Fig 1.
Ligands used for protein-ligand interaction analysis.
The IUPAC name, structure, molecular weight and PubChem CID is provided for the ligands.
Fig 2.
Interaction of dietary flavones with calf thymus (CT) DNA.
(A) UV-Vis spectra of CT-DNA, with 5-aza-2’deoxycytidine, (B) Apigenin, (C) Chrysin, and (D) Luteolin. UV-Vis spectra of CT-DNA with flavones were performed in 0.1 M phosphate buffer solution, with pH 7.4. 0.25 μM DNA solutions were prepared with flavones at 0.5 mM concentration. Absorption spectra of the solutions were recorded from 230 nm to 500 nm using Nanodrop. The experiment was repeated three times with similar results. Details are described in the Materials and Methods section.
Fig 3.
Interaction of dietary flavones with AT-rich and GC-rich DNA sequences.
(A) UV-Vis spectra of AT-rich and GC-rich DNA sequences, with 5-aza-2’deoxycytidine, (B) Apigenin, (C) Chrysin, and (D) Luteolin. 100 bp highly GC-rich oligos mimicking CpG islands ranging from -183 to-83 and AT-rich oligos from the -486 to-386 region of the GSTP1 promoter were synthesized and used as a DNA substrate (2.5 mM) for binding with flavones at a 0.5 mM concentration. Absorption spectra of solutions were recorded from 220 nm to 500 nm using Nanodrop. The experiment was repeated three times with similar results. Details are described in the Materials and Methods section.
Fig 4.
Molecular modeling of the interaction between dietary flavones and DNMT1.
(A) 5-aza-2’deoxycytidine (in sticks) docked into the pocket of DNMT1 (shown as surface), (B) Apigenin, (C) Chrysin, and (D) Luteolin (Left panel). Schematic representation of different non-bonded interactions between ligands and amino acid residues of DNMT1 is shown in the right panel. The pink arrow represents a hydrogen bond and amino acids are colored according to their chemical characteristics. The Glide score for docking of 5-Aza-dC, Apigenin, Luteolin, and Chrysin was -7.16 Kcal/mol, -6.38 Kcal/mol, -5.85 Kcal/mol, and -7.34 Kcal/mol, respectively. Details are described in the Materials and Methods section.
Fig 5.
Effect of dietary flavones on DNA methyltransferase activity and in vitro methylation.
(A) Dose-dependent inhibition of DNMT activity by 10- μM and 20-μM concentrations of 5-Aza-dC, Apigenin, Chrysin and Luteolin. The results are the mean of 3–4 determinations and were analyzed with a one way ANOVA, bars ± SD. **P<0.001 (B) Inhibition of purified recombinant DNA methyltransferase activity by dietary flavones. A 728 bp fragment (-428/+243 relative to the initiation codon) isolated from RWPE-1 cells within the promoter region of the human GSTP1 gene was used as substrate DNA. The methylation reactions were carried out in 1X M.SssI buffer with 160 mM SAM S-adinosylmethionine. The addition of 20 μg of flavones resulted in a detectable decrease in DNA methyltransferase activity (Lanes 4–6), compared to controls (Lane 2). No inhibitory affect was observed with an equal concentration of 5-Aza-dC. (C) MS-PCR for GSTP1 promoter on genomic DNA isolated from RWPE1 cells after treatment with 20 μM of 5-Aza-dC and dietary flavones for 48 h and treatment with methylation using M.SssI. The products generated with primers specific for unmethylated GSTP1 CpG island alleles (U) and for hypermethylated GSTP1 CpG island alleles (M) are displayed. L, DNA ladder. Details are described in the Materials and Methods section.
Fig 6.
Molecular modeling of the interaction between dietary flavones and EZH2.
(A) 3-Deazaneplanocin A (in sticks) docked into the active pocket of EZH2 (shown as surface), (B) Apigenin, (C) Chrysin, and (D) Luteolin (Left panel). Schematic representation of different non-bonded interactions between ligand and amino acid residues of EZH2 is shown in the right panel. The pink arrow represents a hydrogen bond and amino acids are colored according to their chemical characteristics. The Glide scores for docking of DZNep, Apigenin, Luteolin, and Chrysin were -7.62 Kcal/mol, -10.07 Kcal/mol, -9.73 Kcal/mol, and -11.23 Kcal/mol, respectively. Details are described in the Materials and Methods section.
Fig 7.
Effect of dietary flavones on EZH2 catalytic activity and EZH2 and H3K27me3 protein expression.
(A) Dose-dependent inhibition of EZH2 catalytic activity measured through H3K27me3 with 10- μM and 20-μM concentrations of DZNep, Apigenin, Chrysin and Luteolin. The results are the mean of 3–4 determinations and were analyzed with a one way ANOVA, bars ± SD. **P<0.001. (B) Inhibition of EZH2 and H3K27me3 protein expression after treatment with 20 μM DZNep, Apigenin, Chrysin and Luteolin for 48 h. Anti-β-actin and anti-histone H3 were used as loading controls.
Fig 8.
Proposed model of the dual action of dietary flavones in inhibiting DNA methyltransferase and histone methyltransferase.
(A) The two major epigenetic mechanisms, DNA methylation and histone methylation, acted in concert to regulate gene transcription. The DNA double helix backbone is shown in blue and red. In DNA methylation, methyl groups are added to a cytosine that is immediately 5′ to a guanine. In general, an increase in DNA methylation leads to a decrease in gene transcription. Histone methylation is accomplished by trimethylation of histone H3 lysine 27 (H3K27me3), which is catalyzed by the enhancer of zeste homolog 2 enzyme (EZH2) resulting in gene transcriptional repression; occurring to the tails of histones (histones shown in brown, with green H3K27 marks on tails), leading to a condensed chromatin state. (B) Dietary flavones viz. Apigenin, Chrysin and Luteolin can bind to the DNA bases, dock on the catalytic pocket of DNMT and HMT to inhibit methylation resulting in relaxed chromatin, increasing the likelihood of gene transcription.