Cutaneous inflammation is linked to cancer risk. A number of chemokines and cytokines are expressed by skin keratinocytes. Among proinflammatory proteins important in cutaneous disease are a family of calcium binding S100 proteins that are encoded within the Epidermal Differentiation Complex (EDC) on chromosome 1q21). Two of these proteins, S100A7 (psoriasin) and S100A15 (koebnerisin), were first identified in inflamed psoriatic skin. Psoriasis is genetically linked to the human S100A7/A15 subfamily. We have developed bitransgenic mice expressing elevated levels of mS100A7A15 by skin keratinocytes. These mice are primed for an exaggerated cutaneous inflammatory response when challenged by exogenous stimuli (Koebner phenomenon). Inflammation-prone transgenic skin is infiltrated with immune cells and expresses elevated levels of Th1 and Th17 proinflammatory molecules linked to the pathogenesis of psoriasis, which are further amplified upon challenge. Both, inflammation priming and amplification require mS100A7A15 ligand and the receptor of advanced glycated end products (RAGE). mS100A7A15 potentiates inflammation directly as a chemoattractant further enhancing the inflammatory infiltrate in skin from bitransgenic mice. This study models a functional mechanism for a psoriasis candidate gene and emphasizes the link between the epidermal and immune compartments as a pathogenetic model for inflammation priming. Thus, targeting S100A7A15-RAGE may be a novel therapeutic approach for treatment of susceptibility and inflammation in psoriasis, other inflammatory skin disorders and prevention of cancer development. The Epidermal Growth Factor Receptor (EGFR) is another regulator of epidermal homeostasis and in particular participates in immunohomeostasis of skin. Human cancer patients on EGFR inhibitor chemotherapy often display an inflammatory cutaneous reaction as an adverse consequence of therapy. To model and study this cutaneous inflammatory response we crossed Keratin 5 driven Cre recombinase transgenics with EGFR floxed mice in order to ablate EGFR in the epidermis. At 2 weeks of age total lymphocytes are reduced while neutrophils and platelets are increased in double transgenic mice when compared to WT littermates. At the same age increased neutrophils are detected via MPO analysis in the skin of double transgenic mice. Systemic alterations include also higher levels of cytokines and chemokines detected earlier in life. The early detection of CCL17, CCL22, CCL2, CCL11, CXCL1, TNF-alpha, IL1ra, IL-6, IL-17 and IL-18 in the plasma highlights a possible pathogenetic role of these inflammatory mediators in the skin lesions that develop in absence of EGFR expression. Before the skin phenotype develops at 7/8 days of life, we detect higher levels of mRNA for CCL2, TNF-alpha, IL1-beta and IL-17 expression in total skin RNA extracts. In contrast all the other inflammatory mediators detected in the plasma at 2 weeks of age are upregulated later. During the first week of life, CD45 positive (total leukocytes) cells infiltrating the skin of EGFR ablated mice are already higher. TTP (Tristetraprolin) regulates RNA stability by recognition of 3 non-coding sequences and targets the RNA for degradation. Chemokines and cytokines are substrates for this action. Keratinocytes respond to TNF-alpha by increasing the expression levels and phosphorylation status of TTP. The activation of EGFR pathway by TGF-alpha increases TTP to even higher levels than TNF-alpha treatment, and pretreatment with EGFR inhibitors prevents TNF-alpha induction of TTP. Using the transcription inhibitor Actinomycin D, we observe that the half life of GM-CSF mRNA is longer in keratinocytes isolated from EGFR KO mice or in human keratinocytes treated with EGFR inhibitors. TTP siRNA treatment causes a reduction of the decay rate of proinflammatory cytokines like GM-CSF and IL-8 in human keratinocytes. Further, a more potent TNF-alpha/TGF-alpha synergy in the expression of GM-CSF is detected in TTP knockout keratinocytes. Having a mouse model of EGFR ablation in the epidermis gives an opportunity to reveal downstream consequences of EGFR inhibition that might contribute to the cutaneous inflammation. Further, since keratinocytes are transformed to a tumorigenic phenotype by oncogenic ras, and this oncogene is reported to produce resistance to anti-EGFR therapy, we are able to evaluate pathways that might contribute to therapeutic resistance. Microarray analysis was performed on cDNA from Ras transformed mouse keratinocytes derived from wildtype or EGFR null mice or treated with EGFR tyrosine kinase inhibitors AG1478, Tarceva, and PD153035. Comparison of significant gene lists generated by treatment with AG1478, Tarceva, PD153035 and the EGFR null keratinocytes created a 25 gene signature for EGFR ablation. All 25 genes are concordant with 4 genes down-regulated and 21 genes up-regulated. Ingenuity Pathway Analysis of the signature genes revealed direct and indirect linking between the majority of the signature genes. Real time polymerase chain reaction, ELISA and western blot analysis of selected genes showed concordance with the array data. In collaboration with the laboratory of Giorgio Trinchieri, we found that mice ablated for IL-1R or MyD88, an intermediate in the innate immune response through Toll-like/IL-1R signaling, are resistant to topical skin carcinogenesis. Oncogenic RAS initiated MyD88-/- keratinocytes are hyperproliferative but form only a few small tumors in orthotopic grafts and fail to upregulate pro-inflammatory genes or downregulate differentiation markers characteristic of RAS initiated wildtype keratinocytes. Using both genetic and pharmacological approaches, we find that the differentiation and pro-inflammatory functions mediated by oncogenic RAS in keratinocytes require the establishment of an autocrine loop through IL-1alpha, its receptor and MyD88 leading to phosphorylation of IkappaBalpha and NF-kappaB activation. Blocking the IL-1alpha mediated NF-kappaB activation in RAS initiated wildtype keratinocytes reverses the defect in differentiation response and inhibits proinflammatory gene expression. Collectively, these results demonstrate that RAS activation converts normal keratinocytes to an initiated phenotype through a series of potentially reversible feedback signals that provide therapeutic opportunities through inhibition of IL-1 signaling. Inflammatory cytokines and chemokines also play an important role in mediating melanoma development, growth, and metastasis and may originate from keratinocytes and stromal cells, key components of the melanoma microenvironment. In collaboration with Dr. Glenn Merlino we generated HGF-PKCalpha mice, a cross between K5-PKCalpha mice (prone to skin inflammation) and MT1-HGF mice (prone to develop melanoma). HGF-PKCalpha mice that were treated with the DMBA/TPA regimen developed significantly more papillomas than their K5-PKCalpha counterparts, did not develop melanomas until after TPA treatment ended, and the number of melanomas was reduced by 50% compared to MT1-HGF mice. This suggests that keratinocyte-mediated inflammation serves to inhibit melanomagenesis in HGF-PKCalpha mice and that HGF may play a role in promoting papilloma growth. HGF-treated primary FVB/N keratinocytes expressed significantly higher levels of inflammatory cytokines such as CXCL1, CXCL2 and GM-CSF as well as pro-angiogenic factors such as VEGF compared to control keratinocytes. Together, these results present evidence of keratinocyte-mediated regulation of melanoma development and suggest a role of HGF in squamous cell carcinogenesis.
Showing the most recent 10 out of 21 publications
