Endocrine resistance is a major problem with anti-estrogen treatments and how to overcome resistance is a major concern in the clinic. Reliable measurement of cell viability, proliferation, growth inhibition and death is important in screening for drug treatment efficacy in vitro. This report describes and compares commonly used proliferation assays for induced estrogen-responsive MCF-7 breast cancer cell cycle arrest including: determination of cell number by direct counting of viable cells; or fluorescence SYBR®Green (SYBR) DNA labeling; determination of mitochondrial metabolic activity by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay; assessment of newly synthesized DNA using 5-ethynyl-2′-deoxyuridine (EdU) nucleoside analog binding and Alexa Fluor® azide visualization by fluorescence microscopy; cell-cycle phase measurement by flow cytometry. Treatment of MCF-7 cells with ICI 182780 (Faslodex), FTY720, serum deprivation or induction of the tumor suppressor p14ARF showed inhibition of cell proliferation determined by the Trypan Blue exclusion assay and SYBR DNA labeling assay. In contrast, the effects of treatment with ICI 182780 or p14ARF-induction were not confirmed using the MTS assay. Cell cycle inhibition by ICI 182780 and p14ARF-induction was further confirmed by flow cytometric analysis and EdU-DNA incorporation. To explore this discrepancy further, we showed that ICI 182780 and p14ARF-induction increased MCF-7 cell mitochondrial activity by MTS assay in individual cells compared to control cells thereby providing a misleading proliferation readout. Interrogation of p14ARF-induction on MCF-7 metabolic activity using TMRE assays and high content image analysis showed that increased mitochondrial activity was concomitant with increased mitochondrial biomass with no loss of mitochondrial membrane potential, or cell death. We conclude that, whilst p14ARF and ICI 182780 stop cell cycle progression, the cells are still viable and potential treatments utilizing these pathways may contribute to drug resistant cells. These experiments demonstrate how the combined measurement of metabolic activity and DNA labeling provides a more reliable interpretation of cancer cell response to treatment regimens.

As part of a cell’s inherent protection against carcinogenesis, p14ARF is upregulated in response to hyperproliferative signalling to induce cell cycle arrest. This property makes p14ARF a leading candidate for cancer therapy. This study explores the consequences of reactivating p14ARF in breast cancer and the potential of targeting p14ARF in breast cancer treatment. Our results show that activation of the p14ARF-p53-p21-Rb pathway in the estrogen sensitive MCF-7 breast cancer cells induces many hallmarks of senescence including a large flat cell morphology, multinucleation, senescence-associated-β-gal staining, and rapid G1 and G2/M phase cell cycle arrest. P14ARF also induces the expression of the proto-oncogene cyclin D1, which is most often associated with a transition from G1-S phase and is highly expressed in breast cancers with poor clinical prognosis. In this study, siRNA knockdown of cyclin D1, p21 and p53 show p21 plays a pivotal role in the maintenance of high cyclin D1 expression, cell cycle and growth arrest post-p14ARF induction. High p53 and p14ARF expression and low p21/cyclin D1 did not cause cell-cycle arrest. Knockdown of cyclin D1 stops proliferation but does not reverse senescence-associated cell growth. Furthermore, cyclin D1 accumulation in the nucleus post-p14ARF activation correlated with a rapid loss of nucleolar Ki-67 protein and inhibition of DNA synthesis. Latent effects of the p14ARF-induced cellular processes resulting from high nuclear cyclin D1 accumulation included a redistribution of Ki-67 into the nucleoli, aberrant nuclear growth (multinucleation), and cell proliferation. Lastly, downregulation of cyclin D1 through inhibition of ER abrogated latent recurrence. The mediation of these latent effects by continuous expression of p14ARF further suggests a novel mechanism whereby dysregulation of cyclin D1 could have a double-edged effect. Our results suggest that p14ARF induced-senescence is related to late-onset breast cancer in estrogen responsive breast cancers and/or the recurrence of more aggressive breast cancer post-therapy.

No abstract

Many years of oncogenic insult and hormonal flux have important long-term implications for breast cancer emergence and progression. Premature-induced senescence is proposed to be a protective response against cancer development. We hypothesize that the senescence response can be subverted in some breast carcinomas, which exhibit latent hormone-dependent proliferative capacity. The ARF-p53 pathway has long been known to play an important role in tumour surveillance halting cell cycle progression to transiently or permanently stop aberrant cells from proliferating. It is well documented that disruption of this pathway is involved in tumourigenesis, in part, through inactivation of cyclin D1. Cyclin D1 plays an important role in breast cancer progression as a regulator of CDK4/6. In addition cyclin D1 has been reported to stimulate estrogen receptor (ER) activity independent of CDK4/6. Our initial observations showed that reactivation of ARF-p53-p21 is indeed a potent rapid inhibitor of cell proliferation in MCF-7 breast cancer cells with a senescent phenotype, however controversially cyclin D1 expression was increased and predominantly located in the nucleus. High expression of nuclear cyclin D1 and p14ARF inversely correlated with cell proliferation. Our studies showed that p14ARF regulated cyclin D1 at the post-transcriptional level. Furthermore, we have evidence from microRNA array analysis that microRNAs known to bind to the cyclin D1 3'UTR sequence were down regulated upon induction of p14ARF, contributing to cyclin D1 stability. We found that long-term exposure to p14ARF was not a failsafe barrier to tumour progression, senescent-like cells re-entering the cell cycle correlated with reduction of nuclear cyclin D1. Conversely, continuous p14ARF expression resulted in chaotic multinucleation and preceded neoplasia. These aberrant nuclei co-stained with cyclin D1 and Ki-67, a well-known marker of cell proliferation. These results suggest that induction of the p14ARF-p53 pathway does not entirely remove the threat of cancer resulting from hyperproliferative oncogenic signals and cyclin D1 plays a promiscuous role in prevention and progression. Arguably, in some breast cancers premature senescent cells may be precursor, or latent cancer cells. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P1-04-03.