Identifying strategies to increase cancer cell kill while sparing normal tissue is critically important in cancer chemotherapy. Choline kinase (Chk), the enzyme that converts choline to phosphocholine (PC), is elevated in cancer cells and presents a novel target for increasing cell kill. Here, we have examined the effects of transiently down-regulating Chk by small interfering RNA against Chk (siRNA-chk) on PC and total choline-containing compound (tCho) levels and on the viability/proliferation of estrogen receptor–negative and estrogen receptor–positive breast cancer cell lines and a nonmalignant mammary epithelial cell line. We investigated the effects of combination treatment with transient siRNA-chk transfection and the anticancer drug 5-fluorouracil (5-FU) in those cell lines. Microarray analysis of the invasive estrogen receptor–negative MDA-MB-231 cell line was done to characterize molecular changes associated with Chk down-regulation. Chk down-regulation decreased PC and tCho levels in the malignant cell lines, whereas the cell viability/proliferation assays detected a decrease in proliferation in these cells. In contrast, Chk down-regulation had an almost negligible effect on PC and tCho levels as well as cell viability/proliferation in the nonmalignant cell line. A combination of siRNA-chk with 5-FU treatment resulted in a larger reduction of cell viability/proliferation in the breast cancer cell lines; this reduction was evident to a much lesser degree in the nonmalignant cells. Microarray analysis showed that Chk down-regulation affected 33 proliferation-related genes and 9 DNA repair–related genes. Chk down-regulation with siRNA-chk may provide a novel alternative to enhance the effect of anticancer drugs in malignant cells. [Cancer Res 2007;67(23):11284–90]
- choline phospholipid metabolism
- choline kinase
- breast cancer cells
Although there is no dearth of cytotoxic therapies available to kill cancer cells, most of these damage normal cells as well as cancer cells, resulting in unfortunate side effects such as diarrhea, nausea, granulocytopenia, and cardiotoxicity, to name a few ( 1, 2). The ideal therapy would target cancer cells while sparing normal tissue. Targeting pathways or molecules that are up-regulated in cancer cells but not in normal cells can provide new strategies for selective therapy to use either singly or in combination with conventional chemotherapeutic agents.
An increase of cellular phosphocholine (PC) and total choline-containing compounds (tCho) has consistently been observed in cancer cells and tissue ( 3– 5) and is closely related to malignant transformation, invasion, and metastasis ( 3, 6). Choline kinase (Chk), a cytosolic enzyme that catalyzes the phosphorylation of choline to form PC by ATP in the presence of magnesium ( 7), is one of three enzymes that, along with phosphatidylcholine-specific phospholipase C and CTP:phosphocholine cytidylyltransferase, can lead to increased PC levels ( 8, 9). Increased expression of Chk has been observed in human lung, colon, prostate, and breast cancer as well as derived epithelial and hemopoietic cell lines ( 10– 18). Carcinogens such as polycyclic aromatic hydrocarbons and 1,2-dimethylhydrazine induce liver and colon cancers with increased Chk activity ( 17). Chk can also be induced by activating agents including hormones, growth factors, oncogenes, and carcinogens ( 19– 21). Because of its increased expression and activity in cancer, Chk has been proposed as a target for antitumor therapy ( 16, 22, 23). Chk inhibition using the pharmacologic inhibitor MN58b was recently shown to reduce proliferation of cancer cells in vitro and xenografts in vivo ( 22, 24– 27). To avoid the side effects associated with pharmacologic inhibition, we used small interfering RNA (siRNA-chk) to target Chk and showed that both transient transfection and stable expression of siRNA against Chk induced differentiation and reduced proliferation in breast cancer cells ( 15).
Our purpose in this study was twofold. One was to examine the effect of transiently down-regulating Chk in normal and malignant human mammary epithelial cells (HMEC) on cell viability and proliferation, and the second was to determine if combining Chk down-regulation with 5-fluorouracil (5-FU) treatment increased cell kill in cancer cells. 5-FU is commonly used to treat a variety of cancers including breast, head and neck, and colon ( 28, 29). One of the main mechanisms of action of 5-FU is the inhibition of thymidylate synthase by the 5-FU metabolite 5-fluoro-2′-deoxyuridine-5′-monophosphate, which leads to cell death ( 28). This effect is not specific to tumor cells, making 5-FU a highly toxic drug, with a narrow margin of safety, which can cause severe hematologic toxicity and gastrointestinal hemorrhage ( 30). Several other chemotherapeutic agents such as mitomycin C have similar toxicity, and therefore combination therapies with agents that increase cell kill in cancer cells but not in normal cells would be attractive to reduce toxicity and maintain or increase efficacy.
In this study, we examined the effects of transient down-regulation of Chk, using siRNA-chk transfection, on PC and tCho levels and on the viability/proliferation of two human breast cancer cell lines, MDA-MB-231 (estrogen receptor/progesterone receptor negative) and MCF-7 (estrogen receptor/progesterone receptor positive), and a nonmalignant HMEC line, MCF-12A. We investigated the effect of combining transient siRNA-chk transfection with the anticancer drug 5-FU on cell viability/proliferation in these cell lines as well. Because MDA-MB-231 cells were the more aggressive, microarray analysis of these cells was done to characterize molecular changes associated with Chk down-regulation. We chose to examine transient down-regulation because this would be the most likely option in the clinical setting using liposomal or viral mediated delivery of siRNA or short hairpin RNA.
Results from magnetic resonance spectroscopy of cell extracts and Chk expression levels following transient transfection of siRNA-chk showed that Chk plays an important role in generating the high PC and tCho levels observed in MCF-7 and MDA-MB-231 breast cancer cells but not in the nonmalignant HMEC line MCF-12A. The cell viability/proliferation assay showed that Chk down-regulation reduced cell viability/proliferation of both estrogen receptor/progesterone receptor–negative and estrogen receptor/progesterone receptor–positive breast cancer cells but not nonmalignant MCF-12A cells. Microarray analysis of MDA-MB-231 cells showed that Chk down-regulation affected 33 proliferation-related genes and 9 DNA repair–related genes in these cells. Combined treatment with siRNA-chk transfection and 5-FU reduced the cell viability/proliferation to levels that were significantly lower than either treatment alone in both breast cancer cell lines. In contrast, although the nonmalignant HMECs were sensitive to 5-FU treatment, the combination treatment resulted in a smaller reduction of cell viability/proliferation compared with the malignant cells. These data show that for a comparable dose of 5-FU, a greater reduction in cell viability/proliferation was achieved with transient siRNA-chk transfection in breast cancer cells compared with nonmalignant HMECs.
Materials and Methods
Cell culture. MDA-MB-231, an estrogen receptor–negative metastatic human breast cancer cell line, was grown in RPMI 1640 supplemented with 9% fetal bovine serum (FBS) and antibiotics (90 unit/mL penicillin and 90 μg/mL streptomycin). MCF-7, an estrogen receptor–positive poorly metastatic human breast cancer cell line, was cultured in Eagle's MEM supplemented with 9% FBS and antibiotics. MCF-12A, a spontaneously immortalized nonmalignant human mammary epithelial cell line, was cultured in DMEM-Ham's F12 medium supplemented as described previously ( 3). All cell lines were obtained from American Type Culture Collection (ATCC) and were maintained in a humidified atmosphere with 5% CO2 in air at 37°C.
RNA interference experiments. siRNA-chk was designed as previously described ( 15) and purchased from Dharmacon. Approximately 98 nmol/L of siRNA-chk were transfected for 48 h using Oligofectamine and Opti-MEM (both from Invitrogen) based on the observed decrease of Chk message using reverse transcription-PCR analysis within this time ( 15).
Dual-phase cell extraction and magnetic resonance spectroscopy study. Cells were transfected with siRNA-chk for 48 h and water-soluble as well as lipid extracts were obtained from ∼2 × 107 control and siRNA-chk–treated cells using the dual-phase extraction method ( 15). Control cell extracts were obtained from cells without siRNA-chk treatment. Cell extracts were resuspended in 0.6-mL deuterated water for magnetic resonance spectroscopy analysis. Five microliters of 0.75% (w/w) 3-(trimethylsilyl)propionic-2,2,3,3-d4 acid sodium salt (TMSP) in deuterated water were used as an internal standard. Fully relaxed 1H nuclear magnetic resonance spectra of the water-soluble extracts were acquired on a Bruker Avance 500 spectrometer (Bruker BioSpin Corp.) as previously described ( 15). Signal integrals of N-(CH3)3 of free choline (Cho), PC, and glycerophosphocholine (GPC) from within the 3.20 to 3.24 ppm region were determined and normalized to cell numbers and compared with the standard. To determine concentrations, peak integration (In) from 1H spectra for PC, GPC, Cho, and tCho (PC + GPC + Cho) was compared with that of the internal standard TMSP according to the following equation:
In this equation, [metabolite] is the molar concentration per cell of the metabolite expressed as mol/cell, ATMSP is the number of moles of TMSP in the sample, and Ncell is the cell number. Because the number of protons contributing to the signal of all the choline (Cho) metabolites at 3.20 to 3.24 ppm and to the TMSP peak at 0 ppm is the same, correction for differences in the number of protons was not required.
Immunoblot analysis. Approximately 3 × 106 cells were grown in culture medium overnight and transfected with siRNA-chk for 48 h. Cells were scraped into radioimmunoprecipitation assay buffer [50 mmol/L Tris (pH 7.4), 150 mmol/L NaCl, 1 mmol/L EDTA, 1% Triton-100, 1% sodium deoxycholate, 1 mmol/L phenylmethylsulfonyl fluoride, 0.1% SDS, and complete EDTA-free protease inhibitor cocktail (Roche)]. Cell lysates were incubated on ice for 15 min and spun down at 16,000 × g (Refrigerated Centrifuge 5415 R, Eppendorf). Protein samples from whole-cell extracts were resolved on one-dimensional 10% SDS-PAGE gels and transferred onto immunoblot polyvinylidene fluoride membranes (Bio-Rad) using a semidry transfer unit (Bio-Rad). The membranes were incubated overnight in blocking solution [1% dry milk in PBS with 0.1% Tween 20] with custom-made polyclonal Chk antibody (Proteintech Group, Inc.; ref. 15) with an appropriate dilution. Anti–β-actin antibody (Molecular Probes) was used for equal loading assessment. Secondary antibody was horseradish peroxidase–conjugated antimouse immunoglobulin G (Vector Laboratories). Reactions were revealed using SuperSignal West Pico Substrate (Pierce Biotech.) recorded on Blue Bio film (Denville Scientific). Chk was detected at ∼48 kDa.
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. Cells were grown in culture medium without phenol red. Approximately 4,000 cells were seeded in each well of a 96-well plate and cultured overnight. Twenty-four hours later, siRNA-chk was transfected transiently for 48 h, as previously described ( 15). No treatment or Oligofectamine without siRNA was used as negative control. For combination experiments, 5-FU (final concentration, 5 μg/mL) was added 24 h after transfection without changing medium. Forty-eight hours after transfection and/or 24 h after 5-FU treatment, cells were cultured for another 3 days in fresh culture medium. Cell viability/proliferation was evaluated with the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) Cell Proliferation Assay (ATCC) and values were compared with those obtained from untreated cells.
GeneChip microarray gene expression assay. Total cellular RNA was isolated from ∼107 siRNA-chk–treated or control MDA-MB-231 cells as described above using the RNeasy Mini Kit (Qiagen, Inc.) and QIAshredder homogenizer spin columns (Qiagen) as previously described ( 14, 31). The microarray hybridization was done at the JHMI Microarray Core Facility (Dr. Francisco Martinez Murillo, Johns Hopkins University School of Medicine) using the Human Genome U133 Plus 2.0 Array (Affymetrix, Inc.), which contains >47,000 transcripts, and the Affymetrix GeneChip platform. GeneChips were analyzed by fluorescence detection using the Agilent GeneArray Scanner as previously described ( 14, 31). Data acquisition was done using the Micro Array Suite 5.0 software by Affymetrix as previously described ( 14, 31). Experiments were done in duplicate. To estimate the gene expression signals, data analysis was conducted on the chips' cell intensity (CEL) file probe signal values at the Affymetrix probe pair [perfect match (PM) probe and mismatch (MM) probe] level using statistical techniques and package Robust Multiarray Analysis ( 14, 31). The criterion of the posterior probability >0.5, which means that the posterior probability is larger than chance, was used to produce differentially expressed gene lists. All computations were done under the R package EBarrays environment as previously described ( 32– 34).
Statistical analysis. Data were expressed as mean ± SE. The statistical significance of differences in cell metabolite levels and MTT assay result was determined using an unpaired t test (two-tailed). P ≤ 0.03 for cell metabolite levels and P ≤ 0.01 for MTT assay were considered to be significant.
Chk protein level was down-regulated by transient siRNA-chk transfection. The immunoblot assay results with Chk antibody showed significant reduction of Chk protein expression levels in MCF-12A, MCF-7, and MDA-MB-231 cells after transient siRNA-chk transfection ( Fig. 1 ). Chk levels in nonmalignant MCF-12A cells were much lower than the two breast cancer cell lines. This assay was repeated thrice and the results were reproducible. Significantly lower Chk protein levels after transient siRNA-chk transfection confirmed successful transfection and down-regulation of Chk levels.
Magnetic resonance spectroscopy of a nonmalignant HMEC line and two human breast cancer cell lines after transient siRNA-chk transfection. Magnetic resonance spectra showed that basal PC levels in the nonmalignant MCF-12A cells were much lower than in breast cancer cells ( Fig. 2 ). A comparison of PC, GPC, and tCho levels of cells treated with siRNA-chk versus control cells ( Fig. 3 ) revealed a significant reduction of PC and tCho in the malignant MDA-MB-231 breast cancer cells following transient transfection with siRNA-chk compared with control cells. A significant reduction of PC in the estrogen receptor–positive breast cancer cell line MCF-7 was also observed after siRNA-chk transfection. There were no statistically significant changes in PC, tCho, or GPC in nonmalignant MCF-12A cells following transient siRNA-chk transfection ( Fig. 3). We carried out control studies with glyceraldehyde-3-phosphate dehydrogenase siRNA to rule out the possibility of poor transfection of these cells with siRNA (data not shown).
Cell viability/proliferation assay of a nonmalignant HMEC line and two human breast cancer cell lines after transient siRNA-chk transfection. Both breast cancer cell lines MDA-MB-231 and MCF-7 showed significant reduction of viability/proliferation following transient transfection using siRNA-chk alone compared with untreated cells ( Fig. 4 ). In contrast, the reduction of viability/proliferation was negligible in nonmalignant MCF-12A cells ( Fig. 4).
Cell viability/proliferation assay of a nonmalignant HMEC line and two human breast cancer cell lines after treatment with 5-FU alone or in combination with siRNA-chk transfection. Treatment with 5-FU alone resulted in a reduction of viability/proliferation of all cell lines ( Fig. 4). MCF-12A, which has the lowest expression level of Chk protein, showed the highest reduction in cell viability/proliferation by 5-FU alone. The combination treatment of 5-FU and siRNA-chk down-regulation resulted in a 2-fold reduction of cell viability/proliferation in breast cancer cell lines compared with 5-FU treatment alone ( Fig. 4), but only a 10% additional reduction in the nonmalignant cell line. As a result, for a comparable dose of 5-FU, the reduction in cell viability/proliferation was as much or higher for the breast cancer cells compared with the nonmalignant cells, although the nonmalignant cells were more sensitive to 5-FU alone.
GeneChip microarray gene expression assay for proliferation- and DNA repair–related genes. Table 1 shows that 16 proliferation-related genes were significantly overexpressed and 17 proliferation-related genes significantly underexpressed following siRNA-chk transfection in MDA-MB-231 cells. Nine DNA repair–related genes were significantly differentially expressed following transient siRNA-chk transfection in MDA-MB-231 cells as shown in Table 2 . However, none of the genes directly involved in 5-FU metabolism, such as dihydropyrimidine dehydrogenase, thymidine phosphorylase, thymidylate synthase, human concentrative nucleoside transporter 1, or uridine phosphorylase, among others ( 28), were differentially expressed in MDA-MB-231 cells transiently transfected with siRNA-chk compared with control cells (data not shown).
We have previously shown that siRNA-chk induced differentiation and reduced cell proliferation markers in breast cancer cells ( 15). There are at least three isoforms of Chk (chk-α1, chk-α2, and chk-β) in mammalian cells ( 7). Our siRNA-chk was designed using the mRNA sequences transcribed from the chk-α and chk-β genes to allow down-regulation of both of these Chk mRNAs. Our immunoblot assay confirmed successful transient transfection and down-regulation of Chk protein expression levels in all three cell lines used. Magnetic resonance spectroscopy studies confirmed the successful down-regulation of siRNA-chk and showed the significant dependence of the elevated PC levels in breast cancer cells on Chk as evident from the low PC following transient siRNA-chk transfection. Because MCF-12A cells have low basal Chk protein and PC levels, the effect of Chk down-regulation by siRNA-chk was not significant for PC levels in nonmalignant MCF-12A cells. Another possibility may be that PC in MCF-12A cells may not be primarily dependent on Chk and may be derived through other enzymes. The differences in Chk expression levels between malignant and nonmalignant cells obtained here are consistent with data from our previous study ( 15).
Cell viability/proliferation was significantly lower in both estrogen receptor/progesterone receptor–positive and estrogen receptor/progesterone receptor–negative breast cancer cells following transient siRNA-chk transfection. Importantly, transient transfection of nonmalignant MCF-12A cells with siRNA-chk had a negligible effect on cell viability/proliferation. These data suggest that transient Chk silencing may reduce cell viability/proliferation in cells with high Chk such as malignant cells but may not affect nonmalignant cells with low Chk. These data support exploring transient Chk silencing as a broad-spectrum chemotherapeutic agent for cancer cells.
The reduction of cell viability/proliferation following a single dose of 5-FU was highest in the nonmalignant cells with the lowest Chk expression. Malignant cells with higher Chk expression were more resistant to 5-FU treatment. Combined treatment of both breast cancer cell lines with siRNA-chk transfection and 5-FU reduced viability/proliferation to levels that were similar for both estrogen receptor/progesterone receptor–positive and estrogen receptor/progesterone receptor–negative cells and ∼2-fold lower than either treatment alone. Although the nonmalignant MCF-12A cells had the most profound reduction of cell viability/proliferation with 5-FU treatment, combining 5-FU treatment with siRNA-chk transfection resulted in a reduction of cell viability/proliferation in the breast cancer cells that was as much or even more than that in the nonmalignant cells. The increase of cell kill was most likely additive rather than synergistic because the microarray data showed no effect with Chk down-regulation on 5-FU–related genes such as thymidylate synthase.
The microarray analysis detected changes in the expression of 33 proliferation-related genes with Chk down-regulation. One gene significantly influenced by Chk down-regulation was MXI1 (−3.8-fold), which is a c-Myc antagonist and is induced by hypoxia-inducible factor 1. Whereas MXI1 is considered a potential tumor suppressor, it also protects cells from c-Myc-induced apoptosis ( 35, 36). Overexpression of both negative (CDKN1B, CDKN1C, NF1, and SMAD1) and positive (EDN1 and FGFR1OP) regulators of cell proliferation and underexpression of positive (PBEF1, DNAJA2, and SSR1) regulators of cell proliferation were also observed. EDN1 is a potent vasoconstrictor peptide, which is also involved in differentiation, apoptosis, and matrix metalloprotease expression ( 37). In a previous study, indomethacin-mediated down-regulation of PC levels in breast cancer cells was also associated with overexpression of EDN1 as detected by microarray analysis ( 31). Nine DNA repair genes were affected by Chk down-regulation. One of the most significant effects was the 3-fold underexpression of RAD23B (−3.2-fold). The protein encoded by this gene contains an NH2-terminal ubiquitin-like domain that can bind the proteosome ( 38, 39). Rad23/proteosome binding is necessary for efficient nucleotide excision repair, which is the primary mechanism for removing UV-induced DNA lesions ( 40). It has been shown that Rad23 has additional functions that are required for cell survival ( 40). We are currently exploring the role of Chk in multidrug resistance and DNA repair to determine if increased expression of Chk provides protection to cells during chemotherapy.
In conclusion, the down-regulation of Chk by transient transfection of siRNA-chk resulted in a significant reduction of cell proliferation in both estrogen receptor/progesterone receptor–positive and estrogen receptor/progesterone receptor–negative malignant human breast cancer cell lines but not in nonmalignant HMECs. A combination of siRNA-chk transfection and 5-FU treatment resulted in increased cell kill in both cancer cell lines compared with each of these treatments given alone. Although the nonmalignant cells were the most sensitive to 5-FU treatment, the combination treatment resulted in comparable or even higher reduction of cell viability/proliferation in the cancer cells compared with the nonmalignant cells. Microarray analysis showed that Chk down-regulation affected 33 proliferation-related genes and 9 DNA repair–related genes. Our study supports the possibility that Chk silencing may provide a safe and effective therapy against cancer and be a novel alternative to enhance the effect of anticancer drugs. The development of therapy using RNA interference targeting Chk may be a useful and selective treatment strategy for breast cancer.
Grant support: NIH grant P50 CA103175.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Y. Mironchik for laboratory support and Drs. V.P. Chacko and J. Zhang for technical support for magnetic resonance spectroscopy experiments.
- Received July 20, 2007.
- Accepted September 24, 2007.
- ©2007 American Association for Cancer Research.