Hypoxia and nickel inhibit histone demethylase JMJD1A and repress Spry2 expression in human bronchial epithelial BEAS-2B cells

Abstract

Epigenetic silencing of tumor suppressor genes commonly occurs in human cancers via increasing DNA methylation and repressive histone modifications at gene promoters. However, little is known about how pathogenic environmental factors contribute to cancer development by affecting epigenetic regulatory mechanisms. Previously, we reported that both hypoxia and nickel (an environmental carcinogen) increased global histone H3 lysine 9 methylation in cells through inhibiting a novel class of iron- and α-ketoglutarate-dependent histone demethylases. Here, we investigated whether inhibition of histone demethylase JMJD1A by hypoxia and nickel could lead to repression/silencing of JMJD1A-targeted gene(s). By using Affymetrix GeneChip and ChIP-on-chip technologies, we identified Spry2 gene, a key regulator of receptor tyrosine kinase/extracellular signal-regulated kinase (ERK) signaling, as one of the JMJD1A-targeted genes in human bronchial epithelial BEAS-2B cells. Both hypoxia and nickel exposure increased the level of H3K9me2 at the Spry2 promoter by inhibiting JMJD1A, which probably led to a decreased expression of Spry2 in BEAS-2B cells. Repression of Spry2 potentiated the nickel-induced ERK phosphorylation, and forced expression of Spry2 in BEAS-2B cells decreased the nickel-induced ERK phosphorylation and significantly suppressed nickel-induced anchorage-independent growth. Taken together, our results suggest that histone demethylases could be targets of environmental carcinogens and their inhibition may lead to altered gene expression and eventually carcinogenesis.

Fig. 1.

Identification of JMJD1A-targeted genes in BEAS-2B cells. (A) siRNA efficiently knocks down endogenous JMJD1A. BEAS-2B cells were transiently transfected with siRNA against JMJD1A. Three days after transfection, whole cell lysates were collected and the expression of endogenous JMJD1A was detected by immunoblotting. (B) Venn diagram analysis of the potential JMJD1A target genes. The genes in the yellow circle were found to have JMJD1A-binding peaks close to their gene loci as identified by ChIP-on-chip assay, whereas the genes in the green circle showed >2-fold decrease in gene expression after knockdown of JMJD1A in BEAS-2B cells. (C) Distribution of manually curated gene ontology annotations for potential JMJD1A targets. In some instances, a gene is assigned to more than one category. The percentage refers to the number of genes within any particular gene ontology category in relation to the total number of JMJD1A-targeted genes (N = 68).

Fig. 2.

JMJD1A regulates Spry2 gene expression. (A) Spry2 was downregulated in BEAS-2B cells after knockdown of JMJD1A, as shown by semiquantitative PCR. (B) SPRY2 expression in BEAS-2B cells was decreased after knockdown of JMJD1A as measured by western blot. (C) BEAS-2B cells were transfected with two different siRNAs targeting JMJD1A, and expression of SPRY2 was examined using western blot. (D) Overexpression of JMJD1A increased expression of Spry2. BEAS-2B cells were transiently transfected with the control vector, pcDNA3-Flag-JMJD1A, or its mutant vector (pcDNA3-Flag-JMJD1A H1120Y). Expression of Spry2 was then examined by semiquantitative PCR. (E) Overexpression of JMJD1A increased SPRY2 expression in BEAS-2B cells, as measured by western blot.

Fig. 3.

Hypoxia and nickel exposure decreased Spry2 expression. (A) Hypoxia decreases Spry2 gene expression in BEAS-2B cells. BEAS-2B cells were exposed to hypoxia (1% O₂) for 3 days. Expression of Spry2 was examined using semiquantitative PCR. (B) Hypoxia decreases SPRY2 but increases JMJD1A expression in BEAS-2B cells. BEAS-2B cells were transiently transfected with scramble siRNA or JMJD1A siRNA. One day after transfection, cells were replenished with fresh media and incubated in chambers with either 21 or 1% O₂ for 3 days. Expressions of SPRY2 and JMJD1A in these cells were analyzed with immunoblotting. (C) Nickel decreases Spry2 gene expression in BEAS-2B cells. BEAS-2B cells were exposed to 50 or 100 μM NiCl₂ for 8 weeks and expression of Spry2 was examined using semiquantitative PCR. (D) Nickel decreases SPRY2 but increases JMJD1A expression in BEAS-2B cells. The expression of SPRY2 and JMJD1A was analyzed by western blot in the cells treated with 100 μM NiCl₂ for 8 weeks.

Fig. 4.

The demethylase activity of JMJD1A is essential for its direct regulation of Spry2 expression. (A) Knockdown of JMJD1A increases H3K9me2 level at the Spry2 promoter in BEAS-2B cells. BEAS-2B cells were transiently transfected with JMJD1A or scramble siRNA. Three days after transfection, ChIP analysis was performed using anti-JMJD1A and anti-H3K9me2 antibodies. Spry2 promoter DNA in the immunoprecipitated samples was measured using semiquantitative PCR. (B) Hypoxia increases JMJD1A binding at the promoter of Spry2 gene in BEAS-2B cells. BEAS-2B cells were exposed to hypoxia (1% O₂) for 3 days. ChIP assays were performed to analyze the binding of JMJD1A at the promoter of Spry2 gene. (C) Hypoxia increases H3K9me2 level at the Spry2 promoter in BEAS-2B cells. BEAS-2B cells were exposed under the same conditions as in (B). ChIP analysis was performed using anti-H3K9me2 antibody, and Spry2 promoter DNA in the immunoprecipitated samples was measured using semiquantitative PCR. (D) ChIP assays were performed in nickel-treated BEAS-2B cells (100 μM NiCl₂ for 8 weeks) to analyze the binding of JMJD1A and relative levels of H3K9me2 at the promoter of Spry2 gene.

Fig. 5.

Nickel ions induce JMJD1A expression through HIF1α. (A) Nickel ion exposure and hypoxia fail to induce JMJD1a in HIF1α-deficient MEF cells. HIF1α-proficient (HIF1α^+/+) and -deficient (HIF1α^−/−) MEF cells were exposed to 250 μM NiCl₂ or hypoxia (1% O₂) for 24 h. Expression of JMJD1a was then assessed by immunoblotting. (B) Knockdown of HIF1α blocks the induction of JMJD1A by nickel ions in BEAS-2B cells. Cells were transfected with siRNA against HIF1α and were then exposed to 300 μM NiCl₂ for 24 h. Expression of JMJD1A was then assessed by immunoblotting. (C) Knockdown of HIF1α blocks the induction of Jmjd1a mRNA by nickel ions in BEAS-2B cells. Cells were treated under the same condition as described in (B). The relative amount of Jmjd1a mRNA was measured by quantitative real-time PCR and normalized to that of β-actin. Values are means ± SD for triplicates. (D) Binding of HIF-1 to the promoter of Jmjd1a gene. BEAS-2B cells were treated with 300 μM NiCl₂ for 24 h. ChIP assay was performed to analyze the binding of HIF-1α and HIF-1β ((aryl hydrocarbon receptor nuclear translocator) at the hypoxia response element site in the Jmjd1a promoter.

Fig. 6.

Repression of Spry2 expression potentiates nickel-induced ERK phosphorylation. (A) ERK activation in the BEAS-2B cells treated with 100 μM NiCl₂ for 8 weeks. Whole cell extracts were prepared from BEAS-2B and nickel-treated cells, and phosphorylation levels of ERK1/2 at Thr202/Tyr204 and total ERK1/2 were assessed by immunoblotting. (B) Knockdown of SPRY2 or JMJD1A potentiates nickel-induced ERK phosphorylation. Control and nickel-treated BEAS-2B cells were transfected with siRNA against SPRY2 or JMJD1A. The phosphorylation forms and total ERK1/2 were then assessed by immunoblotting. (C) Overexpression of HA-SPRY2 or JMJD1A attenuates ERK phosphorylation in the nickel-treated BEAS-2B cells. Control and nickel-treated BEAS-2B cells were transfected with HA-SPRY2 or Flag-JMJD1A expression vector. The phosphorylation forms and total ERK1/2 were then assessed by immunoblotting.

Fig. 7.

Repression of Spry2 expression potentiates nickel-induced anchorage-independent growth in BEAS-2B cells. (A) Nickel-treated BEAS-2B cells (100 μM NiCl₂ for 8 weeks) have increased anchorage-independent growth. 6 × 10³ control and nickel-treated BEAS-2B cells were seeded in soft agar. After 4 weeks incubation, colonies grown in soft agar were stained, photographed and counted. Values are presented as mean ± SD, N = 3. (B) Attenuation of the nickel ion-induced ERK phosphorylation in BEAS-2B cells that stably express exogenous HA-SPRY2. BEAS-2B cells that stably express HA or HA-SPRY2 were treated with 100 μM NiCl₂ for 3 weeks. Immunoblotting was utilized to assess the expression of endogenous SPRY2, exogenous HA-SPRY2, as well as the phosphorylation forms and total ERK1/2. The HA-SPRY2 vector expresses two bands, a phenomenon reported previously by several groups (30). Western blot band intensities were quantified using an ImageJ program, and results were expressed as top band/bottom band. (C) Stable expression of SPRY2 inhibits nickel-induced anchorage-independent growth in BEAS-2B cells. BEAS-2B cells that stably express HA or HA-SPRY2 were treated with 100 μM NiCl₂ for 3 weeks and then seeded in soft agar. After 4 weeks, colonies were stained with 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride solution, photographed and counted. Values are presented as mean ± SD, N = 3.