Role of hair follicles in the pathogenesis of arsenical-induced cutaneous damage

Abstract

Arsenical vesicants cause skin inflammation, blistering, and pain. The lack of appropriate animal models causes difficulty in defining their molecular pathogenesis. Here, Ptch1^+/- /C57BL/6 mice were employed to investigate the pathobiology of the arsenicals lewisite and phenylarsine oxide (PAO). Following lewisite or PAO challenge (24 h), the skin of animals becomes grayish-white, thick, leathery, and wrinkled with increased bi-fold thickness, Draize score, and necrotic patches. In histopathology, infiltrating leukocytes (macrophages and neutrophils), epidermal-dermal separation, edema, apoptotic cells, and disruption of tight and adherens junction proteins can be visualized. PCR arrays and nanoString analyses showed significant increases in cytokines/chemokines and other proinflammatory mediators. As hair follicles (HFs), which provide an immune-privileged environment, may affect immune cell trafficking and consequent inflammatory responses, we compared the pathogenesis of these chemicals in this model to that in Ptch1^+/- /SKH-1 hairless mice. Ptch1^+/- /SKH-1 mice have rudimentary, whereas Ptch1^+/- /C57BL/6 mice have well-developed HFs. Although no significant differences were observed in qualitative inflammatory responses between the two strains, levels of cytokines/chemokines differed. Importantly, the mechanism of inflammation was identical; both reactive oxygen species induction and consequent activation of unfolded protein response signaling were similar. These data reveal that the acute molecular pathogenesis of arsenicals in these two murine models is similar.

Keywords: animal model; arsenicals; hair follicle; skin injury; vesicants.

Figure 1.

Cutaneous toxicity assessment of lewisite in Ptch1^+/−/C57BL/6 haired mice. (A) Time-dependent (0–24 h) progression of skin damage assessed by gross changes. Black arrows indicate necrotic patches over the lewisite-challenged skin area. (B) Time-dependent alterations in skin bi-fold thickness and Draize score. (C) Histopathology (H&E staining) of the skin showing microvesication (MV) and infiltratory cell infiltrations after 24 h of lewisite exposure. Arrows indicate MV formation. Scale bar: 25 μm. (D) Histogram showing quantitative analysis of MVs (number and size) at 24 h. (E) Pie chart showing size distribution (%) of MVs after 24 h of lewisite exposure. (F) Model showing important protein candidates involved in regulation of tight junctions and adherens junctions. (G) Immunofluorescence staining of tight junction proteins (ZO-1, ZO-2 and claudin-4) and (H) adherens junction proteins (α-E-cat and Yap) in the skin histology from vehicle- and lewisite-treated mice (n = 3–5). Scale bar: 50 μm. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. ns, non-significant.

Figure 2.

Effects of lewisite on the expression of inflammatory mediators in Ptch1+/−/C57BL/6 mice. (A, B) IHC analysis of the macrophage marker F4/80 (A) or the neutrophil marker MPO (B) in vehicle- or lewisite (24 h)-treated mouse skin. Histograms show the quantitative analysis of F4/80- or MPO-positive cells. Several fields of immune cells were photographed, using at least 3–5 individual mice skin samples/group. Scale bar: 50 μm. (C) Western blot and IHC analysis of COX-2 in vehicle- versus lewisite-treated skin sections. Scale bar: 50 μm. (D) ELISA showing protein concentrations of IL-6 and PGE2 in skin samples from the indicated treatment groups. (E) RT-PCR analysis of Il6, Il1b, and Ptgs2 (Cox-2). *p < 0.05, **p < 0.01, and ***p < 0.001. n = 3–5/group. Epi, epidermis; Der, dermis.

Figure 3.

Expression profile of inflammatory genes in the skin of Ptch1^+/−/C57BL/6 mice following lewisite exposure. (A) Table showing significantly up- or down- (>2-fold, p < 0.05) regulated genes identified from a PCR array panel of 84 inflammatory cytokines, chemokines, and receptors. (B) Heat map showing cluster analysis (using the SABiosciences online RT Profiler PCR Array data analysis software) of genes that were upregulated (>2-fold) and (C) downregulated (>2-fold) in the lewisite-treated group compared to vehicle control. (D) RT-PCR analysis of PCR array-identified genes (Cxcl1, Cxcl2, Cxcl5, Cxcl10 and Cxcl15) in vehicle- or lewisite-treated mice skin samples. *p < 0.05 and ****p < 0.0001. (E) nanoString analysis showing the list of upregulated and downregulated differentially expressed genes at the significance level (log2 fold change > |1|, p < 0.05) following exposure of Ptch1^+/−/C57BL/6 mice to lewisite. (F) Venn diagram analysis shows 7 genes (Cxcl10, Itgb2, Il11, Ccl2, Ccr7, Cxcl, and Ccl5) were commonly upregulated in both the nanoString and PCR inflammatory panels. (G) Pathway scores identified by nanoString analysis for various immune cells.

Figure 4.

Comparative analysis of cutaneous damage in hairless Ptch1^+/−/SKH-1 and haired shaved Ptch1^+/−/C57BL/6 mice following PAO exposure. (A) Photographs showing skin of hairless and haired (shaved) mice. Yellows arrows indicate mild skin abrasion due to shaving of haired mice. (B) H&E-stained skin sections from Ptch1^+/−/SKH-1 and Ptch1^+/−/C57BL/6 mice showing the basal level of resident and infiltrated inflammatory cells (indicated by black arrows). Scale bar: 25 μm. (C) RT-PCR mRNA expression analysis of Cxcl10, Cxcl13, Ptgs2 (Cox-2), and Il6 in haired (shaved) and hairless mice at the basal level. (D) (I) Luminex-based protein multiplex assay showing basal levels of various cytokines/chemokines in the skin of Ptch1^+/−/SKH-1 (hairless) and Ptch1^+/−/C57BL/6 (shaved haired) mice. (II) ELISA showing basal level of PGE2 in hairless versus haired (shaved) mice. (E) Time-dependent (0–24 h) changes of skin appearance in haired (shaved) and hairless mice following their exposure to PAO. (F) H&E stained skin sections from haired (shaved) and hairless mice showing infiltration of inflammatory cells in the dermal region. Scale bar:100 μm. (G) Skin bi-fold thickness and Draize score showing nonsignificant changes between the two mouse strains at various time points following PAO exposure. (H) RT-PCR analysis of Il6, Mmp3, Ccl12, Cxcl13, and Cxcl10 in haired and hairless mice at 24 h after PAO exposure. *p < 0.05, **p < 0.01, and ***p < 0.001. ns, not significant; Epi, epidermis; Der, dermis.

Figure 5.

Flow cytometry analysis of immune cell populations in the skin of Ptch1^+/−/SKH-1 and Ptch1^+/−/C57BL/6 mice following PAO exposure. Histograms showing flow cytometry analysis of immune cell infiltration, including (A) infiltrating neutrophils, (B, C) infiltrating monocytes, (D, E, F and G) infiltrating macrophages, (H) M1 macrophages, and (I) M2 macrophages in the skin of Ptch1^+/−/SKH-1 and Ptch1^+/−/C57BL/6 mice exposed to either vehicle or PAO for 3 and 24 h. (n = 4). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. ns, non-significant; C, control; P, PAO.

Figure 6.

Lewisite-induced UPR and apoptosis are not significantly different in the skin of mice with or without developed hair follicles. (A) Bar chart showing relative fluorescence unit of ROS generation in control versus lewisite-treated mouse skin samples obtained from haired (shaved) and hairless strains at 24 h. (B) RT-PCR analysis showing relative mRNA expression of ATF4 and CHOP in control versus lewisite (24 h)-exposed skin of Ptch1^+/−/SKH-1 and Ptch1^+/−/C57BL/6 mice. (C) IHC analysis of ATF4 and CHOP expression and their nuclear localization in the epidermal (Epi) cells of the skin following lewisite exposure in both mouse strains. Magnification = 40X. (D) Western blot analysis of ATF4, CHOP, and cleaved caspse-3 in the skin lysates of control versus lewisite-treated mouse strains. (E) Representative images showing TUNEL staining in the skin of Ptch1^+/−/SKH-1 and Ptch1^+/−/C57BL/6 mice treated with either vehicle control or lewisite for 24 h. Bar chart showing quantitative analysis of green TUNEL-positive epidermal cells. n = 3–5 mice/group. Data are expressed as means ± SEM. *p < 0.05 and ****p < 0.0001

Figure 7.

Lewisite-induced blood arsenic levels in haired (shaved) and hairless mice. Histogram showing ICP-MS analysis of arsenic levels in the whole blood of Ptch1^+/−/SKH-1 or Ptch1^+/−/C57BL/6 mice exposed to lewisite for 24 h. ns, non-significant between two strains; L, lewisite.