Abstract
Background Cholesterol biosynthesis is regulated by the activity of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase. Cholesterol and its derivatives are required in high concentrations by neoplastic proliferating cells for both DNA synthesis and cell growth. Thus, inhibition of HMG-CoA reductase could effect cell cycle progression and proliferation. Therefore, we examined the effect of an HMG-CoA reductase inhibitor (simvastatin) alone and in combination with cytosine arabinoside (ARA-C) on the proliferation of two AML cell lines.
Methods AML blasts derived from two cell lines (HL-60 and AML-2) were incubated with increasing concentrations of either simvastatin alone or simvastatin alone for 24 hours with ARA-C added thereafter. The effect of the drugs on cell proliferation in liquid culture (3 H thymidine uptake) and on clonogenic assay was analyzed.
Results We found that the number of proliferating AML blasts (suspension cultures) and colony formations (agar cultures) of both cell lines declined significantly after incubation with simvastatin. Preincubation of both cell lines with simvastatin by the addition of increasing concentrations of ARA-C produced a degree of growth inhibition that was significantly greater than that of the individual compounds. This antigrowth interaction was additive rather than synergistic.
Conclusions We conclude that simvastatin has a major antiproliferative effect on AML blasts in vitro. Also, the combination of simvastatin and ARA-C significantly enhanced the antiproliferative effect of each drug. These findings may open new avenues in both the laboratory and clinical research of the treatment of leukemia.
INTRODUCTION
Cholesterol is either synthesized intracellularly or obtained from the plasma by receptor-mediated endocytosis of low density lipoprotein.1Cholesterol biosynthesis is regulated by the activity of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase.1The product of this enzyme, mevalonic acid, is an essential precursor of several important cellular constituents. These include, among other, sterols (especially cholesterol) that are a major structural component of cell membranes, and the side chains of isopentenyladenosine and ubiquinone that are involved in the progression of cell cycle and cell survival.2,3Furthermore, it has been shown recently that membrane association is an important function for mevalonate-derived modifications of several important proteins such as cellular membrane G proteins, proteins encoded by oncogenes (Ras proteins), and nuclear proteins (lamins).4,5
Cancer cells seem to require an increased concentration of cholesterol and its precursors. Indeed, proliferating cells posses a high rate of cholesterol production and a high rate of HMG-CoA reductase activity.6Cholesterol is required in these neoplastic cells for cell proliferation and for the progression of cell cycle.7Also, some proteins (such as p21 ras) that are involved in signal transduction undergo posttranslational modification by the addition of isoprenoid group to the carboxyl terminus.8
Because mevalonic acid is the precursor of many metabolites that are involved in both DNA synthesis and cell growth, inhibition of the key enzyme HMG-CoA reductase could affect cell cycle progression and proliferation. Indeed, both in vivo and in vitro studies showed that statins (HMG-CoA reductase inhibitors) inhibit cell proliferation9,10and modulate changes in the cell cycle11and cell morpholgy.12Also, lovastatin slowed growth rate of rat neuroblastoma.12
Only a few studies evaluated the effect of HMG-CoA reductase inhibitors on the survival of blasts of acute myeloid leukemia (AML) and AML cell progenitors. Newman et al13have shown that simvastatin inhibits the proliferation of human AML progenitors that form colonies in semisolid agar to a greater extent than progenitors from normal bone marrow. The same group of scientists recently reported that simvastatin also inhibited proliferation of AML cells in vivo in severe combined immunodeficient mice.14Bansal et al15have suggested that HMG-CoA reductase inhibitors may have a therapeutic value in the treatment of leukemia.
In this study, we investigated and characterized the in vitro effects of simvastatin, an HMG-CoA reductase inhibitor, on the proliferation and colony formation of two AML cell lines (HL60 and AML-2). We also tested the hypothesis that simvastatin would potentiate the inhibitory effect of cytosine arabinoside (ARA-C) on the growth of AML blasts.
MATERIALS AND METHODS
Drugs
Simvastatin (a gift from Merck Sharp and Dohme Ltd, Petah Tikva, Israel) was dissolved in dimethyl sulfoxide (4.0 mg/mL) followed by stepwise dilution with Iscove's Modified Dulbecco's Medium (IMDM) (Gibco, Grand Island, NY). ARA-C (a gift from Upjohn Co, Kalamazoo, Mich) was dissolved in IMDM.
Cell Culture
Cells from two continuous AML cell lines were studied: HL-6016and OCI/AML-2 (a gift from Dr Minden, Ontario Cancer Institute).17The cell lines were maintained in IMDM, supplemented with fetal bovine serum (growth medium; Hyclone Laboratories, Logan, Utah).16Cells were harvested in their logarithmic growth phase for experiments.
Incubation of Cells With Drugs in Liquid Culture
HL-60 cells or AML-2 cells were seeded at 1x105/mL growth medium (in 0.8-mL portions) in 24-well tissue culture dishes. Simvastatin was added to each well (0.5-10 μg/mL) and cell cultures were incubated at 37°C in an atmosphere of 5% CO2/air for 5 days. In other experiments, 24 hours after the addition of simvastatin, ARA-C (10-9 -10-11 M) was added to the wells and cells were incubated for 4 additional days. In the controls in all the experiments, we added dimethyl sulfoxide at the same concentrations as for the corresponding experimental groups in which simvastatin was used. After incubation, cultures were assayed for surviving, proliferating, and colony-forming cells.
Cell Proliferation and 3 H-Thymidine Incorporation
Cell viability and proliferation were determined by trypan blue exclusion and by 3 H-thymidine incorporation. On the fifth day after incubation of the cells with the drug(s), 180 μL of cell suspension was transferred to each well in 96-well plates in quadruplicates. Cultures were pulsed with 1 μCi/mL of 3 H-thymidine (1 mCi/mol/L; Nuclear Research Center, Negev, Israel; final volume, 200 μL) and harvested 4 hours later onto glass filter with an automated cell harvester under hypotonic conditions. Filter discs were dried and counted in a Packard Tri-carb liquid scintillation counter (Downers Grove, Ill).
Clonogenic Assay
HL-60 cells or AML-2 cells, grown in liquid culture in the presence of the drug(s), were harvested on day 5, cell viability was determined, and 5x103 viable cells were plated in 35-mm Petri dishes containing IMDM, 15% fetal bovine serum, and 0.3% agar. Cultures were incubated at 37°C in a humidified atmosphere of 5% CO2 /air. On day 8, cell aggregates containing 40 cells or more were enumerated as colonies.
Statistical Analysis
Statistical analyses were performed using the analysis of variance with repeated measures test.
RESULTS
Effect of Exposure to Simvastatin
The number of proliferating AML blasts (suspension cultures) and colony formation (agar cultures) of both cell lines declined significantly after incubation with simvastatin. On the fifth day of culture, the number of HL-60 cells and AML-2 cells not treated with drugs (control cells) was 19±1.9x105/mL (n=6) and 10.9±0.6x105/mL (n=4), respectively. The proliferation of HL-60 cells or AML-2 cells in liquid culture in the presence of 1 μg/mL of simvastatin for 5 days was significantly inhibited, as determined by trypan blue exclusion (by 66% and 42%, respectively; P <0.05) (Figure 1A), and DNA synthesis was significantly inhibited by 90%, as determined by 3 H-thymidine incorporation (P <0.05) (Figure 1B). When higher concentrations of simvastatin (5 μg/mL) were used, viability was reduced to less than 10% (data not shown), and therefore, the rest of the experiments were performed at concentrations of 0.5 and 1 μg/mL.
The effect of simvastatin on clonogenic potential of both cell lines was evaluated in the next set of experiments. Figure 2 demonstrates that, at a concentration of 1 μg/mL, a significant decrease in clonogenic cells was observed (up to 30% of the controls; P <0.05). A similar reduction in clonogenic potential was found in both cell lines.
Combined Effect of Simvastatin and ARA-C
The effect of simvastatin in combination with increasing concentrations of ARA-C was explored in each cell line. ARA-C alone inhibited cell proliferation and colony formation in a concentration-dependent fashion in the range of concentrations used. Preincubation of AML-2 cells with simvastatin (0.5 or 1 μg/mL) followed by the addition of 10-11 -10-9 M ARA-C resulted in a marked and significant inhibition of proliferation (compared with ARA-C alone), as determined by the trypan blue exclusion test (Figure 3A), [3H] thymidine incorporation assay (Figure 3B), and clonogenic assay (Figure 3C). Similar statistically significant inhibition was also demonstrated in HL60 cells (Figure 4A-C). Thus, combination of simvastatin with ARA-C produced an antigrowth interaction that was more active than the individual compounds in inhibiting cell growth. To determine whether the combined effect of simvastatin and ARA-C was additive or synergistic, we used analysis of variance. We found the combination to be additive.
DISCUSSION
The main finding of our study is that simvastatin significantly reduced cell proliferation and clonogenic potential of two AML cell lines, HL-60 and AML-2. In addition, preincubation of the cells with simvastatin before exposure to ARA-C increased the cytotoxic effect of ARA-C significantly in an additive fashion and decreased DNA synthesis as manifested by [3H] thymidine uptake. These results lend further support to the concept of antiproliferative effects of simvastatin and may open new directions in the clinical management of AML.
The precise mechanism of action of statins on the proliferation of malignant cells is unknown. Farnesyl residues derived from the conversion of mevalonate to cholesterol have a major regulatory role in cell growth.18It was repeatedly shown that posttranslational fernesylation of ras protein (P21 ras protein) and the nuclear membrane protein lamin B is required for their attachment to the cell membrane and is essential for oncogenic transformation.5,19,20The inhibition of mevalonate synthesis would limit the availability of farnesyl pyrophosphate and thereby limit the cell growth promotion of the ras oncoprotein.21However, Clutterbuck et al14have shown recently that ketoconazole, which inhibits cholesterol biosynthesis after farnesyl pyrophosphate synthesis, had a similar effect to simvastatin on HL-60 colony development. Furthermore, the clonogenicity of N-ras mutated primary AML cells was not more sensitive to simvastatin than those of a population without the mutation. It was concluded that inhibition of AML cell proliferation by simvastatin may be independent of the ras signaling pathway. Because statins could interfere with the function of more than 40 other proteins that are normally isoprenylated, it is possible that the inhibition of proliferation by simvastatin may be related to the interference of function of other proteins.22
ARA-C is the most important agent in the treatment of AML. In this study, the combination of simvastatin and ARA-C enhanced the antiproliferative effect of the drugs. In previous in vitro studies, HMG-CoA reductase inhibitors had a synergistic or additive effect when used in combination with various antineoplastic agents (BCNU, interferon-β, mitomycin-C) in different malignant cell lines (hepatoma cell culture-4, human glioma cells).23,24Simvastatin was used in our study at concentrations that are usually achieved in the plasma of patients treated for hypercholesterolemia (0.1-10 μm). At these concentrations, simvastatin alone inhibited cell proliferation significantly. The combination of the drugs induced a major increase in the inhibition of cell proliferation over each drug alone. This may allow the combination of lower concentrations of antineoplastic agents with therapeutic doses of simvastatin to achieve the same inhibition of cell growth. It also may allow the increase of neoplastic cell kill over what is achieved with chemotherapy alone, because it was already shown that AML progenitors have greater sensitivity to the antiproliferative properties of simvastatin compared with normal bone marrow progenitors.13
The possible effects of HMG-CoA reductase inhibitors on tumor growth have drawn much attention. An inhibitory effect was demonstrated in rats,25mice,26or severe combined immunodeficient mice implanted with human tumors.27A phase I study in patients with high-grade glioma was disappointing.28We suggest that the model we describe be used in future clinical studies in patients with leukemia resistant to chemotherapy. The possible therapeutic options include either simvastatin alone or in combination with chemotherapeutic medications.