulated normal human epidermal keratinocytes.
Reduction in serine protease activity correlates with improved rosacea severity in a small, randomized pilot study of a topical serine protease inhibitor.
38995-8&id=gr1.jpg){kind=link}
38995-8&id=gr2.jpg){kind=link}
38995-8&id=gr3.jpg){kind=link}
38995-8&id=gr4.jpg){kind=link}
38995-8&id=gr5.jpg){kind=link}
38995-8&id=gr6.jpg){kind=link}