Use of global gene expression patterns in mechanistic studies of oestrogen action in MCF7 human breast cancer cells

Over the years, the MCF7 human breast cancer cell line has provided a model system for the study of cellular and molecular mechanisms in oestrogen regulation of cell proliferation and in progression to oestrogen and antioestrogen independent growth. Global gene expression profiling has shown that oestrogen action in MCF7 cells involves the coordinated regulation of hundreds of genes across a wide range of functional groupings and that more genes are down regulated than upregulated. Adaptation to long-term oestrogen deprivation, which results in loss of oestrogen-responsive growth, involves alterations to gene patterns not only at early time points (0-4 weeks) but continuing through to later times (20-55 weeks), and even involves alterations to patterns of oestrogen-regulated gene expression. Only 48% of the genes which were regulated >= 2-fold by oestradiol in oestrogen-responsive cells retained this responsiveness after long-term oestrogen deprivation but other genes developed de novo oestrogen regulation. Long-term exposure to fulvestrant, which resulted in loss of growth inhibition by the antioestrogen, resulted in some very large fold changes in gene expression up to 10,000-fold. Comparison of gene profiles produced by environmental chemicals with oestrogenic properties showed that each ligand gave its own unique expression profile which suggests that environmental oestrogens entering the human breast may give rise to a more complex web of interference in cell function than simply mimicking oestrogen action at inappropriate times. (C) 2009 Elsevier Ltd. All rights reserved.

Effect of sulphation on the oestrogen agonist activity of the phytoestrogens genistein and daidzein in MCF-7 human breast cancer cells

The phytoestrogens genistein, daidzein and the daidzein metabolite equol have been shown previously to possess oestrogen agonist activity. However, following consumption of soya diets, they are found in the body not only as aglycones but also as metabolites conjugated at their 4'- and 7-hydroxyl groups with sulphate. This paper describes the effects of monosulphation on the oestrogen agonist properties of these three phytoestrogens in MCF-7 human breast cancer cells in terms of their relative ability to compete with [H-3]oestradiol for binding to oestrogen receptor (ER), to induce a stably transfected oestrogen-responsive reporter gene (ERE-CAT) and to stimulate cell growth. In no case did sulphation abolish activity. The 4'-sulphadon of genistein reduced oestrogen agonist activity to a small extent in whole-cell assays but increased the relative binding affinity to ER. The 7-sulphation of genistein, and also of equol, reduced oestrogen agonist activity substantially in all assays. By contrast, the position of monosulphation of daidzein acted in an opposing manner on oestrogen agonist activity. Sulphation at the 4'-position of daidzein resulted in a modest reduction in oestrogen agonist activity but sulphation of daidzein at the 7-position resulted in an increase in oestrogen agonist activity. Molecular modelling and docking studies suggested that the inverse effects of sulphation could be explained by the binding of daidzein into the ligand-binding domain of the ER in the opposite orientation compared with genistein and equol. This is the first report of sulphation enhancing activity of an isoflavone and inverse effects of sulphation between individual phytoestrogens.

Changes in oestrogen receptor-alpha and -beta during progression to acquired resistance to tamoxifen and fulvestrant (Faslodex, ICI 182,780) in MCF7 human breast cancer cells

Cell culture models of antioestrogen resistance often involve applying selective pressures of oestrogen deprivation simultaneously with addition of tamoxifen or fulvestrant (Faslodex, ICI 182,780) which makes it difficult to distinguish events in development of antioestrogen resistance from those in loss of response to oestrogen or other components. We describe here time courses of loss of antioestrogen response using either oestrogen-maintained or oestrogen-deprived MCF7 cells in which the only alteration to the culture medium was addition of 10(-6) M tamoxifen or 10(-7) M fulvestrant. In both oestrogen-maintained and oestrogen-deprived models, loss of growth response to tamoxifen was not associated with loss of response to fulvestrant. However, loss of growth response to fulvestrant was associated in both models with concomitant loss of growth response to tamoxifen. Measurement of oestrogen receptor alpha (ER alpha) and oestrogen receptor beta (ER beta) mRNA by real-time RT-PCR together with ER alpha and ER beta protein by Western immunoblotting revealed substantial changes to ER alpha levels but very little alteration to ER beta levels following development of antioestrogen resistance. In oestrogen-maintained cells, tamoxifen resistance was associated with raised levels of ERa mRNA/protein. However by contrast, in oestrogen-deprived MCF7 cells, where oestrogen deprivation alone had already resulted in increased levels of ERa mRNA/protein, long-term tamoxifen exposure now reduced ER alpha levels. Whilst long-term exposure to fulvestrant reduced ERa. mRNA/protein levels in the oestrogen-maintained cells to a level barely detectable by Western immunoblotting and non-functional in inducing gene expression (ERE-LUC reporter or pS2), in oestrogen-deprived cells the reduction was much less substantial and these cells retained an oestrogen-induction of both the ERE-LUC reporter gene and the endogenous pS2 gene which could still be inhibited by antioestrogen. This demonstrates that whilst ER alpha can be abrogated by fulvestrant and increased by tamoxifen in some circumstances, this does not always hold true and mechanisms other than alteration to ER must be involved in the development of antioestrogen resistant growth. (c) 2006 Elsevier Ltd. All rights reserved.

Comparative study of oestrogenic properties of eight phytoestrogens in MCF7 human breast cancer cells

Previous studies have compared the oestrogenic properties of phytoestrogens in a wide variety of disparate assays. Since not all phytoestrogens have been tested in each assay, this makes inter-study comparisons and ranking oestrogenic potency difficult. In this report, we have compared the oestrogen agonist and antagonist activity of eight phytoestrogens (genistein, daidzein, equol, miroestrol, deoxymiroestrol, 8-prenylnaringenin, coumestrol and resveratrol) in a range of assays all based within the same receptor and cellular context of the MCF7 human breast cancer cell line. The relative binding of each phytoestrogen to oestrogen receptor (ER) of MCF7 cytosol was calculated from the molar excess needed for 50 % inhibition of [H-3]oestradiol binding (IC50), and was in the order coumestrol (35x)/8-prenylnaringenin (45 x)/deoxymiroestrol (50 x) > miroestrol (260x) > genistein (1000x) > equol (4000x) > daidzein (not achieved: 40 % inhibition at 10(4)-fold molar excess) > resveratrol (not achieved: 10 % inhibition at 10(5)-fold molar excess). For cell-based assays, the rank order of potency (estimated in terms of the concentration needed to achieve a response equivalent to 50 % of that found with 17 beta-oestradiol (IC50)) remained very similar for all the assays whether measuring ligand ability to induce a stably transfected oestrogen-responsive ERE-CAT reporter gene, cell growth in terms of proliferation rate after 7 days or cell growth in terms of saturation density after 14 days. The IC50 values for these three assays in order were for 17 beta-oestradiol (1 x 10-(11) M, 1 x 10-(11) M, 2 x 10(-11) M), and in rank order of potency for the phytoestrogens, deoxymiroestrol (1 x 10(-10) M, 3 x 10(-11) M, 2 x 10(-11) M) > miroestrol (3 x 10(-10) M, 2 x 10(-11) M, 8 x 10(-11) M) > 8-prenylnaringenin (1 x 10(-9) M, 3 x 10(-10) M, 3 x 10(-10) M) > cournestrol (3 x 10(-8) M, 2 x 10(-8) M, 3 x 10(-8) M) > genistein (4 x 10(-8) M, 2 x 10(-8) M, 1 x 10(-8) M)/equol (1 x 10(-7) M, 3 x 10(-8) M, 2 x 10(-8) M) > daidzein (3 x 10(-7) M, 2 x 10(-7) M, 4 x 10(-8) M) > resveratrol (4 x 10(-6) M, not achieved, not achieved). Despite using the same receptor context of the MCF7 cells, this rank order differed from that determined from receptor binding. The most marked difference was for cournestrol and 8-prenylnaringenin which both displayed a relatively potent ability to displace [3H]oestradiol from cytosolic ER compared with their much lower activity in the cell-based assays. Albeit at varying concentrations, seven of the eight phytoestrogens (all except resveratrol) gave similar maximal responses to that given by 17 beta-oestradiol in cell-based assays which makes them full oestrogen agonists. We found no evidence for any oestrogen antagonist action of any of these phytoestrogens at concentrations of up to 10(-6) M on either reporter gene induction or on stimulation of cell growth. (c) 2005 Elsevier Ltd. All rights reserved.

The intracellular genistein metabolite 5,7,3 ',4 '-tetrahydroxyisoflavone mediates G2-M cell cycle arrest in cancer cells via modulation of the p38 signaling pathway

The cellular actions of genistein are believed to mediate the decreased risk of breast cancer associated with high soy consumption. We have investigated the intracellular metabolism of genistein in T47D tumorigenic and MCF-10A nontumorigenic cells and assessed the cellular actions of resultant metabolites. Genistein selectively induced growth arrest and G2-M phase cell cycle block in T47D but not MCF10A breast epithelial cells. These antiproliferative effects were paralleled by significant differences in the association of genistein to cells and in particular its intracellular metabolism. Genistein was selectively taken up into T47D cells and was subject to metabolism by CYP450 enzymes leading to the formation of both 5,7,3',4'-tetrahydroxyisoflavone (THIF) and two glutathionyl conjugates of THIF THIF inhibited cdc2 activation via the phosphorylation of p38 MAP kinase, suggesting that this species may mediate genistein's cellular actions. THIF exposure activated p38 and caused subsequent inhibition of cyclin B1 (Ser 147) and cdc2 (Thr 161) phosphorylation, two events critical for the correct functioning of the cdc2-cyclin B1 complex. We suggest that the formation of THIF may mediate the cellular actions of genistein in tumorigenic breast epithelial cells via the activation of signaling through p38. (c) 2006 Elsevier Inc. All rights reserved.

Investigating the mechanism of enhanced cytotoxicity of HPMA copolymer-Dox-AGM in breast cancer cells

Recently we have described an HPMA copolymer conjugate carrying both the aromatase inhibitor aminoglutethimide (AGM) and doxorubicin (Dox) as combination therapy. This showed markedly enhanced in vitro cytotoxicity compared to the HPMA copolymer-Dox (FCE28068), a conjugate that demonstrated activity in chemotherapy refractory breast cancer patients during early clinical trials. To better understand the superior activity of HPMA copolymer-Dox-AGM, here experiments were undertaken using MCF-7 and MCF-7ca (aromatase-transfected) breast cancer cell lines to: further probe the synergistic cytotoxic effects of AGM and Dox in free and conjugated form; to compare the endocytic properties of HPMA copolymer-Dox-AGM and HPMA copolymer-Dox (binding, rate and mechanism of cellular uptake); the rate of drug liberation by lysosomal thiol-dependant proteases (i.e. conjugate activation), and also, using immunocytochemistry, to compare their molecular mechanism of action. It was clearly shown that attachment of both drugs to the same polymer backbone was a requirement for enhanced cytotoxicity. FACS studies indicated both conjugates have a similar pattern of cell binding and endocytic uptake (at least partially via a cholesterol-dependent pathway), however, the pattern of enzyme-mediated drug liberation was distinctly different. Dox release from PK1 was linear with time, whereas the release of both Dox and AGM from HPMA copolymer-Dox-AGM was not, and the initial rate of AGM release was much faster than that seen for the anthracycline. Immunocytochemistry showed that both conjugates decreased the expression of ki67. However, this effect was more marked for HPMA copolymer-Dox-AGM and, moreover, only this conjugate decreased the expression of the anti-apoptotic protein bcl-2. In conclusion, the superior in vitro activity of HPMA copolymer-Dox-AGM cannot be attributed to differences in endocytic uptake, and it seems likely that the synergistic effect of Dox and AGM is due to the kinetics of intracellular drug liberation which leads to enhanced activity. (c) 2006 Elsevier B.V All rights reserved.

Antitumor activity of Titanocene Y against freshly explanted human breast tumor cells and in xenografted MCF-7 tumors in mice

Bis-[(p-methoxybenzyl)cyclopentadienyl] titanium dichloride, better known as Titanocene Y, is a newly synthesized transition metal-based anticancer drug. We studied the antitumor activity of Titanocene Y with concentrations of 2.1, 21 and 210 mu mol/l against a freshly explanted human breast cancer, using an in-vitro soft agar cloning system. The sensitivity against Titanocene Y was highly remarkable in the breast cancer tumor in the full concentration range. Titanocene Y showed cell death induction at 2.1 mu mol/l, well comparable to cisplatin, given at a concentration of 1.0 mu mol/l. A further preclinical development of Titanocene Y was warranted and therefore an MCF-7 human breast cancer xenograft nonobese diabetic/severe combined immunodeficient mouse model was used. Titanocene Y was given for 21 days at 30 mg/kg/ day (75% of the maximum tolerable dose of Titanocene Y), which resulted in the reduction of the tumor volume to around one-third, whereas no mouse was lost because of the surprisingly low toxicity of Titanocene Y.

In vitro immunomodulatory activity of Lactobacillus fermentum CECT5716 and Lactobacillus salivarius CECT5713: two probiotic strains isolated from human breast milk

Commensal bacteria, including some species of lactobacilli commonly present in human breast milk, appear to colonize the neonatal gut and contribute to protection against infant infections, suggesting that lactobacilli could potentially modulate immunity. In this study, we evaluated the potential of two Lactobacillus strains isolated from human milk to modulate the activation and cytokine profile of peripheral blood mononuclear cell (PBMC) subsets in vitro. Moreover, these effects were compared to the same probiotic species of non-milk origin. Lactobacillus salivarius CECT5713 and Lactobacillus fermentum CECT5716 at 105, 106 and 107 bacteria/mL were co-cultured with PBMC (106/mL) from 8 healthy donors for 24 h. Activation status (CD69 and CD25 expressions) of natural killer (NK) cells (CD56+), total T cells (CD3+), cytotoxic T cells (CD8+) and CD4+ T cells was determined by flow cytometry. Regulatory T cells (Treg) were also quantified by intracellular Foxp3 evaluation. Regarding innate immunity, NK cells were activated by addition of both Lactobacillus strains, and in particular, the CD8+ NK subset was preferentially induced to highly express CD69 (90%, p<0.05). With respect to acquired immunity, approximately 9% of CD8+ T cells became activated after co-cultivation with L. fermentum or L salivarius. Although CD4+ T cells demonstrated a weaker response, there was a preferential activation of Treg cells (CD4+CD25+Foxp3+) after exposure to both milk probiotic bacteria (p<0.05). Both strains significantly induced the production of a number of cytokines and chemokines, including TNFα, IL-1β, IL-8, MIP-1α, MIP-1β, and GM-CSF, but some strain-specific effects were apparent. This work demonstrates that L salivarius CECT5713 and L. fermentum CECT5716 enhanced both natural and acquired immune responses, as evidenced by the activation of NK and T cell subsets and the expansion of Treg cells, as well as the induction of a broad array of cytokines.

Differential effects of overexpression of ERalpha and ERbeta in MCF10A immortalised, non-transformed human breast epithelail cells

Background: Cellular effects of oestrogen are mediated by two intracellular receptors ERα and ERβ. However, to compare responses mediated through these two receptors, experimental models are needed where ERα and ERβ are individually stably overexpressed in the same cell type. Methods: We compared the effects of stable overexpression of ERα and ERβ in the MCF10A cell line, which is an immortalised but non-transformed breast epithelial cell line without high endogenous ER expression. Results: Clones of MCF10A cells were characterised which stably overexpressed ERα (10A-ERα2, 10A-ERα13) or which stably overexpressed ERβ (10A-ERβ12, 10A-ERβ15). Overexpression of either ERα or ERβ allowed induction of an oestrogen-regulated ERE-LUC reporter gene by oestradiol which was not found in the untransfected cells. Oestradiol also increased proliferation of 10A-ERα13 and 10A-ERβ12 cells, but not untransfected cells, by 1.3-fold over 7 days. The phytoestrogen, genistein, which is reported to bind more strongly to ERβ than to ERα, could induce luciferase gene expression from an ERE-LUC reporter gene at concentrations of 10−6 M and 10−5 M but only in the clones overexpressing ERβ and not in those overexpressing ERα. Clone 10A-ERβ12 also yielded growth stimulation with 10-6 M genistein. Finally, the overexpression of ERα, but not ERβ, gave rise to increased growth in semi-solid methocel suspension culture in the presence of 70 nM oestradiol, suggesting that overexpression of ERα, but not ERβ, produces characteristics of a transformed phenotype. Conclusions: This provides a model system to compare effects of oestradiol with other oestrogenic ligands in cells stably overexpressing individually ERα or ERβ.

Enhanced sensitivity to rapamycin following long-term oestrogen deprivation in MCF-7, T-47-D and ZR-75-1 human breast cancer cells

Human breast cancer cells (MCF-7, T-47-D and ZR-75-1) can adapt to circumvent any reduced growth rate during long-term oestrogen deprivation, and this provides three model systems to investigate mechanisms of endocrine resistance in breast cancer. In this paper we report consistent differences in the effects of three growth inhibitors following long-term oestrogen deprivation in all three cell models. Long-term oestrogen deprivation of MCF-7, T-47-D and ZR-75-1 cells resulted in reduced growth inhibition by PD98059 (2–10 µg/ml), implying a loss of dependence on mitogen-activated protein kinase pathways for growth. The growth inhibitor LY294002 (2–10 µM) inhibited growth of both oestrogen-maintained and oestrogen-deprived cells with similar dose–responses, implying continued similar dependence on phosphoinositide 3-kinase (PI3K) pathways with no alteration after adaptation to oestrogen independent growth. However, by contrast, long-term oestrogen deprivation resulted in an increased sensitivity to growth inhibition by rapamycin, which was not reduced by readdition of oestradiol. The enhanced inhibition of long-term oestrogen-deprived MCF-7-ED, T-47-D-ED and ZR-75-1-ED cell growth by combining rapamycin with LY294002 at concentrations where each alone had little effect, offers preclinical support to the development of therapeutic combinations of rapamycin analogues with other PI3K inhibitors in endocrine-resistant breast cancer.

Equol: a comparison of the effects of the racemic compound with that of the purified S-enantiomer on the growth, invasion, and DNA integrity of breast and prostate cells in vitro

It has been postulated that the R- and S-equol enantiomers have different biological properties given their different binding affinities for the estrogen receptor. S-(-)equol is produced via the bacterial conversion of the soy isoflavone daidzein in the gut. We have compared the biological effects of purified S-equol to that of racemic (R and S) equol on breast and prostate cancer cells of varying receptor status in vitro. Both racemic and S-equol inhibited the growth of the breast cancer cell line MDA-MB-231 (> or = 10 microM) and the prostate cancer cell lines LNCaP (> or = 5 microM) and LAPC-4 (> or = 2.5 microM). The compounds also showed equipotent effects in inhibiting the invasion of MDA-MB-231 and PC-3 cancer cells through matrigel. S-equol (1, 10, 30 microM) was unable to prevent DNA damage in MCF-7 or MCF-10A breast cells following exposure to 2-hydroxy-4-nonenal, menadione, or benzo(a)pyrene-7,8-dihydrodiol-9,10-epoxide. In contrast, racemic equol (10, 30 microM) prevented DNA damage in MCF-10A cells following exposure to 2-hydroxy-4-nonenal or menadione. These findings suggest that racemic equol has strong antigenotoxic activity in contrast to the purified S-equol enantiomer implicating the R-, rather than the S-enantiomer as being responsible for the antioxidant effects of equol, a finding that may have implications for the in vivo chemoprotective properties of equol.

Molecular mechanisms of oestrogen action on growth of human breast cancer cells in culture

Growth responses to oestrogen can be reproducibly obtained using a selection of oestrogen-receptor-containing human breast cancer cell lines, and molecular mechanisms have been shown to include modulation to growth factor/receptor/signalling pathways, cell-cycle proteins, apoptosis, differentiation, adhesion, motility and migration. Considerable progress has been made in understanding the molecular basis of oestrogen action on gene expression through the ligand-activated transcription factors human oestrogen receptor α (ERα) and ERβ and the resulting effects on global gene expression patterns, but the full profile of coordination of the alterations, which brings about changes in cell growth through genomic and non-genomic mechanisms remain to be fully elucidated. Oestrogen regulation of cell growth involves a complex cross-talk between oestrogen receptor and growth factor signalling pathways such that inhibition of one pathway may lead to stimulation of another, which may explain the remarkable ability of human breast cancer cells to escape from any mode of imposed growth inhibition be it oestrogen deprivation or administration of antioestrogen. Although studies on cell growth have focused to date on the effects of physiological oestrogens, many hundreds of environmental chemicals with oestrogenic properties have now been measured in the human breast. Whether or not the weight of evidence eventually establishes any causal link of complex mixtures of environmental oestrogenic chemicals with breast cancer, the presence of so many oestrogenic chemicals in the breast must influence resulting oestrogenic responses, and the impact of this additional oestrogenic burden needs to be taken into account in future studies on growth regulation of human breast cancer cells.

HPMA copolymer-aminoglutethimide conjugates inhibit aromatase in MCF-7 cell lines

N-(2-Hydroxypropyl)methacrylamide (HPMA) copolymer–doxorubicin (Dox) has already shown clinical activity in breast cancer patients. Moreover, we have recently found that an HPMA conjugate containing a combination of both Dox and the aromatase inhibitor aminoglutethimide (AGM) shows significantly increased anti-tumour activity in vitro. To better understand the mechanism of action of HPMA copolymer–AGM conjugates several models were used here to investigate their effect on cell growth and aromatase inhibition. Cytotoxicity of HPMA copolymer conjugates containing AGM, Dox and also the combination AGM–Dox was determined by MTT assay in MCF-7 and MCF-7ca cells. Androstenedione (5 × 10− 8 M) stimulates the growth of MCF-7ca cells. Both free AGM and polymer-bound AGM (0.2–0.4 mg/ml) were shown to block this mitogenic activity. When MCF-7ca cells were incubated [3H]androstenedione both AGM and HPMA copolymer–GFLG–AGM (0.2 mg/ml AGM-equiv.) showed the ability to inhibit aromatase. Although, free AGM was able to inhibit isolated human placental microsomal aromatase in a concentration dependent manner, polymer-bound AGM was not, suggesting that drug release is essential for activity of the conjugate. HPMA copolymer conjugates containing aromatase inhibitors have potential for the treatment of hormone-dependant cancers, and it would be particularly interesting to explore further as potential therapies in post-menopausal women as components of combination therapy.

Daidzein, R-(+)equol and S-(-)equol inhibit the invasion of MDA-MB-231 breast cancer cells potentially via the down-regulation of matrix metalloproteinase-2

PURPOSE: Soy isoflavones may inhibit tumor cell invasion and metastasis via their effects on matrix metalloproteinases (MMPs) and their tissue inhibitors (TIMPs). The current study investigates the effects of daidzein, R- and S-equol on the invasion of MDA-MB-231 human breast cancer cells and the effects of these compounds on MMP/TIMP expression at the mRNA level. METHODS: The anti-invasive effects of daidzein, R- and S-equol (0, 2.5, 10, 50 μM) on MDA-MB-231 cells were determined using the Matrigel invasion assay following 48-h exposure. Effects on MMP-2, MMP-9, TIMP-1 and TIMP-2 expression were assessed using real-time PCR. Chiral HPLC analysis was used to determine intracellular concentrations of R- and S-equol. RESULTS: The invasive capacity of MDA-MB-231 cells was significantly reduced (by approximately 50-60 %) following treatment with 50 μM daidzein, R- or S-equol. Anti-invasive effects were also observed with R-equol at 2.5 and 10 μM though overall equipotent effects were induced by all compounds. Inhibition of invasion induced by all three compounds at 50 μM was associated with the down-regulation of MMP-2, while none of the compounds tested significantly affected the expression levels of MMP-9, TIMP-1 or TIMP-2 at this concentration. Following exposure to media containing 50 μM R- or S-equol for 48-h intracellular concentrations of R- and S-equol were 4.38 ± 1.17 and 3.22 ± 0.47 nM, respectively. CONCLUSION: Daidzein, R- and S-equol inhibit the invasion of MDA-MB-231 human breast cancer cells in part via the down-regulation of MMP-2 expression, with equipotent effects observed for the parent isoflavone daidzein and the equol enantiomers.

Increasing doxorubicin activity against breast cancer cells using PPARγ-ligands and by exploiting circadian rhythms

Doxorubicin is effective against breast cancer, but its major side effect is cardiotoxicity. The aim of this study was to determine whether the efficacy of doxorubicin on cancer cells could be increased in combination with PPARγ agonists or chrono-optimization by exploiting the diurnal cycle. We determined cell toxicity using MCF-7 cancer cells, neonatal rat cardiac myocytes and fibroblasts in this study. Doxorubicin damages the contractile filaments of cardiac myocytes and affects cardiac fibroblasts by significantly inhibiting collagen production and proliferation at the level of the cell cycle. Cyclin D1 protein levels decreased significantly following doxorubicin treatment indicative of a G1 /S arrest. PPARγ agonists with doxorubicin increased the toxicity to MCF-7 cancer cells without affecting cardiac cells. Rosiglitazone and ciglitazone both enhanced anti-cancer activity when combined with doxorubicin (e.g. 50% cell death for doxorubicin at 0.1 μM compared to 80% cell death when combined with rosiglitazone). Thus, the therapeutic dose of doxorubicin could be reduced by 20-fold through combination with the PPARγ agonists, thereby reducing adverse effects on the heart. The presence of melatonin also significantly increased doxorubicin toxicity, in cardiac fibroblasts (1 μM melatonin) but not in MCF-7 cells. Our data show, for the first time, that circadian rhythms play an important role in doxorubicin toxicity in the myocardium; doxorubicin should be administered mid-morning, when circulating levels of melatonin are low, and in combination with rosiglitazone to increase therapeutic efficacy in cancer cells while reducing the toxic effects on the heart.