CYP24A1 is overexpressed in keloid keratinocytes and its inhibition alters profibrotic gene expression

Abstract

Background: Keloids are disfiguring, fibrotic scar-like lesions that are challenging to treat and commonly recur after therapy. A deeper understanding of the mechanisms driving keloid formation is necessary for the development of more effective therapies. Reduced vitamin D receptor (VDR) expression has been observed in keloids, implicating vitamin D signaling in keloid pathology. Vitamin D exhibits anti-proliferative and anti-inflammatory properties, suggesting it could have therapeutic utility in keloid disorder. The current study investigated vitamin D-regulated gene expression in keloid keratinocytes and the effects of inhibiting an enzyme involved in vitamin D metabolism on the phenotype of keloid-derived keratinocytes.

Methods: Normal and keloid-derived primary keratinocytes were isolated from normal skin and keloid lesions, respectively, and were cultured in the absence or presence of vitamin D. In some experiments, inhibitors of the vitamin D metabolizing enzyme CYP24A1, ketoconazole or VID400 were added in the absence or presence of vitamin D. Cellular proliferation, migration and gene expression were measured.

Results: We observed significant overexpression of CYP24A1 mRNA in keloid versus normal keratinocytes and increased CYP24A1 protein levels in keloids versus normal skin. CYP24A1 encodes 24 hydroxylase and is induced by vitamin D in a feedback loop that regulates vitamin D levels; thus, inhibition of CYP24A1 activity may locally increase active vitamin D levels. Ketoconazole, a non-specific cytochrome P-450 inhibitor, reduced proliferation of keloid and normal keratinocytes, but VID400, a specific CYP24A1 inhibitor, only significantly affected keloid keratinocyte proliferation. Neither inhibitor significantly reduced keratinocyte migration. The two inhibitors had different effects on vitamin D target gene expression in keratinocytes. Specifically, ketoconazole treatment reduced CYP24A1 expression in normal and keloid keratinocytes, whereas VID400 increased CYP24A1 expression. Both inhibitors decreased expression of profibrotic genes, including periostin and hyaluronan synthase 2, in keloid-derived cells. Combined treatment of keloid keratinocytes with vitamin D and ketoconazole or VID400 increased the effects of vitamin D treatment on target genes, although the effects were gene- and cell type-specific.

Conclusions: The data suggest that reduction of vitamin D inactivation with CYP24A1 inhibitors may reduce profibrotic gene expression in keloid-derived cells. Therefore, CYP24A1 inhibitors may serve as adjunctive therapies to suppress keloid-associated gene expression changes.

Keywords: 24-hydroxylase; CYP24A1; Epithelial-mesenchymal transition; Keloid; Keratinocyte; Vitamin D.

Figure 1

Proliferation of normal and keloid-derived keratinocytes treated with vitamin D. Normal keratinocytes (NK, white bars; n = 3 strains) and keloid keratinocytes (KK, black bars; n = 4 strains) were cultured in the presence of vehicle or active vitamin D (0–100 nM) for 24 h, and proliferation was measured using an MTT assay. Absorbance at 570 nm was measured, and data are plotted as means ± standard deviations. Absorbance values are proportional to cell number, with higher values reflecting greater proliferation levels. Although a trend toward decreasing proliferation with increasing vitamin D concentrations was observed, no statistically significant differences were found among different vitamin D concentrations or between normal and keloid keratinocyte groups. MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

Figure 2

Vitamin D-stimulated gene expression in normal and keloid keratinocytes. Normal keratinocytes (left) and keloid-derived keratinocytes (right) respond to vitamin D treatment (100 nM for 24 h) with significant upregulation of target genes CYP24A1, cathelicidin antimicrobial peptide (CAMP), and CD14. Expression levels in vitamin D-treated cells (black bars) were normalized to values for untreated keratinocytes (control; white bars) of the same type. Mean values ± standard deviations are plotted (n = 6 for normal keratinocytes and n = 9 for keloid keratinocytes). Asterisks indicate statistically significant differences (^*p < 0.05; ^**p < 0.01; ^***p < 0.001)

Figure 3

Expression of VDR and genes encoding enzymes involved in vitamin D metabolism in normal and keloid keratinocytes. Gene expression in normal keratinocytes (NK; n = 6 strains, white bars) and keloid-derived keratinocytes (KK; n = 9 strains, black bars) was analyzed by qPCR. Results were normalized to the mean expression in normal keratinocytes and data are plotted as means ± standard deviations. Statistical comparisons between normal and keloid keratinocytes were analyzed via t test; only CYP24A1 expression was significantly different (^**p < 0.01) between normal and keloid-derived cells. VDR Vitamin D receptor

Figure 4

Expression of CYP24A1 protein in keloids and normal skin samples. Immunohistochemistry was used to localize CYP24A1 in sections of keloids (a–c) and normal skin (g–i). Negative controls were processed identically but without primary antibody (d–f for keloid samples; j–l for normal skin samples). Levels of CYP24A1 were variable among samples within each group, keloid or normal skin, but staining levels were higher overall in the keloid samples; representative sections illustrating this variability are shown. Keloid samples from donors 843K (a, d), 934K (b, e), and 991K (c, f) are shown, as are normal skin samples from donors 880 (g, j), 886 (h, k) and 974 (i, l); see Table 1 for donor demographic information. The scale bar in panel l (0.1 mm) is the same for all images

Figure 5

Effects of CYP24A1 inhibition, without or with vitamin D treatment, on keratinocyte proliferation. Normal keratinocytes (NK; white bars; n = 3 strains) and keloid keratinocytes (KK; black bars; n = 4 strains) were cultured in the absence or presence of ketoconazole, a non-specific CYP24A1 inhibitor (a), or VID400, a specific CYP24A1 inhibitor (b), with or without vitamin D (c), and proliferation was measured via MTT assay. Absorbance at 570 nm was measured, and data are plotted as means ± standard deviations. Statistical analyses were performed using one-way ANOVA, and pairwise comparisons were analyzed using a post hoc Tukey test. (a) Normal and keloid keratinocytes were cultured for 24 h with ketoconazole at concentrations ranging from 0 to 30 μM, as indicated. A statistically significant dose–response relationship was observed in both normal and keloid keratinocytes. (b) Normal and keloid keratinocytes were cultured for 24 h with VID400 at concentrations ranging from 0 to 400 nM, as indicated. Statistically significant differences were only observed for keloid keratinocytes. (c) Normal and keloid keratinocytes were treated for 24 hs with 100 nM vitamin D, ±10 μM ketoconazole or 100 nM VID400, as indicated. No statistically significant differences were observed among any of the normal keratinocyte groups, although the difference between untreated and ketoconazole-treated normal keratinocytes approached statistical significance (^#p = 0.05). In keloid keratinocyte groups, ketoconazole treatment caused a significant reduction in keloid keratinocyte proliferation in the absence or presence of vitamin D. Statistically significant differences are indicated in each plot by asterisks (^*p < 0.05; ^**p < 0.01; ^***p < 0.001). ANOVA Analysis of variance

Figure 6

Effects of CYP24A1 inhibition and vitamin D on keratinocyte migration. Keratinocytes were cultured in the absence (non-patterned bars) or presence (patterned bars) of 100 nM vitamin D, and in the absence or presence of CYP24A1 inhibitors ketoconazole or VID400 (red bars). Migration was measured using an in vitro scratch wound assay at 6 and 24 h after wounding, and the percent of the wound closed was calculated at each time point. Data are plotted as means ± standard deviations. Statistical analyses were performed using one-way ANOVA, and pairwise comparisons were analyzed using a post hoc Tukey test. (a) In normal keratinocytes (n = 3 donor strains), neither inhibitor had a significant effect on cell migration at 6 or 24 h in the absence of vitamin D. Vitamin D treatment added with either ketoconazole or VID400 significantly increased migration rates at 24 h compared to either inhibitor alone. Additionally, VID400 + vitamin D significantly increased migration compared with untreated control cells. Statistically significant differences are indicated by asterisks (^*p < 0.05; ^**p < 0.01). (b) In keloid keratinocytes, no significant differences in migration among groups were observed at either time point. However, there was greater variability in migration rates among the different keloid keratinocyte cell strains, with correspondingly large standard deviations, which may have impacted this analysis. ANOVA analysis of variance

Figure 7

Gene expression in normal and keloid keratinocytes in response to vitamin D and CYP24A1 inhibition. Keratinocytes were cultured in the absence (non-patterned bars) or presence (patterned bars) of 100 nM vitamin D, and in the absence or presence of CYP24A1 inhibitors ketoconazole (gray bars) or VID400 (red bars); see legend (bottom right of figure) for details. Gene expression was analyzed by quantitative PCR, and expression levels were normalized to expression in control, untreated cells of the same type, normal or keloid, as indicated (white, open bars). Plotted are mean normalized expression levels ± standard deviations for CYP24A1 (a), VDR (b), CAMP (c), CD14 (d), EDN1 (e), FZD7 (f), WNT5A (g), MMP1 (h), MMP3 (i), POSTN (j) and HAS2 (k). Statistical comparisons within each cell type for each gene (normal keratinocytes, left; keloid keratinocytes, right) were performed using one-way ANOVA with post hoc Tukey test, and significant differences are indicated by asterisks: ^*p < 0.05; ^**p < 0.01; ^***p < 0.001. ANOVA Analysis of variance, CAMP Cathelicidin antimicrobial peptide, EDN1 Endothelin 1, FZD7 Frizzled 7, HAS2 Hyaluronan synthase 2, MMP1 Matrix metallopeptidase 1, MMP3 Matrix metallopeptidase 3, POSTN Periostin

Figure 8

Production of vitamin D in the skin. Vitamin D3 (D3) is synthesized in keratinocytes from the precursor 7-dehydrocholesterol in response to ultraviolet radiation in sunlight. D3 is metabolized to 25-hydroxy vitamin D (25(OH)D), the form most commonly measured in the blood as a marker of vitamin D status, by CYP27A1. 25(OH)D is metabolized to 1,25-dihydroxyvitamin D (1,25(OH)₂D), the hormonally active form of vitamin D, by CYP27B1. The enzyme CYP24A1 inactivates both 25(OH)D and 1,25(OH)₂D. Transcription of the gene encoding CYP24A1 is induced by 1,25(OH)₂D, which serves as a feedback loop to modulate levels of active vitamin D. The inhibitors ketoconazole and VID400 block the activity of CYP24A1, which may lead to decreased profibrotic gene expression by decreasing inactivation of 1,25(OH)₂D