Influence of cell cycle on responses of MCF-7 cells to benzo[a]pyrene

Abstract

Background: Benzo[a]pyrene (BaP) is a widespread environmental genotoxic carcinogen that damages DNA by forming adducts. This damage along with activation of the aryl hydrocarbon receptor (AHR) induces complex transcriptional responses in cells. To investigate whether human cells are more susceptible to BaP in a particular phase of the cell cycle, synchronised breast carcinoma MCF-7 cells were exposed to BaP. Cell cycle progression was analysed by flow cytometry, DNA adduct formation was assessed by 32P-postlabeling analysis, microarrays of 44K human genome-wide oligos and RT-PCR were used to detect gene expression (mRNA) changes and Western blotting was performed to determine the expression of some proteins, including cytochrome P450 (CYP) 1A1 and CYP1B1, which are involved in BaP metabolism.

Results: Following BaP exposure, cells evaded G1 arrest and accumulated in S-phase. Higher levels of DNA damage occurred in S- and G2/M- compared with G0/G1-enriched cultures. Genes that were found to have altered expression included those involved in xenobiotic metabolism, apoptosis, cell cycle regulation and DNA repair. Gene ontology and pathway analysis showed the involvement of various signalling pathways in response to BaP exposure, such as the Catenin/Wnt pathway in G1, the ERK pathway in G1 and S, the Nrf2 pathway in S and G2/M and the Akt pathway in G2/M. An important finding was that higher levels of DNA damage in S- and G2/M-enriched cultures correlated with higher levels of CYP1A1 and CYP1B1 mRNA and proteins. Moreover, exposure of synchronised MCF-7 cells to BaP-7,8-diol-9,10-epoxide (BPDE), the ultimate carcinogenic metabolite of BaP, did not result in significant changes in DNA adduct levels at different phases of the cell cycle.

Conclusions: This study characterised the complex gene response to BaP in MCF-7 cells and revealed a strong correlation between the varying efficiency of BaP metabolism and DNA damage in different phases of the cell cycle. Our results suggest that growth kinetics within a target-cell population may be important determinants of susceptibility and response to a genotoxic agent.

Figure 1

BaP delays escape from S-phase. A- MCF-7 cells were synchronised in G0/G1-phase by serum-deprivation for 48 hours, after which cells were exposed for 12 h to BaP (2.5 μM) or DMSO. B- MCF-7 cells were enriched in S-phase by serum-deprivation for 48 h, then left to grow for 18 h, after which they were treated by either BaP (2.5 μM) or DMSO for 12 h. C- MCF-7 cells were synchronised in G2/M-phase by exposing them to aphidicolin (1 μg/mL) for 24 h followed by colchicine (0.25 μM) for 12 h. Subsequently, they were released into media containing either BaP (2.5 μM) or DMSO for 12 h. Cell cycle distribution was examined by flow cytometry. The profiles are representative of three independent experiments.

Figure 2

DNA adduct levels in synchronised MCF-7 cells. Cell were synchronised in G0/G1, S and G2/M phases with different methods, after which they were exposed to 2.5 μM BaP or 0.5 μM BPDE for 12 h. DNA was isolated with a standard phenol/chloroform method and DNA adducts were assessed by the ³²P-postlabelling method. Results were expressed as DNA adducts/10⁸ nucleotides.

Figure 3

Cell cycle effects on microarray results. A- Hierarchical clustering of genes in different conditions; B- Principal component analysis (PCA). Both methods were performed in GeneSpring on a list of genes that had good confidence measurement and revealed a cell cycle response in gene expression profiles. Before BaP (2.5 μM) treatment for 12 h, MCF-7 cells were synchronised in different phases of the cell cycle. In hierarchical clustering (A), red colour denotes up-regulation and green denotes down-regulation. In PCA analysis (B), squares with black dots denote BaP-treated samples.

Figure 4

Venn diagram of the gene lists in different phases. Only genes that exhibited 1.5-fold or greater change after BaP treatment are shown. The gene lists represent expression profiles of MCF-7 cells synchronised in different phases of the cell cycle. Although there was overlap of the expression profiles between the phases, the majority of the changes were cell-cycle dependent.

Figure 5

Ingenuity Pathway Analysis (IPA) on genes modulated by BaP in G1-enriched MCF7 cell cultures. Gene lists of 1.5-fold differentially expressed genes in different phases were imported to IPA software, which revealed the involvement of several pathways and genes in the response to BaP. Two networks are shown here. Red colour denotes up-regulation and green colour denotes down-regulation. The IPA legend is shown in Additional file 6.

Figure 6

Ingenuity Pathway Analysis (IPA) on genes modulated by BaP in S-enriched MCF-7 cell cultures. Gene lists of 1.5-fold differentially expressed genes in different phases were imported to IPA software, which revealed the involvement of several pathways and genes in the response to BaP. Two networks are shown here. Red colour denotes up-regulation and green colour denotes down-regulation. The IPA legend is shown in Additional file 6.

Figure 7

Ingenuity Pathway Analysis (IPA) on genes modulated by BaP in G2/M-enriched MCF-7 cell cultures. Gene lists of 1.5-fold differentially expressed genes in different phases were imported to IPA software, which revealed the involvement of several pathways and genes in the response to BaP. Three networks are shown here. Red colour denotes up-regulation and green colour denotes down-regulation. The IPA legend is shown in Additional file 6.

Figure 8

Relative CYP1A1 mRNA expression levels in synchronised MCF7 cells with and without BaP treatment. mRNA quantification was carried out by RT-PCR. The highest expressed sample in the RT-PCR was set to 100% and other samples' expressions are shown relative to that. Values represent mean ± SD from 3 determinations.

Figure 9

Expression of CYP1A1, CYP1B1, AHR, p53, and p21 protein levels in MCF7 cells in different phases of the cell cycle in response to BaP detected by Western blots. Cells were treated with DMSO (0.3%) or BaP (2.5 μM) for 12 h. 15 μg protein was loaded in each lane. β-actin was used as a loading control.