c‐Src activation as a potential marker of chemical‐induced skin irritation using tissue‐engineered skin equivalents

Abstract Skin irritancy to topically applied chemicals is a significant problem that affects millions of people worldwide. New or modified chemical entities must be tested for potential skin irritancy by industry as part of the safety and toxicity profiling process. Many of these tests have now moved to a non‐animal‐based format to reduce experiments on animals. However, these tests for irritancy potential often rely on monolayer cultures of keratinocytes that are not representative of the skin architecture or tissue‐engineered human skin equivalents (HSE) using complex multi‐gene expression panels that are often cumbersome and not amenable for high throughput. Here, we show that human skin equivalents increase abundance of several phosphorylated kinases (c‐Src, c‐Jun, p53, GSK3α/β) in response to irritant chemical stimulation by phosphokinase array analysis. Specific phosphorylation of c‐SrcY419 was confirmed by immunoblotting and was plasma membrane‐associated in basal/spinous cells by phospho‐specific immunohistochemistry. Moreover, c‐SrcY419 phosphorylation in response to the irritants lactic acid and capsaicin was inhibited by the c‐Src inhibitors KB‐SRC and betaine trimethylglycine. These data provide the first evidence for c‐Src specific activation in response to chemical irritants and point to the development of new modes of rapid testing by immunodetection for first‐pass screening of potential irritants.

Examples based on monolayer keratinocyte luciferase reporterbased assay systems include KeratinoSens™ 2,3 and LuSens, 4 where chemically induced activation of the transcription factor, nuclear factor erythroid 2 (Nrf2) mediates its translocation to the nucleus, where it heterodimerises with other transcription factors (Mif/c-Jun) and binds to the antioxidant response element (ARE) in the promoter region, driving reporter luciferase expression. However, these tests are based on monolayer culture, whereas skin is comprised of a stratified squamous epithelium containing keratinocytes which display increasing levels of differentiation and keratinisation that play a major role in skin permeability to topically applied chemicals. The deficiencies of simple monoculture assays can be overcome by the use of tissue-engineered human skin equivalents (HSE) that accurately mimic the structure of human skin for skin irritancy testing.
A number of studies have used HSE to identify increased expression of several genes in response to many chemicals and known irritants. [5][6][7][8] However, these gene signature sets are often large comprising of between 7 and 25 genes that would be cumbersome for high-throughput screening. Analysing the immediate up-stream signalling cascade events upon treatment with a chemical irritant may not only identify activation of specific kinases in the epidermis that are important in irritant-response pathways but may also ascertain markers of irritancy that could be tested in more rapid assay formats.

| QUE S TION ADDRE SS ED
Do chemicals that cause skin irritancy induce common intracellular signalling events such as protein kinase phosphorylation and can these cellular events be used as biomarkers for chemicals with skin irritancy potential?

| E XPERIMENTAL DE S IG N
See Supporting Information.

| RE SULTS
Phosphokinase array analysis of HSE protein extracts showed a dramatic and significant increased abundance of phospho-c-Src Y419 when the known skin irritant, lactic acid (LA), was topically applied to HSE for 15 minutes, in comparison to the non-irritants methylparaben (MP) and cocamide diethanolamine (Co-DEA), or water applied as carrier control ( Figure 1). In our previous study, we identified a seven-gene signature panel to identify chemical irritant from non-irritant. 8 Four of the genes identified in this panel (IL-6, PTGS2, MAP3K8, MMP-3) are regulated by activation of the transcription factors AP-1 (c-Fos/c-Jun) and p65/NFκB. In line with these data, we found increased phosphorylation of both c-Jun S63 ( Figure 1) and p65 S536 ( Figure S1) in response to irritants but not non-irritants. In addition to phospho-c-Src Y419 , increased abundance of other phosphorylated kinases including phospho-glycogen synthase kinase-3 (GSK3)α/β and phospho-p53 were also markedly increased in the array compared with both non-irritants and water control ( Figure 1).
Phosphorylated heat shock proteins 27 and 60 were abundant in the control, LA and MP but not Co-DEA ( Figure 1). The majority of kinases displayed no difference in phosphorylation status between treatments.
Phospho-specific immunoblot analysis for c-Src Y419 confirmed the array data, showing significantly increased abundance of phospho-c-Src Y419 for the irritant LA, and even more so in response to cinnamaldehyde (CIN) and capsaicin (CA) stimulation, when compared with MP, Co-DEA and water control. However, this difference was not as apparent as observed in the array, with the non-irritant MP displaying levels of phospho-c-Src Y419 similar to the control and Co-DEA displaying similar levels to LA ( Figure 1J,K). This is likely due to differences in the binding affinity of the two anti-phospho-c-Src specific antibodies used in the two immunoblot methods.
To our knowledge, this is the first observation of c-Src-induced phosphorylation in response to irritants by HSE. Moreover, phosphorylation was at tyrosine 419, the main site of phosphorylation within the activation loop that results in Src autophosphorylation and activation status. 9 c-Src is a ubiquitously expressed non-receptor protein tyrosine kinase that is phosphorylated and activated by other protein kinases (e.g., activated epidermal growth factor receptor, adhesion and cytokine receptors as well as several G-proteincoupled receptors). Upon activation c-Src acts by phosphorylating other proteins involved in regulating cell morphology, adhesion, motility, apoptosis, proliferation and survival. 10 There is currently no evidence for a role of c-Src in skin irritancy, although this signalling molecule does appear to be important in hyperproliferative epidermal disorders and epidermal wound healing. 11,12 Histological analysis of HSE sections showed no difference in skin structure between the treatments ( . Densitometry analysis of array immunoblots showing fold-change abundance of phosphokinases in LA, MP or CO-DEA treated HSE relative to water controls for Src Y419 (E), GSK3α/β S21/S9 (F), p53 S15 (G), p53 S46 (H) and c-Jun S63 (I) (n = 2 independent experiments). Immuno-blot analysis of HSE for phospho-c-Src Y419 abundance compared with total c-Src upon stimulation with water control (con), LA, MP, Co-DEA, CIN or CAP for 15 min (J). Densitometric analysis showing fold change in phospho-c-Src Y419 relative to total c-Src (K). Data are mean ± SD for n = 3 independent experiments *p < 0.05, **p < 0.01 by one-way ANOVA with Tukey's multiple comparison post hoc test. The map accompanying the phosphokinase array is shown in Figure S2. The advantage of searching for up-stream phosphorylated signal transduction targets is that activation is quick -within minutes; additionally, the activity is largely restricted to the epidermis so the cell lysate can be subjected to rapid testing formats such as immuno-based rapid antigen tests, rather than rely of complex multi-gene panel analysis. Further screening against a large panel of irritant/non-irritant chemicals is now required to verify these findings.

| CON CLUS I ON S AND PER S PEC TIVE S
This study demonstrates that specific kinases, in particular c-Src, are activated in HSE in response to chemical irritants. c-Src has been im-