Ultraviolet-B induces ERCC6 repression in lens epithelium cells of age-related nuclear cataract through coordinated DNA hypermethylation and histone deacetylation

Abstract

Background: Ultraviolet-B (UVB) exposure attributes to the formation of age-related nuclear cataract (ARNC), which is mediated with DNA damage. DNA damage, an important factor for pathogenesis of ARNC, is induced by UVB, and is generally resolved by the nucleotide excision repair (NER) repair mechanism. Cockayne syndrome complementation group B (CSB) protein coded by ERCC6 is a vital component for NER. However, we found no association between selected ERCC6 polymorphisms and ARNC. In this study, we investigated whether UVB exposure could alter ERCC6 expression and the process could involve epigenetic changes of DNA methylation and/or histone acetylation of ERCC6 in the lens epithelial cells (LECs). We also assessed the involvement of those coordinated changes in lens tissue from ARNC patients.

Results: mRNA and protein expression of ERCC6 in lens tissue (LECs) were lower in ARNCs than those in the controls. This reduction corresponded to methylation of a CpG site at the ERCC6 promoter and histone modifications (methylation and acetylation) nearby this site. UVB-treated human lens epithelium B3 (HLE-B3) and 239T cell presented (1) increased apoptosis, suggesting reduced UV-damage repair, (2) hypermethylation of the CpG site located at position -441 (relative to transcription start site) within the binding region for transcriptional factor Sp1 in the ERCC6 promoter, (3) the enhancement of histone H3K9 deacetylation, (4) induction in DNA methyltransferases 3b (DNMT3b) and histone deacetylase1 (HDAC1) associated to the CpG site of ERCC6 by CHIP assay.

Conclusions: These findings suggest an orchestrated mechanism triggered by UVB radiation where the concurrent association of specific hypermethylation CpG site, H3K9 deacetylation of ERCC6, and repression of ERCC6 gene expression. Taken together, with the similar changes in the lens tissue from ARNC patients, our data unveiled a possible mechanism of epigenetic modification of DNA repair gene in the pathogenesis of ARNC.

Keywords: Age-related nuclear cataract (ARNC); DNA methylation; ERCC6; Histone deacetylation; Lens epithelial cells (LECs); Sp1; Ultraviolet (UV).

Fig. 1

Relative expression of mRNA and protein levels of ERCC6 in LECs of controls and ARNCs. a qRT-PCR analysis of the mRNA expression of ERCC6 in LECs of controls and ARNCs. The mRNA levels were normalized by using the GAPDH as the inner control. Data were depicted as the mean ± SD of three independent experiments. *P < 0.01. b The amount of ERCC6 protein in LECs of controls and ARNCs was measured by Western bolt analysis (Samples labeled as #1–#15 of the controls and #1–#15 of the ARNCs out of total 30 samples in each group). c Relative ERCC6 protein level to GAPDH is presented as mean ± SD. *P < 0.01

Fig. 2

Methylation status of the CpG island at promoter of ERCC6 in LECs of controls and ARNCs (30 controls and 30 ARNCs, the samples cover those analyzed in Fig. 1). a The CpG island (-603/-396, black arrow) located in the promoter of ERCC6. There are twelve CpG sites (the red marker letter) and a putative binding site for Sp1 (-446/-437) relative to transcriptional starting site (red arrow) in the region. b In LECs of ARNCs, the CpG site 8 displayed hypermethylation compared to the controls. *P < 0.01. c Representative pyrosequencing results of the ARNC. d Representative pyrosequencing results of the control

Fig. 3

Identification of ERCC6 minimal promoter. a Schematic illustration of deletion constructs of ERCC6 proximal promoter pGL3 -603/-396, Met-pGL3 -466/-396, pGL3 control, and pGL3 enhancer. b Luciferase activity of the deleted constructs in 293T cells. Data are expressed as mean ± SD of three independent experiments. The thick lines represent retained parts of overall length (from -466 to -396, relative to the TSS). The dash lines represent the removed parts of overall length. The thin lines only represent the link to plasmid. Met-pGL3 -466/-396 represents methylated pGL3 -466/-396 without changing the sequences

Fig. 4

Relative expression of mRNA and protein level of DNMTs in LECs of controls and ARNCs. a qRT-PCR analysis of the mRNA expression of DNMTs in LECs of controls and ARNCs. The mRNA levels were normalized by using the GAPDH as the inner control. Values represent mean ± SD. *P < 0.01. b Western bolt analysis for protein level of DNMT3b in LECs of controls and ARNCs (Samples labeled as #1–#15 of the controls and #1–#15 of the ARNCs out of total 30 samples in each group). c Relative DNMT3b protein level to GAPDH is presented as mean ± SD. *P < 0.01

Fig. 5

The staining changes of ERCC6 and DNMT3b immunoreactivity in LECs of controls and ARNCs (Samples labeled as #16–#30 of the controls and #16–#30 of the ARNCs out of total 30 samples in each group). The LECs of controls showed low levels of DNMT3b and high levels of ERCC6 staining (a, c). DNMT3b immunoreactivity increased significantly and the ERCC6 immunoreactivity decreased significantly in LECs of ARNCs (b, d). Bar graph illustrated the immunoreactivity of ERCC6 and DNMT3b in LECs of controls and ARNCs (e). *P < 0.01. Scale bars: 50 μm

Fig. 6

Protein level and promoter methylation status (site 8) of ERCC6 in HLE-B3 after treatment with UVB exposure. a 12 h after treatment with UVB exposure, the promoter of ERCC6 in the cells displayed hypermethylation compared to the control cells. *P < 0.01. b Protein levels of ERCC6 in control cells and in cells after treatment with UVB exposure. c Relative ERCC6 protein level to GAPDH is presented as mean ± SD. *P < 0.01. d Protein level of active caspases-3 in control cells and HLE-B3 cells after expose to UVB was measured by Western bolt analysis. e Relative active caspases-3 protein level to GAPDH is presented as mean ± SD. *P < 0.01

Fig. 7

Relative mRNA expression and protein levels of ERCC6 in HLE-B3 after treatment with 5-aza-dC. a mRNA expression of ERCC6 in control cells and in cells after treatment with 5-aza-dC for 48 h. b Protein level of ERCC6 in control cells and in cells after treatment with 5-aza-dC. c Relative ERCC6 protein level to GAPDH is presented as mean ± SD. *P < 0.01

Fig. 8

Electrophoretic mobility shift detected the DNA-binding ability of Sp1 on the ERCC6 promoter. The DNA-protein complex (indicated by black arrow) can be observed (lanes 3, 4, 5, 7, and 9). In contrast, no nucleoprotein complex was observed when used the M-probe and labeled mutated type probe (lanes 1 and 8). The competition assay was done by the addition of 50-, 100-, or 200-fold molar excess of unlabeled methylated probes to the incubation mixtures (lanes 4, 5, and 6). The complex formation was fully suppressed by the addition of a 200-fold molar excess of unlabeled wild type probe (lane 6) and not was suppressed by 200-fold molar excess of unlabeled mutated type probe (lane 9). A supershift band (indicated by black arrowhead, lane 7) was detected when anti-Sp1 antibody was incubated

Fig. 9

Knockdown of endogenous Sp1 decreased the mRNA expression and protein level of ERCC6 and Sp1 in 239T cells. a Detection of ERCC6 and Sp1 mRNA expression by qRT-PCR. Values represent mean ± SD. *P < 0.01. b Protein level of ERCC6 and Sp1 was examined by Western blot analysis. Results are presented as mean ± SD. c Relative ERCC6 protein level to GAPDH is presented as mean ± SD. *P < 0.01

Fig. 10

UVB irradiation decreased the occupancy of endogenous Sp1 and H3K9 acetylation on the ERCC6 promoter, but increased occupancy of DNMT3b and HDAC1 on the ERCC6 promoter. a HLE-B3 cells were treated with UVB irradiation. ChIP was performed using anti-Sp1, anti-DNMT3b, anti-HDAC1, and anti-ac-H3 (K9). b Statistically significant differences from control cells were indicated by *(P < 0.01)

Fig. 11

Relative mRNA expression and protein levels of ERCC6 in HLE-B3 after treatment with MS275. a Detection of ERCC6 mRNA expression by qRT-PCR at different time point. b Protein level of ERCC6 was examined by Western blot analysis at different time point. c Relative ERCC6 protein level to GAPDH is presented as mean ± SD. *P < 0.01