Characterization of CYP26B1-Selective Inhibitor, DX314, as a Potential Therapeutic for Keratinization Disorders

Abstract

Inhibition of CYP450-mediated retinoic acid (RA) metabolism by RA metabolism blocking agents increases endogenous retinoids and is an alternative to retinoid therapy. Currently available RA metabolism blocking agents (i.e., liarozole and talarozole) tend to have fewer adverse effects than traditional retinoids but lack target specificity. Substrate-based inhibitor DX314 has enhanced selectivity for RA-metabolizing enzyme CYP26B1 and may offer an improved treatment option for keratinization disorders such as congenital ichthyosis and Darier disease. In this study, we used RT-qPCR, RNA sequencing, pathway, upstream regulator, and histological analyses to demonstrate that DX314 can potentiate the effects of all-trans-RA in healthy and diseased reconstructed human epidermis. We unexpectedly discovered that DX314, but not all-trans-RA or previous RA metabolism blocking agents, appears to protect epidermal barrier integrity. In addition, DX314-induced keratinization and epidermal proliferation effects are observed in a rhino mice model. Altogether, the results indicate that DX314 inhibits all-trans-RA metabolism with minimal off-target activity and shows therapeutic similarity to topical retinoids in vitro and in vivo. Findings of a barrier-protecting effect require further mechanistic study but may lead to a unique strategy in barrier-reinforcing therapies. DX314 is a promising candidate compound for further study and development in the context of keratinization disorders.

Figure 1:. DX314 potentiates atRA gene expression effects in healthy RHE.

(a) Relative expression of HBEGF, IVL, and CYP26A1 mRNA by RT-qPCR. Symbol underneath indicates comparison group (n=3-4 times in duplicate; mean±95% CI; *p≤0.05, **p≤0.01, ***p≤0.001; one-way ANOVA with Tukey’s correction; *vs Control, ^†vs 1nM atRA, ^¥vs 10nM atRA, ^⨂vs DX314-alone, ^⊕vs Liarozole-alone). (b) IVL localization in healthy RHE. (c) RNAseq: Relative mRNA expression of retinoid-responsive genes in healthy RHE. Non-grey cells differ from controls (FDR≤0.05; n=3-5). Adjacent green cell indicates likely RAR-mediated effect based on presence of RARE-promotor for respective gene. Predicted activation z-score of (d) upstream regulators or (e) canonical pathways determined by IPA software utilizing RNAseq data. Black dots indicate statistical insignificance (p≤0.05 and z-score ≥2 or ≤−2). Additional abbreviations: Table S4.

Figure 2:. DX314 potentiates the effects of atRA on CYP26A1 mRNA expression in keratinocytes from individuals with keratinization disorders.

Relative CYP26A1 mRNA expression by RT-qPCR in; (a) Darier disease RHE, (b) recessive x-linked ichthyosis (RXLI) full-thickness RHE, (c) lamellar ichthyosis RHE, and (d) RXLI monolayer keratinocyte cultures. RHE were treated for 4 days and monolayer keratinocytes for 20hrs. Statistical significance was computed with (a) autoscaled or (b-d) raw dCt values. Symbol below each treatment indicates comparison group (n=3 independent replicates with technical duplicates; mean±95% CI; *p≤0.05; **p≤0.01; ***p≤0.001; one-way ANOVA with Tukey’s correction; *vs Control, ^†vs 1nM atRA, ^¥vs 10nM atRA, ^&vs 100nM atRA, ^◆vs 1000nM atRA, ^⊗vs DX314-alone, ^⊕vs Liarozole-alone, ^∅vs Talarozole-alone).

Figure 3:. DX314 potentiates the effects of atRA on the expression and localization of keratin 10 (KRT10) in Darier disease (DD) RHE. atRA, but not DX314, induces a loss of stratum granulosum.

(a) HE staining and (b) immunofluorecent staining of KRT10 (green) localization with nuclear stain (blue), in DD RHE treated for 4 days. Scale bars: black = 20μm, white =50 μm. (c) Relative KRT10 mRNA expression by qPCR. Symbol below each treatment indicates comparison group. (n=3 independent replicates with technical duplicates; mean±95% CI; *p≤0.05; **p≤0.01; ***p≤0.001; one-way ANOVA with Tukey’s correction on autoscaled values; *vs Control, ^†vs 1nM atRA, ^¥vs 10nM atRA, ^&vs 100nM atRA, ^⊗vs DX314-alone, ^∅vs Talarozole-alone).

Figure 4:. DX314 protects barrier function in RHE.

(a) Transepithelial electrical resistance (TEER) in healthy RHE. TEER was normalized to control RHEs for each run, then pooled for analysis. Graph shows Tukey’s boxplot with outliers. Sample sizes (n) are shown above x-axis. (b) LI RHE TEER (top), transepidermal water loss (middle), and the linear correlation between the two measures (bottom), (c) HE staining of lamellar ichthyosis (LI) RHE. Scale bar = 50μm. (d) Semi-quantitative analysis of relative stratum granulosum (SG) surface area in healthy RHE. (e) Relative expression of epidermal differentiation complex (EDC) genes and regulators by RNAseq. Colored (non-grey) cells indicate statistical significance from control (FDR≤0.05; n=3-5). All RHE received a 4-day treatment. (a,b,d) (*p≤0.05; **p≤0.01; ***p≤0.001; one-way ANOVA with Dunnett’s correction vs control).

Figure 5:. DX314 reduces rhino mouse skin abnormalities.

Representative HE staining of skin biopsies from rhino mice topically treated for 11 days with vehicle (acetone) or 1% DX314. ImageJ software was used to quantify comedonal number, profile (d/D, ratio of opening to inner diameter), and epidermal thickness. Epidermal thickness was measured at multiple points across each sample by measuring the sum of epidermal areas (yellow), excluding the corneal layer, and dividing by the sum of the length of the basal layers (dotted blue line). Scale bar =200μm.

Figure 6:. DX314 treatment reduces comedonal number, induces epidermal thickening, and increases comedonal profile, while having no effect on transepidermal water loss (TEWL) in treated rhino mice.

Semi-quantitative analysis of changes in (a) total (open + closed) and (b) open comedonal number, (c) comedonal profile, and (d) epidermal thickness in rhino mice topically treated with vehicle (acetone) or 1% DX314 over 11 days, (e) Daily TEWL measurements did not reveal any statistically significant differences between treatment groups. (n=5-6 mice per treatment; mean±SD; *p≤0.05; **p≤0.01; Student’s t-test vs vehicle control).