Dioxin Suppresses the Checkpoint Protein, MAD2, by an Aryl Hydrocarbon Receptor-independent Pathway¹

The cell cycle checkpoint governing mitotic spindle assembly functions through a highly conserved signal transduction pathway, linking the initiation of anaphase to spindle assembly and the completion of the chromosome-microtubule attachment (1, 2, 3, 4). The molecules implicated in the spindle checkpoint were first identified in yeast (1). Vertebrate homologues of mitotic checkpoint genes (MAD1, MAD2, MAD2B, and BUB 1) were also shown to be required for the execution of the mitotic checkpoint (3, 4, 5). Loss of function of these mitotic checkpoint genes is associated with chromosomal instability in cancer cells (6). In addition, mutations in these genes and/or reduced levels of the resulting proteins have been identified in several human cancers (5, 7, 8, 9). These data strongly suggest that defects in mitotic checkpoint regulation contribute to carcinogenesis through chromosomal instability. The compound TCDD³ is ubiquitously present in the environment. Exposure to TCDD results in a variety of toxic effects, including immunotoxicity, hepatotoxity, teratogenicity, and tumor promotion (10, 11). TCDD toxicity is mediated by the AhR, bHLH/PAS transcription factor (12). However, the mechanism of TCDD toxicity and its contribution to tumor development are still poorly understood. In this study, we analyzed the effects of TCDD on Mad2 expression in MEFs and HeLa cells. In addition, we analyzed the phenotype of HeLa cells after the reduction of Mad2 resulting from TCDD treatment.

TCDD (Cambridge Isotope Laboratories, Inc., Andover, MA), nocodazole (Sigma Chemical Co., St. Louis, MO), and aphidicolin (Wako Pure Chemical Industries, Ltd., Osaka, Japan) were prepared in DMSO.

MEFs were produced as described previously (13) from C57BL/6J female mice. AhR^−/− MEFs were generated from the intercrossed AhR^+/− mice (14). HeLa cells were obtained from American Type Culture Collection. MEFs and HeLa cells were grown in DMEM (Life Technologies, Inc., Rockville, MD), supplemented with 2 mml-glutamine and 10% fetal bovine serum at 37°C in a 5% CO₂ environment. To measure the mitotic index, we calculated the percentage of round-shaped cells, assumed to be mitotic. Data were obtained from >250 cells in four different fields. For cell counting analysis, HeLa cells were treated with nocodazole (50 ng/ml) with or without 1 μm TCDD for 0, 6, 12, or 24 h. At each time point, we harvested the cells from 60-mm dishes and suspended them into 500 μl of PBS. Then we counted the number of cells using an erythrocytometer. We performed independent experiments three times for cell count analysis.

Equal amounts of protein (10 μg/lane) were subjected to 12% SDS-PAGE and transferred to a Hybond-ECL membrane (Amersham Pharmacia Biotech UK Ltd., Buckinghamshire, United Kingdom). The protein samples were confirmed by an equal amount of loading by Ponceau staining of transferred membrane (Sigma). Membranes were probed with an antihuman Mad2 polyclonal antibody (C-19; Santa Cruz Biotechnology, Santa Cruz, CA). The signals were detected with the ECL plus Western blotting detection system (Amersham Pharmacia Biotech UK, Ltd.).

We extracted cellular RNA from MEFs and HeLa cells using ISOGEN (Nippon Gene, Tokyo, Japan). Total RNA (10 μg) was fractionated by electrophoresis on a 1.2% agarose formaldehyde denaturing gel and then transferred to a Hybond-N membrane (Amersham Pharmacia Biotech UK, Ltd.) in 10× SSC. The mouse MAD2 cDNA probe was obtained by reverse transcription-PCR using the primer pair, 5′-ATGGCACAGCAGCTCGCCCGA-3′ and 5′-TCAGTCATTGACAGGGGTTTT-3′, and labeled with ³²P using the Rediprime II random prime labeling system (Amersham Pharmacia Biotech UK, Ltd.).

To investigate the effect of TCDD on the expression of the checkpoint protein, Mad2, we cultured MEFs for 6 h with or without 1 μm TCDD and then examined the protein samples of the cells by immunoblotting with an anti-Mad2 antibody (Fig. 1 A). The results demonstrated that the levels of Mad2 protein were reduced by TCDD treatment.

HeLa cells, a stable, well-characterized cell line, are more suitable than MEFs for cell cycle studies. Mad2 expression in HeLa cells treated with 1 μm TCDD decreased with increasing exposure to 6 h (Fig. 1 A).

To investigate whether the reduced expression of Mad2 is regulated by a transcriptional or posttranscriptional mechanism, we performed Northern blot analysis of total RNA extracted from the MEFs treated for 2 h with several concentrations of TCDD. Using a cDNA fragment corresponding to the full Mad2 open reading frame as a probe, we discovered that Mad2 mRNA levels did not change with TCDD treatment at levels <10 nm (Fig. 1,B). mRNA levels decreased in a dose-dependent manner, however, after treatment with >100 nm TCDD, indicating that the suppression of Mad2 expression by TCDD occurs at the mRNA level. Moreover, we confirmed a decrease of Mad2 mRNA in HeLa cells treated with 1 μm TCDD for 6 h (Fig. 1 B).

We next examined whether TCDD reduction of Mad2 mRNA was dependent on gene transcription and/or mRNA stability. To assess this point, we performed transcription stability analysis in HeLa cells using an RNA synthesis inhibitor, actinomycin D, with or without 1 μm TCDD. Decay curves for Mad2 mRNA in the cells were not significantly different during the 6-h time period (data not shown). From these results, we concluded that TCDD affected the Mad2 transcription.

The effects of TCDD are believed to be mediated by the AhR (15). Studies in AhR null mice, however, suggest that AhR-independent mechanisms may contribute to TCDD-induced developmental toxicity (16). To determine the involvement of the AhR-mediated pathway in the TCDD-induced reduction of Mad2, we examined Mad2 expression in the AhR-deficient (AhR^−/−) MEFs treated with 1 μm TCDD for either 2 h for Northern blot analysis or 6 h for immunoblot analysis (Fig. 1, C and D). These results suggested that reductions in Mad2 after TCDD treatment is mediated by a novel pathway operating independently of the typical AhR-mediated pathway.

Reductions in Mad2 may affect the cell cycle behavior of affected cells. To determine the effect of alterations in cell cycle speed by TCDD treatment, HeLa cells were arrested by aphidicolin (1 μg/ml) for 12 h and then washed twice and released into fresh medium with or without 1 μm TCDD. Cells were harvested every 3 h over 12 h and then analyzed by flow cytometry. The results demonstrate that TCDD does not affect cell cycle speed in HeLa cells within 12 h of treatment (data not shown). Conversely, these data also showed that reductions in Mad2 levels are not a simple response to changes in cell cycle progression.

Mad2 is required for mitotic arrest in response to spindle disruption (3, 17). In this study, TCDD treatment reduced Mad2 protein levels in MEFs and HeLa cells. To confirm the effect of this reduction on the mitotic stress response, we treated HeLa cells with nocodazole (50 ng/ml) with or without 1 μm TCDD for 12 h. Many of the control cells became rounded after nocodazole treatment, indicative of cell cycle arrest in mitosis (Fig. 2,A). In contrast, fewer of the cells treated with TCDD became rounded after nocodazole treatment. The average MI after 12 h of nocodazole treatment was ∼64%; the MI of TCDD treated cells was reduced to 41% (Fig. 2,B). MI was not as large as values reported previously following the inhibition by the anti-Mad2 antibody or the complete disruption of Mad2 function (3, 17). It is possible, however, that TCDD could not penetrate all of the cells because of its extreme hydrophobicity. It is also conceivable that the small reduction in Mad2 levels was not sufficient to escape the arrest in mitosis after nocodazole treatment. Furthermore, to confirm that TCDD treated cells were cycling after nocodazole treatment, we performed cell count analysis. We treated HeLa cells with nocodazole (50 ng/ml) with or without 1 μm TCDD for 0, 6, 12, or 24 h, and then we harvested the cells and counted cell numbers. Fig. 2 C showed that only TCDD-treated cells were increased. From these results, we confirmed that TCDD-treated cells failed to arrest in mitosis after nocodazole treatment. Thus, we conclude that reductions of Mad2 by TCDD treatment in HeLa cells inactivate the cell cycle checkpoint monitoring mitotic spindle assembly.

In the present study, we demonstrated that TCDD treatment reduces the expression of the mitotic checkpoint protein, Mad2, in a dose-dependent manner through an AhR-independent pathway. We also showed that reductions of Mad2 in HeLa cells after TCDD treatment inactivate the mitotic spindle assembly cell cycle checkpoint.

Recent molecular studies of mitotic checkpoint genes have demonstrated close ties between cell cycle checkpoints and chromosome instability (6, 18). Today, chromosomal instability is thought to be the one of the main mechanisms governing human carcinogenesis. Mad2 is a key molecule monitoring chromosomal segregation in the spindle assembly checkpoint. The loss of Mad2 causes both chromosome missegregation and failure to properly arrest at this checkpoint (3, 17). This cell cycle disregulation caused by reductions in Mad2 may contribute to carcinogenesis; Percy et al. (8) discovered a decreased expression of Mad2 in human breast cancer. Although the detailed molecular mechanism of tumor formation by TCDD is still unclear, our data connect carcinogenesis by checkpoint dysfunction to TCDD exposure.

The AhR mediates many of the biological effects of TCDD and controls the transcriptional regulation of genes encoding a number of xenobiotic metabolizing enzymes (12). We confirmed the suppression of Mad2 after TCDD treatment, even in AhR^−/− MEFs. One nm TCDD was sufficient to activate drug-metabolizing genes by AhR. Low concentrations of TCDD (<10 nm), however, did not suppress Mad2 expression (Fig. 1 B). Although we cannot exclude the involvement of the AhR signaling pathway in Mad2 suppression after TCDD treatment, these data suggest that an AhR-independent pathway is the main contributor to TCDD-induced MAD2 suppression.

In this study, TCDD had a minimal effect on the cell cycle profile of HeLa cells. AhR, however, induces the p27^kip1 cyclin/cdk inhibitor (19) and interacts with retinoblastoma protein (20), suggesting that TCDD has the potential to induce cell cycle delay using AhR-mediated signaling. TCDD treatment for 48 h induced cell cycle delay using fetal thymocytes (19). The discrepancy between this study and our data may have resulted from differing cell types or a shorter treatment period.

In summary, our data demonstrated that TCDD has the potential to induce chromosomal instability in human cells through a novel TCDD signaling pathway. Further study of the upstream mechanism of Mad2 suppression by TCDD will hopefully define this AhR-independent TCDD toxicity.

We thank Naomi Fukushima for assistance in preparing the manuscript.