Rosacea and acne are the most common inflammatory chronic skin diseases. Clinically, we frequently see acne patients with rosacea or patients who develop from acne to rosacea. Genetics, microbiomes, innate and adaptive immunological dysregulation, and skin barrier dysfunction have been implicated as shared pathogenetic factors in both rosacea and acne5. When rosacea and acne coexist, rosacea is frequently missed or misdiagnosed due to the similarity of their clinical manifestations. However, few studies have investigated their relationship. The purpose of this study was to explore the relationship between rosacea and acne. In this study, hub genes were identified by screening DEGs and constructing weighted gene co-expression and PPI networks. And TFs that regulate the expression of these hub genes have been identified. Finally, we performed immune cell infiltration analysis to compare the immune cell infiltration characteristics of rosacea and acne lesions.
A total of 151 up-regulated and 18 down-regulated co-DEGs were identified. KEGG pathway analysis of up-regulated co-DEGs demonstrated that the they were mainly involved in IL-17 signaling pathway, NF-κB signaling pathway, Toll-like receptor signaling pathway, and TNF signaling pathway. This is consistent with previous findings that these pathways are involved in the development of both rosacea and acne6–12. 10 co-DEGs, IL1B, PTPRC, CXCL8, MMP9, CCL4, CXCL10, CD163, CCR5, CXCR4, and TLR8, were identified as hub genes. The GO and the KEGG pathways analysis of hub genes revealed that they were mainly involved in immunological and inflammatory processes. And 14 pathways were significantly positively correlated with hub genes in both rosacea and acne datasets. The expression of CXCL10, MMP9, IL1B, CXCL8, and CXCR4 were co-regulated by IRF1, STAT1, STAT3, IKBKB, HDAC1, ETS1, and CEBPB, which were highly expressed in rosacea and acne lesions. Moreover, MMP9 was significantly positively correlated with M0 macrophages in both rosacea and acne lesions. The infiltration of gamma delta T cells was significantly increased and positively correlated with almost all hub genes in both rosacea and acne, suggesting that gamma delta T cells may play a crucial role in the pathogenesis of these two diseases.
Interleukin (IL)-1β is a potent proinflammatory cytokine which belongs to the IL-1 family. Previous studies have indicated that IL-1β plays a crucial role in the development of inflammation in rosacea and acne. Meng et al. found that the expression of IL-1β in lesions of rosacea patients was significantly higher than age- and sex-matched healthy volunteers, especially PPR. In acne, all patients with inflammatory acne have dermal and epidermal IL-1β expression, and the expression level correlates with disease severity13. Ultraviolet B and P. acnes are triggers for IL-1β secretion, and the inflammation induced by P. acnes causes the massive secretion of IL-1β from monocytes14–16.
Matrix metalloproteases 9 (MMP9) is a member of the MMPs family that can degrade and remodel extracellular matrix proteins17. MMPs mediate the rupture of the pilosebaceous follicle, whereby promotes the exacerbation of inflammation18. The level of MMP9 expression is significantly increased in the lesions and serum of rosacea patients19,20. In addition, the expression of MMP9 in the dermis of granulomatous rosacea lesions is significantly higher than that of non-granulomatous rosacea lesions, especially at the granuloma’s center21. MMP9 induced by ultraviolet radiation has a role in the development of granuloma by promoting tissue remodeling and enhancing recruitment of inflammatory cells into the granuloma21. MMP9 also participates in various cellular processes, including angiogenesis and vasodilation 17,22. Additionally, MMP9 contributes to the progression of acne. Kang et al. found that the expression level of MMP9 in acne lesions was significantly higher than in non-lesional skin23. As one of the major pathogeneses of acne, P. acne was reported to induce the production of MMP9. Isotretinoin, quercetin, and daylight photodynamic therapy are all effective acne therapies that have been shown to decrease MMP9 expression24–26. MMP9 induces the production of pro-inflammatory cytokine, participates in angiogenesis, vasodilation, and tissue remodeling, which may make it the common hub gene in rosacea and acne.
CD163 is a macrophage-specific protein that is a member of group B of the scavenger receptor cysteine-rich superfamily and plays a crucial role in the regulation of the immune response27,28. The upregulation of CD163 is one of the major alterations in the macrophage transition to alternate activated phenotypes during inflammation27. In rheumatoid arthritis, psoriasis, atopic dermatitis (AD), and rosacea, CD163 levels have been shown to be elevated29–34. In a study including 50 rosacea patients, significant overexpression of CD163 was observed in the skin samples of PPR patients compared to the age- and sex-matched healthy volunteers29. In addition, CD163+ macrophages also play a crucial role in the progression of atherosclerosis (AS). Both rosacea and acne are closely related to AS. Several studies have shown a higher risk of AS in rosacea patients35–37. Massive uptake of oxidized lipid by macrophages and neutrophils (called foam cells) is believed to contribute to the common pathogenesis of acne and AS38. In AS, CD163+ macrophages are associated with plaque progression, microvascularity, and a high level of HIF1α (hypoxia-inducible factor 1α) and VEGF-A (vascular endothelial growth factor-A) expression39. Therefore, CD163+ macrophages may contribute to the development of AS in rosacea and acne patients.
Toll-like receptor 8 (TLR8), a type I transmembrane protein localized on the surface of endosomes, is expressed in monocytes, macrophages, myeloid dendritic cells, and neutrophils40. TLR8 induces NF-κB activation via the myeloid differentiation factor 88 (MyD88)-mediated signaling pathway and promotes the production of inflammatory factors41, which is crucial for innate immunity40. Previous studies demonstrated that TLR8 is involved in inflammatory dermatosis and autoimmune diseases, which is in the upstream of NF-κB-NLRP3 and associated with the production of IL-1β, interferon (IFN)-α, and IL-642–45. However, the role of TLR8 in the pathogenesis of rosacea and acne requires further exploration.
CXCL8, CXCL10, CCL4, CCR5, and CXCR4 all belong to the chemokine superfamily and play important roles in inflammatory processes and immune responses46. CXCL10 is a member of the ELR−CXC subfamily chemokines, which play a crucial role in a variety of autoimmune diseases47. CXCL10 binds to its ligand CXCR3 to guide CXCR3+ effector CD4+ and CD8+ T cells selectively migrated to autoimmune sites and tumor sites48. CXCL10 and CXCR3 were found in keratinocytes and dermal infiltrates from active psoriasis plaques, and psoriasis patients had higher serum levels of CXCL1049. Buhl et al. demonstrated that the expression of CXCL10 was increased in the lesional skin of rosacea patients50. Similar results were observed in a cohort study, the CXCR3 ligands CXCL9, CXCL10 and CXCL11 were overexpressed in acne lesions9. All of these findings indicate that the expression levels of CXCL10 are elevated in rosacea and acne lesions, and its role in the pathogenesis of these two diseases requires further study. In addition, we also found that the expression level of CXCR4 is increased in these two diseases in our study. Helfrich et al. found an 8-fold increase in CXCR4 expression in lesional skin from ETR patients compared to HC51. Su et al. demonstrated that in the mouse model of allergic contact dermatitis, the CXCL12/CXCR4 signaling pathway induces itching and pain feeling52. It requires further study to determine whether the CXCL12/CXCR4 signaling pathway has a role in some rosacea and acne patients with itching and pain.
By the prediction and verification analysis of TFs, we found that IRF1, STAT1, STAT3, IKBKB, HDAC1, ETS1 and CEBPB regulate the expression of hub genes in a synergistic manner. Several previous studies have demonstrated that these TFs play a crucial role in the pathogenesis of rosacea and acne. Signal Transducers and Activators of Transcription 1 (STAT1) is a member of the STAT family, it is activated by Janus kinases (JAK)53. STAT1 is critical in inflammatory diseases, such as systemic lupus erythematosus, inflammatory bowel disease and psoriasis54–56. Activation of STAT1 leads to the upregulation of several pro-inflammatory chemokines in the epidermis, such as CXCL9, CXCL10, and CCL257. Saez-de-Ocariz et al. found that rosacea is a striking feature in family members with a STAT1 gain-of-function mutation, indicating that STAT1 may contribute to the development of chronic inflammation58. Deng et al. performed RNA-seq on lesional skin from rosacea patients and found that the epidermal STAT1/IRF1 signature were observed in all rosacea subtypes59. STAT1-regulated gene transcription mediates many of the immune and inflammatory actions of IFN-γ60. Previous studies have implicated that IFN-γ is implicated in the pathogenesis of rosacea and acne. P. acnes can induce IFN-γ secretion by Th17 cells and CD4+ T cells61,62. In an in vitro experiment, acne-associated P. acnes phylotypes induced 2–3 times higher levels of IFN-γ in PBMCs than healthy phylotypes63. In addition, the expression level of IFN-γ was significantly increased in rosacea lesions, which was validated by immunocytochemistry to have a largely higher staining for IFN-γ in lesions50. Whether STAT1 has a role in the pathogenesis of rosacea and acne by regulating gene transcription to mediate the immune and inflammatory actions of IFN-γ remains to be confirmed in the future.
Signal Transducers and Activators of Transcription 3 (STAT3) is a component of the IL-6 activated acute phase response factor complex and is also a member of the STAT family64. It has been demonstrated that the IL-6/STAT3 signaling pathway plays a role in inflammation65–67. Wang et al. found that STAT3 and related pathways are up-regulated in rosacea, and they speculated that STAT3 expression is up-regulated in keratinocyte after skin barrier dysfunction, leading to the secretion of inflammatory factors and immune infiltration68. In a model in which HaCaT cells treated with the antibacterial peptide LL-37 to simulate rosacea caused by Demodex folliculorum (D. folliculorum) infection, JAK 2 and STAT3 expression levels were increased, indicating that rosacea caused by D. folliculorum infection may be associated with the activation of the JAK/STAT signaling pathway69. These evidences indicate that STAT3 may be an important gene in the pathogenesis of rosacea and acne, and it remains to be confirmed whether medications that target STAT3 may effectively treat rosacea and acne.
NF-κB is normally sequestered in the cytoplasm, bound to inhibitory κB (IκB) proteins as an inactive complex70. In response to lipopolysaccharide (LPS), an endotoxin recognized by the TLR4 receptor on immune cells, cellular IκB kinase (IKK) complex is activated and the cytoplasmic IκB protein is phosphorylates71. The phosphorylated forms of IκB are subjected to ubiquitination and then degraded in the proteasome, which frees NF-κB to translocate to the nucleus where it regulates gene transcription72. Then NF-κB moves into the nucleus and binds to the κB motif of inflammation and immune genes, such as IL-1β73. As a component of the IKK complex, IKKβ is intimately associated with the expression of NF-κB. And in our study, we found that the expression levels of IKKβ is increased in both rosacea and acne lesions, and that it may regulate the expression of hub genes. It has been demonstrated that the NF-κB signaling pathway plays a crucial role in the pathogenesis of rosacea and acne23,74−77. Most inflammatory stimuli, including LPS, require the IKKβ subunit for NF-κB activation78. Benzothiazolone derivatives have been used in the treatment of acne79. Kim et al. found that LYR-71, a benzothiazolone derivative, is a potent IKKβ inhibitor that prevents NF-κB activation in macrophages and may help in inhibiting the expression of inflammatory cytokines73. These findings suggest that IKKβ may play a role in the pathogenesis of rosacea and acne by influencing the NF-κB signaling pathway and regulating the secretion of pro-inflammatory cytokines.
In addition, we also performed immune infiltration analysis for the rosacea and acne datasets, respectively. Notably, MMP9 was significantly positively correlated with M0 macrophages in both rosacea and acne lesions in our study. And the infiltration of gamma delta T cells in rosacea and acne lesions was higher than in HC and positively correlated with the most of hub genes. However, few previous studies have reported the role of gamma delta T cells in this process.
Gamma delta T cells (also called γδT cells) are an unconventional population of T lymphocytes. They are significantly enriched in mucosal and epithelial sites, such as the skin and respiratory, digestive, and reproductive tracts80. γδT cells has a variety of functions, including cytokine and chemokine production, antigen-presenting functions, and regulation abilities81, and therefore play a crucial role in autoimmune, infection, allergy, cancer, among others82–85. They produce a variety of cytokines, such as IL-17 and IFN-γ80. IL-17-producing γδT cells induces the recruitment of neutrophils and monocytes and increases the inflammation response80. In a variety of immune skin diseases, including AD, alopecia areata, and psoriasis, the number of intracutaneous γδT cells has been demonstrated to be increased86–88. Cai et al. found that epidermal hyperplasia and inflammatory response induced by IL-23 and IMQ were significantly decreased in T cell receptor δ deficient mice89. These evidences suggest that γδT cells are crucial in inflammatory diseases and may be therapeutic targets in the future. As for the role of γδT cells in the progression of rosacea and acne, additional research is required.
In addition, we also found that there was a significant positive correlation between MMP9 and M0 macrophages in rosacea and acne lesions. It has been demonstrated that MMP9 expression and M0 macrophage infiltration were significantly increased in both rosacea and acne lesions19,20,23,90, which is consistent with our results. MMP9 can degrade extracellular matrix and plays a crucial role in the development and spread of inflammatory response18. M0 macrophages (resting macrophages) can be polarized into M1 macrophages (classically activated macrophages) under the stimulation of IFN-γ and LPS and into M2 macrophages (alternatively activated macrophages) under the stimulation of IL-4 and IL-13. M1 macrophages then secrete a large number of pro-inflammatory factors, which mainly promote the development of inflammation. Anti-inflammatory factors are secreted by M1 macrophages, which serve an anti-inflammatory function91. Inflammation in rosacea and acne can be aggravated by polarizing macrophages towards the M1 phenotype92–95, indicating that macrophages play a crucial role in the pathogenesis of these two diseases. Previous studies have shown a significant positive correlation between MMP9 expression and M0 macrophages infiltration in inflammation-related diseases, such as coronary artery disease (CAD), AS, adhesive capsulitis and cancers96–99. However, the causal relationship between them has not yet been clarified. Linton et al. found that M0 macrophages secrete MMP9 in the early phases of pancreatic cancer to promote tumor progression 100. Therefore, we hypothesized that M0 macrophages may secrete MMP9 and act in concert with it to promote the development of inflammatory responses in rosacea and acne.
Despite the fact that rosacea and acne frequently occur together clinically and have cross pathogenesis, it is important to note that rosacea and acne are two separate diseases with distinct pathogenesis, and their clinical manifestations are not identical. The main difference between the clinical manifestations of rosacea and acne is that rosacea patients have persistent facial erythema, flushing, a burning sensation, telangiectasia, and in some cases hyperplasia of the facial tissue1. Rosacea is primarily caused by the dysregulation and upregulation of the innate immune system, leading to in excessive inflammation and vasodilation. In addition, hyperreactive neurovascular and exogenous factors play a significant role in the pathogenesis of rosacea10. Transient receptor potential (TRP) cation channels are widely expressed on keratinocytes and endothelial cells. Activation of the TRP family of channels leads to the release of mediators of neurogenic inflammation and pain, such as substance P and calcitonin gene-related peptide. Persistent facial flushing in rosacea patients is mainly induced by these vasoregulatory neuropeptides6. The decreased thermal pain threshold and increased burning sensation in the skin of rosacea patients may be related to the increased activity of the TRPs. In addition, connective tissue hyperplasia in some rosacea patients is related to the persistence of inflammation, the activation of mast cells and the release of MMP1, MMP9, IL-6, and histamine, and the increased activity of the TRPs11. However, acne is mainly caused by follicular hyperkeratinization, excess sebum, inflammation, and P. acnes, which does not typically involve neurovascular dysergulation75. Consequently, persistent erythema, flushing, burning sensation, and telangiectasia are not typically observed in acne patients. Therefore, despite the fact that rosacea and acne share some clinical manifestations and pathogenesis, great care must be taken not to confuse these two diseases in order to avoid missed and misdiagnosis.
However, our study also has some limitations. First, there were little datasets on rosacea and acne in the GEO database. Although we pooled two acne datasets, the samples size remained small. Second, we were unable to find an additional rosacea dataset to validate hub genes. Third, the conclusions of our study need to be further validated in an in vitro model, which will be the focus of our future studies. In addition, we believe that our study may provide ideas and directions for future research.
In summary, we identified 169 common DEGs after analyzing the datasets of rosacea and acne. Then we identified 10 hub genes that may play important roles in the pathogenesis of these two diseases by validating in another dataset, and we also identified TFs that may regulate the expression of these hub genes. Then we found the pathogenesis of rosacea and acne had some similarity in terms of immune responses through immune cell infiltration analysis. This is the first study to explore the common hub genes and critical immune cells of rosacea and acne, which helped to further clarify the common molecular pathogenesis of these two diseases.