Identification of Novel Melanin Synthesis Inhibitors From Crataegus pycnoloba Using an in Vivo Zebrafish Phenotypic Assay

Abstract

Zebrafish has emerged as a powerful model organism for high throughput drug screening. Several morphological criteria, transgenic lines and in situ expression screens have been developed to identify novel bioactive compounds and their mechanism of action. Here, we used the inhibition of melanogenesis during early zebrafish embryo development to identify natural compounds that block melanogenesis. We identified an extract from the Greek hawthorn Crataegus pycnoloba as a potent inhibitor of melanin synthesis and used activity based subfractionation to identify active subfractions and eventually three single compounds of the same family (dibenzofurans). These compounds show reversible inhibition of melanin synthesis and do not act via inhibition of tyrosinase. We also showed that they do not interfere with neural crest differentiation or migration. We identified via in silico modeling that the compounds can bind to the aryl hydrocarbon receptor (AHR) and verified activation of the Ahr signaling pathway showing the induction of the expression of target genes.

Keywords: Crataegus; dibenzofuran; melanin synthesis inhibitors; phenotype-driven screens; zebrafish.

FIGURE 1

Screening for inhibitors of melanogenesis from Crataegus pycnoloba extracts. Crude extract (A) and fractions (C–J) were tested for anti-pigmentation ability by in vivo zebrafish assay. Synchronized 24 hpf embryos were treated with 0.04 mg/mL of the crude extract and 0.1 mg/mL (CPC-Fr6-8) or 0.01 mg/mL (CPC-Fr9-13) from the different fractions for 24 h. Images were obtained at 48 hpf using a stereomicroscope. Test compounds were dissolved in DMSO. (B) Control treated embryo. Scale bar: 100 μM.

FIGURE 2

Effects of isolated compounds from the dibenzofuran family on melanin synthesis in zebrafish. (A) Embryos were treated from 24 to 48 hpf with 0.01 mg/mL of the compounds and the effects on the pigmentation were assessed using a stereomicroscope. (B) Melanin content was quantified by a photometric method. PTU was used as positive control. Results shown are the mean of three independent experiments ± SEM. ^∗p < 0.001, versus DMSO control.

FIGURE 3

Structures of polychlorinated biphenyls (PCBs). 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and the five compounds employed in this study. Compounds 1–4 are the dibenzo-p-furan derivatives isolated from the total C. pycnoloba extract, including the atom numbering of the newly discovered compound 2. Compound 5 is ursolic aldehyde, isolated from the active fragments of C. pycnoloba but has no melanin synthesis inhibition activity.

FIGURE 4

Dibenzofuran compounds inhibit melanogenesis in a dose dependent manner. Synchronized embryos were treated with test compounds at the indicated concentrations from 24 hpf for 24 h. The effects on the pigmentation of zebrafish were observed at 48 hpf under the stereomicroscope. Treatment with 0.002 mg/mL had no effect on pigmentation for any of the tested compounds (A,D,G,J). Mild effect on melanogenesis was observed at 0.005 mg/mL (B,E,K). At 0.01 mg/mL there was a strong inhibition of melanogenesis (C,F,L). Compound 3 had no effect at any of the tested concentrations (G–I). Scale bar: 100 μM.

FIGURE 5

Isolated compounds from the dibenzofuran family repress reversibly in vivo pigmentation in zebrafish. Embryos were pretreated at 24 hpf with 0.01 mg/mL test compounds for 24 h. At 48 hpf pigmentation levels were recorded (A–F) and the embryos were divided into two groups. One group was washed and bathed immediately in fresh medium, whereas the other group was incubated with dibenzofuran compounds for further 24 h. At 72 hpf, the effect on melanogenesis was assessed using a stereomicroscope for both the compound treatment stopped group (G–L) and the embryos that were continuously treated with the dibenzofuran compounds for 48 h (M–R). PTU (E,K,Q) and DMSO (F,L,R) were used as positive and negative controls, respectively. (S) A schematic representation for the schedule of pigmentation rescue study. Scale bar: 100 μM.

FIGURE 6

Molecular modeling suggests that compounds 1 and 2 can bind to the AHR. (A) Homology model of the PAS-B domain of human AHR in complex with a dibenzofuran derivative (orange C and red O spheres). (B) Close-up view of the internal cavity of the AHR model, illustrating a bound conformation of 2 and the surrounding hydrophobic residues (cyan C, blue N, yellow S). The asterisk indicates position-1 of the dibenzofuran ring.

FIGURE 7

The effect of compounds from the dibenzofuran family on the mRNA expression of Ahr dependent genes. Gene expression analysis from 30 hpf zebrafish embryos treated for 6 h with 50 μg/mL of 1, 4 or 5. Shown are mRNA levels of cyp1b1, aldh3a1, and the ugt1a superfamily, known to be downstream targets of the Ahr. mRNA expression was normalized against Actin, and the gene expression of embryos treated with DMSO (vehicle) was set as 1. Data are mean ± standard error of the mean (SEM), n = 3, ^∗P < 0.005.