Choline kinase is overexpressed in breast cancer cells and activated by oncogenes and mitogenic signals, making it a potential target for cancer therapy. Here, we have examined, for the first time, the effects of RNA interference (RNAi)–mediated down-regulation of choline kinase in nonmalignant and malignant human breast epithelial cell lines using magnetic resonance spectroscopy (MRS) as well as molecular analyses of proliferation and differentiation markers. RNAi knockdown of choline kinase reduced proliferation, as detected by proliferating cell nuclear antigen and Ki-67 expression, and promoted differentiation, as detected by cytosolic lipid droplet formation and expression of galectin-3. The functional importance of RNAi-mediated choline kinase down-regulation on choline phospholipid metabolism was confirmed by the significant reduction of phosphocholine detected by MRS. These results strongly support the targeting of choline kinase in breast cancer cells with RNAi and show the potential ability of noninvasive MRS to detect and evaluate future treatments incorporating such strategies.

Overexpression and increased activity of choline kinase in malignant cells and tumors (15) and oncogene-mediated increase in the activity and basal levels of choline kinase (610) have been reported in several studies. Both increased enzymatic activity as well as overexpression of choline kinase correlated strongly with histologic tumor grade and negative estrogen receptor status in breast carcinomas (1). Because choline kinase converts free choline to phosphocholine, magnetic resonance studies over the past decade have also consistently detected an elevation of phosphocholine and total choline-containing compounds in cancer cells in vivo and in vitro (1118).

Ras oncogene transformation has been related to choline kinase stimulation and to an increase of phosphocholine levels (6, 7, 9, 19). The effect of ras signaling on choline kinase activity is mediated through a combination of its effectors Ral-GDP dissociation stimulator, a nucleotide exchange factor, and phosphatidylinositol 3-kinase, a lipid kinase, downstream of Ras (8). Transformation of murine NIH3T3 fibroblasts with other oncogenes, such as v-raf, v-mos, and src, were also found to increase basal choline kinase activity and cellular phosphocholine levels (20). Choline kinase was also activated by serum (21), hormones (22, 23), or other mitogenic/growths factors, such as insulin (21, 24), platelet-derived growth factor (PDGF; ref. 25), fibroblast growth factor (FGF; ref. 25), epidermal growth factor (24), prolactin (26), and estrogens (22, 27). Expression of human choline kinase in fibroblasts increased the mitogenic potential of insulin, insulin-like growth factor-I, FGF, and PDGF (28). Stress induced chemically with carcinogens was also found to induce choline kinase expression (23, 29), possibly due to a xenobiotic responsive element in the putative promoter region of the chk-α gene (29).

These observations have led to pharmacologic inhibitors of choline kinase being explored as potential antitumor agents (13, 18, 20). Strong antitumor activity of these inhibitors has been shown in vivo against breast, colon, and epidermoid cancer xenografts (3, 10, 30). Although the toxicity and side effects of these drugs have largely been minimized with newly designed chemical compounds (20), the long-term down-regulation of choline kinase and its effect on the tumor phenotype are as yet unknown. RNA interference (RNAi) provides a molecular approach to down-regulate the expression of specific target genes in mammalian cells (3133), without the associated side effects of chemical agents, especially in long-term studies.

In this study, RNAi knockdown of choline kinase was investigated, for the first time, in a nonmalignant immortalized human mammary epithelial cell (HMEC) line and in malignant breast cancer cell lines. The effects of choline kinase down-regulation were assessed with transiently transfected small interfering RNA (siRNA) specific for choline kinase mRNA and in a breast cancer cell line stably expressing short hairpin RNA (shRNA) specific for choline kinase mRNA. Functional evaluation of RNAi with magnetic resonance spectroscopy (MRS) of cell extracts, together with molecular analyses of choline kinase expression and markers of proliferation and differentiation, showed that RNAi-mediated down-regulation of choline kinase had a profound effect on choline kinase expression, choline phospholipid metabolism, proliferation, and differentiation in breast cancer cells. These studies strongly support the use of specific targeting of choline kinase with RNAi for breast cancer treatment and show the feasibility of using MRS to detect such targeting in preclinical models in vivo and potentially in the clinical setting.

Cell lines. An immortalized nonmalignant human mammary epithelial cell line, MCF-12A, was used as model of normal human mammary epithelium (34). MCF-7, an estrogen-sensitive poorly metastatic breast cancer cell line (35), and MDA-MB-231 and MDA-MB-435, two metastatic human breast cancer cell lines, were used as models of malignant human breast cancer cells (17). All cell lines (American Type Culture Collection, Rockville, MD) were maintained as previously described (17, 34, 35).

RNA interference experiments. siRNA specific for choline kinase (siRNA-chk) was designed with the sequence 5′-CAUGCUGUUCCAGUGCUCC-3′ and purchased as a duplex (Dharmacon, Lafayette, CO). Approximately 106 cells at 40% to 50% confluence were transfected with 2.6 nmol of the annealed siRNA-chk (Dharmacon) using 772 μL oligofectamine (Invitrogen, Carlsbad, CA) in Opti-MEM (Invitrogen). Depending upon the number of cells required, proportionately larger or smaller amounts of siRNA-chk, oligofectamine, and Opti-MEM were applied. Cells were treated with siRNA-chk for 48 hours, because this incubation period resulted in the most effective reduction of choline kinase levels as determined by reverse transcriptase-PCR (RT-PCR) and Western blotting.

Generation of stable clones. An oligonucleotide cDNA insert encoding a shRNA specific for down-regulating choline kinase mRNA was designed using the RNA Oligo Retriever software (http://katahdin.cshl.org:9331/RNAi/; ref. 33). The oligonucleotides selected encoded the same region of the chk gene used for synthetic siRNA transfection experiments. The oligonucleotides were synthesized, purified (DNA/RNA Synthesis Core Facility, Johns Hopkins University Bloomberg School of Public Health), annealed, and ligated into the pSHAG vector containing the U6 promoter for RNA Polymerase III catalyzed transcription (33). The U6 promoter plus shRNA sequence for choline kinase was cut from this vector and ligated into the multiple cloning site of a pCR3.1 vector (Invitrogen) devoid of any promoter. This final construct, referred to as U6-shRNA-chk vector, was transfected into malignant MDA-MB-231 breast cancer cells selected with G418 (Invitrogen). Forty-five stable clones expressing this vector were screened using quantitative real-time RT-PCR (qRT-PCR) analysis and Western blotting using choline kinase antibody. Selection of the two clones used in these studies was based on the significantly low amounts of choline kinase mRNA and protein levels in these two clones compared with parental MDA-MB-231 cells.

Total RNA isolation, reverse transcriptase-PCR, and quantitative real-time reverse transcriptase-PCR analysis. Total cellular RNA was isolated from ∼3 × 106 MDA-MB-231 or MCF-12A cells using the QIAshredder homogenizer spin columns (Qiagen, Inc., Valencia, CA) and RNeasy Mini Kit (Qiagen), and its concentration, purity, and integrity was assessed according to the manufacturer's protocol. Reverse transcription was done using SuperScript II (Invitrogen) according to the manufacturer's protocol. One microliter of the resulting cDNA reaction mix was used for PCR. The human ribosomal protein 36B4 was used as internal loading control for conventional as well as quantitative RT-PCR. The PCR primers for amplifying 36B4 and choline kinase were as follows for both the conventional RT-PCR and qRT-PCR: 36B4 sense strand, 5′-GATTGGCTACCCAACTGTTGCA-3′; antisense strand, 5′-CAGGGGCAGCAGCCACAAAGGC-3′; choline kinase sense strand, 5′-ATCCCACCAAGAAACAACAGC-3′; antisense strand, 5′-TGGTGGAAATAGGCATCAAAG-3′. Conventional PCR reactions were carried out using the Rapid Cycler (Idaho Technology, Salt Lake City, UT), and a buffer containing 50 mmol/L Tris (pH 8.3), 2.5 mg/mL bovine serum albumin (BSA), and 2 mmol/L magnesium chloride (Idaho buffer 10× with BSA and 20 mmol/L magnesium chloride, Idaho Technology). The denaturation temperature was 94°C, the annealing temperature was 56°C, and the extension temperature was 74°C. Thirty amplification cycles were used. qRT-PCR was done using the iQ SYBR Green Supermix (Bio-Rad, Hercules, CA) and the iCycler (Bio-Rad). PCR reactions were also done on RNA samples without prior reverse transcription to ensure that DNA contamination of the samples had not occurred.

Protein lysates, gel electrophoresis, and Western blotting. Approximately 3 × 106 cells at 70% to 80% confluence were homogenized with lysis buffer containing a protease inhibitor cocktail (P8340, Sigma-Aldrich, St. Louis, MO) as previously described (36). Twenty micrograms of total protein, as determined by a modified Lowry assay (Bio-Rad), was loaded in each lane, and two lanes were loaded with molecular weight standard (BenchMark, Life Technologies, Rockville, MD). Ten percent SDS-PAGE and Western blotting were done as previously described (36). Membranes were incubated for 1 hour with an appropriate dilution of primary antibody followed by incubation with horseradish peroxidase–conjugated second-step antibody (Amersham, Arlington Heights, IL), and visualization using the Supersignal West Pico chemiluminescent substrate kit (Pierce Biotechnology, Rockford, IL) recorded on Blue Bio film (Denville Scientific, Metuchen, NJ). The films were scanned and densitometry was done using the Gel-analysis-tool in ImageJ (NIH, Bethesda, MD).

Choline kinase antibody. The polyclonal choline kinase antibody (IgG) used here was raised against a hydrophilic synthetic peptide (Proteintech Group, Inc., Chicago, IL) specific for choline kinase. The peptide was designed by employing Hopp-Woods hydropathy plots of human choline kinase and visualized with the Swiss-Pdb Viewer (http://us.expasy.org/spdbv/text/download.htm) to assess surface exposure of choline kinase amino acids. The peptide representing the amino acids 73 to 90 of choline kinase was synthesized at 90% purity as verified by mass spectrometry, conjugated with keyhole limpet hemocyanin, and used to immunize two rabbits using a standard immunization protocol (Proteintech Group). Antiserum specificity was assessed by ELISA against the synthetic peptide. Antibodies were isolated and purified by affinity chromatography using the peptide as affinity ligand. Using this antibody, choline kinase was detected in Western blots at an apparent molecular weight of 48 kDa.

Dual-phase extraction, magnetic resonance data acquisition, and processing. Approximately 108 cells were harvested and both lipid and water-soluble cell extract fractions were obtained using a dual-phase extraction method based on methanol/chloroform/water (1:1:1, v/v/v) as previously described (18, 37). Briefly, cells were harvested by trypsinization, and a small aliquot was counted for quantitation. The cells were washed twice with 10 mL of saline and pooled into a glass centrifuge tube. Four milliliters of ice-cold methanol were added to the pelleted cells, vigorously vortexed, and kept on ice for 10 minutes. Four milliliters of chloroform were added, vortexed, and kept on ice for 10 minutes. Finally, 4 mL of water were added, vortexed, and kept at 4°C overnight for phase separation. The samples were centrifuged for 30 minutes at 35,000 × g at 4°C, and the phases were carefully separated. The water/methanol phase containing the water-soluble cellular metabolites, such as choline, phosphocholine, and glycerophosphocholine, was treated with 10 mg of chelex (Sigma-Aldrich) for 10 minutes on ice to remove divalent cations. The chelex beads were then removed. Methanol was removed by rotary evaporation, and the remaining water phase was lyophilized and stored at −20°C. The chloroform phase containing the cellular lipids, such as phosphatidylcholine, was dried in a stream of N2 and stored under N2 at −20°C. The samples were dissolved in deuterated solvents containing 3-(trimethylsilyl)propionic-2,2,3,3-d4 acid (TSP; Sigma-Aldrich) in the case of water-soluble fractions or tetramethylsilane (TMS; Cambridge Isotope Laboratories, Inc., Andover, MA) in the case of lipid fractions to serve as concentration standards and chemical shift references. Fully relaxed 1H magnetic resonance spectra of the extracts were acquired on a Bruker Avance 500 spectrometer operating at 11.7 Tesla (Bruker BioSpin Corp., Billerica, MA) using a 5-mm HX inverse probe as previously described (18). Magnetic resonance spectra were analyzed using Bruker XWIN-NMR 3.5 software (Bruker BioSpin). Integrals of the N-(CH3)3 signal of free choline at 3.209 ppm, phosphocholine at 3.227 ppm, and glycerophosphocholine at 3.236 ppm in the 1H magnetic resonance spectra of water-soluble metabolites, and of phosphatidylcholine at 3.220 ppm, and the methylene groups in fatty acids (Fmix) at 1.245 to 1.364 ppm in the 1H magnetic resonance spectra of lipids, were determined and normalized to cell size and number as previously described (17, 18). Briefly, the signal integrals of the N-(CH3)3 groups of choline, phosphocholine, glycerophosphocholine, and the standard TSP in the water-soluble 1H magnetic resonance spectra, and phosphatidylcholine and the standard TMS in the lipid 1H magnetic resonance spectra were determined using XWIN-NMR 3.5. The concentration of each metabolite in mM was calculated according to the following equation:

\[[\mathrm{metabolite}]\ =\ (\mathit{I}\mathrm{_{metabolite}}\ {\times}\ [\mathrm{standard}])/(\mathit{I}\mathrm{_{standard}}\ {\times}\ \mathrm{cell\ number}\ {\times}\]
\[\mathrm{cell\ volume})\]

In this equation, [metabolite] represents the intracellular concentration of the metabolite of interest expressed in mmol/L, Imetabolite represents the signal integral of the metabolite of interest divided by the number of protons, and ([standard]) represents the amount of TSP (water-soluble metabolites) or TMS (lipids) used in mol divided by the number of protons. The number of cells in each sample (cell number) was counted before extraction. The cell volumes were determined as previously described (17).

Differentiation assays: Nile red staining, fluorescence microscopy, and marker proteins. Cells were grown on Permanox chamber slides (Nalge Nunc, Naperville, IL) to 60% to 70% confluence, treated as described above in the case of siRNA-chk experiments, washed with PBS, and fixed with 3% (w/v) paraformaldehyde. Cells were washed with PBS and incubated with a 1:1,000 dilution in PBS of a 1 mg/mL stock solution of Nile red (9-diethylamino-5H-benzo[α]phenoxazine-5-one; Sigma-Aldrich) in acetone for 10 minutes at room temperature. Nile red (Sigma-Aldrich) stains lipid droplets (3840), which are a marker of differentiation in epithelial cells (41, 42). Cell nuclei were counterstained with Hoechst H-33342 (Molecular Probes, Eugene, OR). Cells were washed and mounted using Faramount aqueous mounting medium (DakoCytomation, Carpinteria, CA). Fluorescence microscopy was done with a Zeiss LSM 510 META confocal laser-scanning microscope (Carl Zeiss, Inc., Thornwood, NY) using a Plan-Apochromat 63×/1.4 oil immersion lens (Zeiss). Nile red and H-33342 were excited at 488 and 405 nm, respectively, and fluorescence emission was detected with photomultipliers using 530-nm long pass and 420- to 480-nm band pass filters, respectively. Confocal Z-sections of 1-μm thickness containing the maximal number of lipid droplets were imaged. Numbers of lipid droplets per cell were quantified using customized software that was developed in-house as previously described (36). Additionally, SDS-PAGE and immunoblotting using the differentiation marker protein galectin-3 (refs. 43, 44; ab2785, Abcam, Cambridge, MA) were done.

Statistical analysis of experiments. An unpaired two-tailed t test (α = 0.05) was used to detect significant differences between experimental groups using the JMP software package (Brooks/Cole-Thomson Learning, Belmont, CA). Ps < 0.05 were considered significant.

The custom-made choline kinase antibody consistently detected an immunoreactive band at 48 kDa, the anticipated molecular weight of choline kinase (29), which increased with increasing malignancy of the HMECs (Fig. 1). Nonmalignant MCF-12A cells had almost undetectable levels of choline kinase, which increased for the poorly metastatic malignant MCF-7 breast cancer cell line and were highest for the metastatic malignant MDA-MB-231 and MDA-MB-435 breast cancer cell lines (Fig. 1). Densitometry of the Western blots showed that choline kinase levels were significantly higher in all three breast cancer cell lines (P < 0.01, n = 3) compared with the nonmalignant MCF-12A HMEC line (Fig. 1B).

Figure 1.

A, choline kinase (Chk) protein expression levels, detected with the polyclonal choline kinase antibody, in a panel of nonmalignant (MCF-12A) and malignant HMEC lines with increasing malignancy (MCF-7 < MDA-MB-231 < MDA-MB-435 cells). Actin immunoblotting was done as a loading control. B, densitometry of Western blots showed semiquantitative changes relative to nonmalignant MCF-12A cells (n = 3). Relative fold changes in choline kinase were normalized to the choline kinase immunoreactive band in nonmalignant MCF-12A cells, which was set to 1. Columns, mean; bars, SD. **, P < 0.01, compared with MCF-12A.

Figure 1.

A, choline kinase (Chk) protein expression levels, detected with the polyclonal choline kinase antibody, in a panel of nonmalignant (MCF-12A) and malignant HMEC lines with increasing malignancy (MCF-7 < MDA-MB-231 < MDA-MB-435 cells). Actin immunoblotting was done as a loading control. B, densitometry of Western blots showed semiquantitative changes relative to nonmalignant MCF-12A cells (n = 3). Relative fold changes in choline kinase were normalized to the choline kinase immunoreactive band in nonmalignant MCF-12A cells, which was set to 1. Columns, mean; bars, SD. **, P < 0.01, compared with MCF-12A.

Close modal

To evaluate the role of choline kinase in breast cancer, cellular choline kinase mRNA levels were knocked down in nonmalignant MCF-12A cells and in malignant MDA-MB-231 breast cancer cells by transiently transfecting these cells with siRNA capable of down-regulating choline kinase mRNA (siRNA-chk). RT-PCR analysis of choline kinase mRNA was done following siRNA-chk treatment at 24, 48, and 72 hours after transfection. The ribosomal protein 36B4 mRNA was amplified as an internal loading control. A significant reduction in choline kinase mRNA was detected within 48 hours of treatment with siRNA-chk, as shown in Fig. 2A. Also evident were larger basal choline kinase mRNA levels (Fig. 2A) and choline kinase protein levels (Fig. 2B) in malignant MDA-MB-231 cells compared with nonmalignant MCF-12A cells (Fig. 2A). The corresponding choline kinase protein levels were determined by immunoblotting using choline kinase antibody as shown in Fig. 2B. Nonmalignant MCF-12A cells and malignant MDA-MB-231 cells were treated for 48 hours with oligofectamine alone (control oligo), or siRNA-chk premixed with oligofectamine (Fig. 2B). Choline kinase protein levels were significantly (P < 0.05, n = 4) lower in siRNA-chk-treated nonmalignant MCF-12A cells and malignant MDA-MB-231 cells compared with the controls, which were treated with oligofectamine alone (Fig. 2C). Fully relaxed 1H magnetic resonance spectra showed that phosphocholine levels, the product of choline kinase enzyme activity, as well as total choline-containing compound levels were significantly lower in control nonmalignant MCF-12A cells compared with control malignant MDA-MB-231 cells (Fig. 2D and E). No significant changes in cellular phosphocholine, total choline-containing compound, glycerophosphocholine, choline, and phosphatidylcholine levels were detected following siRNA-chk treatment for 48 hours in nonmalignant MCF-12A cells (Fig. 2D and E). By comparison, phosphocholine and total choline-containing compounds were significantly (P < 0.01 and P < 0.05, respectively, n = 6) lower in the malignant MDA-MB-231 breast cancer cells, following choline kinase knockdown (Fig. 2D and E). Glycerophosphocholine levels increased slightly following siRNA-chk treatment in malignant MDA-MB-231 cells (Fig. 2D and E), whereas choline (Fig. 2D and E), and lipid phosphatidylcholine levels (data not shown) remained constant. In summary, choline kinase mRNA, choline kinase protein, phosphocholine, and total choline-containing compound levels significantly decreased following transient choline kinase knockdown in malignant MDA-MB-231 breast cancer cells, whereas nonmalignant MCF-12A cells were hardly affected because of their low basal choline kinase and phosphocholine levels.

Figure 2.

A, RT-PCR of the time course of choline kinase (Chk) knockdown in nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells. 36B4 mRNA was amplified as internal loading control. B, Western blots of oligofectamine-treated controls and siRNA-chk-treated nonmalignant MCF-12A cells and malignant MDA-MB-231 cells at the 48-hour time point. Actin levels were determined as loading controls. C, densitometry of Western blots showed semiquantitative changes relative to the respective control cells (n = 4). Relative fold changes in choline kinase were normalized to the choline kinase immunoreactive band in the respective control oligo-treated cells, which was set to 1. Columns, mean; bars, SD. *, P < 0.05, compared with respective control oligo. D, expanded regions of 1H magnetic resonance spectra and (E) quantitation of 1H magnetic resonance spectra (n = 6) from cell extracts of nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells following 48 hours of transient siRNA-chk treatment mediating choline kinase knockdown. Abbreviations: Cho, free choline; GPC, glycerophosphocholine; PC, phosphocholine; tCho, total choline-containing compounds; ns, not significant. Columns, mean; bars, ±SE. *, P < 0.05; **, P < 0.01.

Figure 2.

A, RT-PCR of the time course of choline kinase (Chk) knockdown in nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells. 36B4 mRNA was amplified as internal loading control. B, Western blots of oligofectamine-treated controls and siRNA-chk-treated nonmalignant MCF-12A cells and malignant MDA-MB-231 cells at the 48-hour time point. Actin levels were determined as loading controls. C, densitometry of Western blots showed semiquantitative changes relative to the respective control cells (n = 4). Relative fold changes in choline kinase were normalized to the choline kinase immunoreactive band in the respective control oligo-treated cells, which was set to 1. Columns, mean; bars, SD. *, P < 0.05, compared with respective control oligo. D, expanded regions of 1H magnetic resonance spectra and (E) quantitation of 1H magnetic resonance spectra (n = 6) from cell extracts of nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells following 48 hours of transient siRNA-chk treatment mediating choline kinase knockdown. Abbreviations: Cho, free choline; GPC, glycerophosphocholine; PC, phosphocholine; tCho, total choline-containing compounds; ns, not significant. Columns, mean; bars, ±SE. *, P < 0.05; **, P < 0.01.

Close modal

Stable transfection of malignant MDA-MB-231 breast cancer cells was done with a construct expressing a shRNA, analogous to the siRNA-chk used in the transient transfection experiments, under the control of a U6 promoter (U6-shRNA-chk vector). Stable expression of the U6-shRNA-chk vector in malignant MDA-MB-231 breast cancer cells significantly reduced choline kinase mRNA levels in these cells compared with empty vector control, as displayed in Fig. 3A. Quantitative RT-PCR revealed that choline kinase mRNA decreased in excess of 10 cycles in U6-shRNA-chk vector expressing cells compared with empty vector control cells. Immunoblotting with choline kinase antibody showed that the corresponding choline kinase protein levels were significantly (P < 0.01, n = 6; Fig. 3B and C) reduced in U6-shRNA-chk vector expressing MDA-MB-231 cells compared with vector control (Fig. 3C). Stably knocking down choline kinase expression significantly reduced cellular phosphocholine (P < 0.01, n = 5) and total choline-containing compound (P < 0.05, n = 5) levels in 1H magnetic resonance spectra of MDA-MB-231 U6-shRNA-chk vector cells compared with empty vector cells, as shown in Fig. 3D and E. Lipid phosphatidylcholine levels were slightly lower in malignant MDA-MB-231 breast cancer cells with stably knocked down choline kinase compared with vector controls (Fig. 3E). Glycerophosphocholine and choline levels remained constant (Fig. 3E). Thus, similar to the transient transfection, the stable expression of U6-shRNA-chk in malignant MDA-MB-231 breast cancer cells resulted in a significant reduction of choline kinase mRNA, choline kinase protein, phosphocholine, and total choline-containing compound levels. Low expression of choline kinase mRNA and protein, and 1H MRS-detectable phosphocholine and total choline-containing compound levels remained stable in the clones, which were maintained in cell culture for up to 21 passages, over a period of 155 days. Changes in phospholipid metabolite levels were comparable with transient and stable choline kinase down-regulation in MDA-MB-231 cells as evident in Fig. 2E and Fig. 3E.

Figure 3.

A, conventional RT-PCR of choline kinase (Chk) mRNA levels in U6-shRNA-chk expressing MDA-MB-231 cells compared with MDA-MB-231 cells stably expressing the control vector. 36B4 mRNA levels served as an internal control. B, choline kinase Western blot of wild-type, control vector, and U6-shRNA-chk vector expressing MDA-MB-231 cells. Actin was probed as a loading control. C, densitometry of Western blots showed semiquantitative changes relative to the respective control cells (n = 6). Relative fold changes in choline kinase were normalized to the choline kinase immunoreactive band in MDA-MB-231 wild type cells, which was set to 1. Columns, mean; bars, SD. **, P < 0.01, compared with wild type; ##, P < 0.01, compared with control vector. D, expanded regions of 1H magnetic resonance spectra and (E) quantitation of these 1H magnetic resonance spectra (n = 5) from cell extracts of U6-shRNA-chk expressing MDA-MB-231 compared with control vector expressing MDA-MB-231 cells. Abbreviations: Cho, free choline; GPC, glycerophosphocholine; PC, phosphocholine; tCho, total choline-containing compounds; ns, not significant. Columns, mean; bars, ±SE. *, P < 0.05; **, P < 0.01.

Figure 3.

A, conventional RT-PCR of choline kinase (Chk) mRNA levels in U6-shRNA-chk expressing MDA-MB-231 cells compared with MDA-MB-231 cells stably expressing the control vector. 36B4 mRNA levels served as an internal control. B, choline kinase Western blot of wild-type, control vector, and U6-shRNA-chk vector expressing MDA-MB-231 cells. Actin was probed as a loading control. C, densitometry of Western blots showed semiquantitative changes relative to the respective control cells (n = 6). Relative fold changes in choline kinase were normalized to the choline kinase immunoreactive band in MDA-MB-231 wild type cells, which was set to 1. Columns, mean; bars, SD. **, P < 0.01, compared with wild type; ##, P < 0.01, compared with control vector. D, expanded regions of 1H magnetic resonance spectra and (E) quantitation of these 1H magnetic resonance spectra (n = 5) from cell extracts of U6-shRNA-chk expressing MDA-MB-231 compared with control vector expressing MDA-MB-231 cells. Abbreviations: Cho, free choline; GPC, glycerophosphocholine; PC, phosphocholine; tCho, total choline-containing compounds; ns, not significant. Columns, mean; bars, ±SE. *, P < 0.05; **, P < 0.01.

Close modal

A significant increase in the mixed fatty acid signal (Fmix) at 1.3 ppm was observed in transiently siRNA-chk-transfected nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells compared with control cells, whereas the phosphatidylcholine signal at 3.22 ppm remained constant in 1H magnetic resonance spectra of the cellular lipid fractions (Fig. 4A). Thus, the Fmix/phosphatidylcholine signal integral ratio was significantly higher in siRNA-chk-treated cells versus control cells for both nonmalignant MCF-12A cells (P < 0.05, n = 3, Fig. 4B) and malignant MDA-MB-231 cells (P < 0.05, n = 3, Fig. 4B), indicating an increase in cellular triacylglycerol. These data were verified by two-dimensional scalar-correlated spectroscopy (ref. 45; data not shown). Similarly, a significant (P < 0.001, n = 5) increase in triacylglycerol was detected in MDA-MB-231 breast cancer cells stably expressing U6-shRNA-chk vector compared with empty vector control cells, as quantified from the Fmix/phosphatidylcholine ratio (Fig. 4C). The increase of the Fmix/phosphatidylcholine ratio relative to controls was comparable in MDA-MB-231 breast cancer cells with transient or stable choline kinase down-regulation (Fig 4B and C).

Figure 4.

A, representative 1H magnetic resonance spectra and (B) quantitation of the Fmix/phosphatidylcholine ratio from these 1H magnetic resonance spectra (n = 3) obtained from lipid cell extract fractions of nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells treated with siRNA-chk. C, Fmix/phosphatidylcholine ratio from 1H magnetic resonance spectra (n = 3) of MDA-MB-231 cells with stably knocked down choline kinase (Chk) compared with empty-vector control. Abbreviation: PtdCho, phosphatidylcholine. Columns, mean; bars, ±SE. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 4.

A, representative 1H magnetic resonance spectra and (B) quantitation of the Fmix/phosphatidylcholine ratio from these 1H magnetic resonance spectra (n = 3) obtained from lipid cell extract fractions of nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells treated with siRNA-chk. C, Fmix/phosphatidylcholine ratio from 1H magnetic resonance spectra (n = 3) of MDA-MB-231 cells with stably knocked down choline kinase (Chk) compared with empty-vector control. Abbreviation: PtdCho, phosphatidylcholine. Columns, mean; bars, ±SE. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

The increase in triacylglycerol correlated with an increase in cellular lipid droplets in transiently siRNA-chk-transfected nonmalignant MCF-12A cells and malignant MDA-MB-231 cells, as evident in the confocal fluorescence images shown in Fig. 5A. An accumulation of cellular lipid droplets containing mobile triacylglycerol indicates differentiation in human breast epithelial cells (41, 42). Results from the analyses of lipid droplets in 20 to 22 cells from four independent siRNA-chk experiments each are summarized in Fig. 5B. Treatment with siRNA-chk resulted in a significant increase in the number of lipid droplets in nonmalignant MCF-12A (P < 0.001, n = 20) and malignant MDA-MB-231 cells (P < 0.001, n = 22). In addition, malignant MDA-MB-231 breast cancer cells were found to contain significantly (P < 0.01, n = 20-22) fewer lipid droplets compared with nonmalignant MCF-12A HMECs (Fig. 5B). In addition, as shown in Fig. 5C, in the stable clones, U6-shRNA-chk vector expressing MDA-MB-231 cells displayed significantly more lipid droplets compared with wild-type MDA-MB-231 cells (P < 0.001, n = 21) or empty vector expressing MDA-MB-231 cells (P < 0.001, n = 20). This increase in lipid droplets was more dramatic in transient choline kinase down-regulation (Fig. 5B and C). Baseline levels of Fmix/phosphatidylcholine did not differ significantly between nonmalignant MCF-12A cells and malignant MDA-MB-231 cells, although the number of lipid droplets was slightly higher in nonmalignant MCF-12A cells. This small difference was most likely not detected by the Fmix/phosphatidylcholine ratio due to poor sensitivity of this variable to subtle changes. The more pronounced changes in lipid droplets following transient or stable choline kinase down-regulation were detected by the Fmix/phosphatidylcholine ratio (Fig. 4B-C and Fig. 5B-C).

Figure 5.

A, number of lipid droplets visualized by Nile red staining (red) following siRNA-chk treatment in nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells. Nuclei (green). Bar, 10 μm. B, quantitation of the number of lipid droplets from 20 to 22 cells obtained from four independent experiments following siRNA-chk treatment in nonmalignant MCF-12A cells and malignant MDA-MB-231 cells. C, quantitation of the number of lipid droplets obtained from MDA-MB-231 cells with stably knocked down choline kinase compared with wild-type and empty-vector control cells. ##, P < 0.01, compared with MCF-12A; ###, P < 0.001, compared with wild-type MDA-MB-231; ***, P < 0.001, compared with control vector.

Figure 5.

A, number of lipid droplets visualized by Nile red staining (red) following siRNA-chk treatment in nonmalignant MCF-12A cells and malignant MDA-MB-231 breast cancer cells. Nuclei (green). Bar, 10 μm. B, quantitation of the number of lipid droplets from 20 to 22 cells obtained from four independent experiments following siRNA-chk treatment in nonmalignant MCF-12A cells and malignant MDA-MB-231 cells. C, quantitation of the number of lipid droplets obtained from MDA-MB-231 cells with stably knocked down choline kinase compared with wild-type and empty-vector control cells. ##, P < 0.01, compared with MCF-12A; ###, P < 0.001, compared with wild-type MDA-MB-231; ***, P < 0.001, compared with control vector.

Close modal

Proliferation was significantly (P < 0.001, n = 5) slower in MDA-MB-231 breast cancer cells stably expressing U6-shRNA-chk compared with empty vector controls (Fig. 6A). The cell doubling time of U6-shRNA-chk expressing MDA-MB-231 cells was significantly longer (34.7 hours) than the doubling time for MDA-MB-231 cells with the control vector (28.8 hours). Expression levels of both proliferation markers, proliferating cell nuclear antigen (PCNA; refs. 46, 47; Fig. 6B) and Ki-67 (ref. 48; Fig. 6E), were significantly (P < 0.01 and P < 0.05, respectively, n = 3; Fig. 6D and G) higher in the two metastatic malignant breast cancer cell lines MDA-MB-231 and MDA-MB-435 compared with the poorly metastatic malignant MCF-7 human breast cancer cell line and the nonmalignant MCF-12A HMEC line (Fig. 6). Both PCNA (Fig. 6C) and Ki-67 (Fig. 6F) levels were significantly (P < 0.01, n = 3; Fig. 6D and G) lower in MDA-MB-231 cells stably expressing U6-shRNA-chk compared with empty vector MDA-MB-231 cells. PCNA and Ki-67 protein levels in malignant MDA-MB-231 cells expressing U6-shRNA-chk were comparable with the expression levels in nonmalignant MCF-12A cells (Fig. 6B-G).

Figure 6.

A, proliferation rate of MDA-MB-231 cells containing stably knocked down choline kinase (Chk) compared with vector control. Points, mean; bars, ±SE. B, for comparison, PCNA immunoblotting was done in a panel of human breast epithelial cell lines exhibiting different degrees of malignancy. C, PCNA immunoblotting in MDA-MB-231 cells stably expressing U6-shRNA-chk compared with control vector cells. D, densitometry of Western blots showed semiquantitative fold changes relative to nonmalignant MCF-12A cells (n = 3). Relative fold changes in PCNA were normalized to the PCNA immunoreactive band in nonmalignant MCF-12A cells, which was set to 1. Columns, mean; bars, SD. E, Ki-67 immunoblotting of the nonmalignant MCF-12A HMEC line, the poorly invasive malignant MCF-7 breast cancer cell line, compared with the highly invasive malignant breast cancer cell lines MDA-MB-231 and MDA-MB-435. F, Ki-67 protein expression levels in MDA-MB-231 cells stably expressing U6-shRNA-chk compared with control vector. Actin was probed as a loading control in all blots. G, densitometry of Western blots showed semiquantitative changes relative to nonmalignant MCF-12A cells (n = 3). Relative fold changes in Ki-67 were normalized to the Ki-67 immunoreactive band in MCF-12A cells, which was set to 1. Columns, mean; bars, SD. *, P < 0.05; **, P < 0.01. Abbreviation: ns, not significant compared with MCF-12A.

Figure 6.

A, proliferation rate of MDA-MB-231 cells containing stably knocked down choline kinase (Chk) compared with vector control. Points, mean; bars, ±SE. B, for comparison, PCNA immunoblotting was done in a panel of human breast epithelial cell lines exhibiting different degrees of malignancy. C, PCNA immunoblotting in MDA-MB-231 cells stably expressing U6-shRNA-chk compared with control vector cells. D, densitometry of Western blots showed semiquantitative fold changes relative to nonmalignant MCF-12A cells (n = 3). Relative fold changes in PCNA were normalized to the PCNA immunoreactive band in nonmalignant MCF-12A cells, which was set to 1. Columns, mean; bars, SD. E, Ki-67 immunoblotting of the nonmalignant MCF-12A HMEC line, the poorly invasive malignant MCF-7 breast cancer cell line, compared with the highly invasive malignant breast cancer cell lines MDA-MB-231 and MDA-MB-435. F, Ki-67 protein expression levels in MDA-MB-231 cells stably expressing U6-shRNA-chk compared with control vector. Actin was probed as a loading control in all blots. G, densitometry of Western blots showed semiquantitative changes relative to nonmalignant MCF-12A cells (n = 3). Relative fold changes in Ki-67 were normalized to the Ki-67 immunoreactive band in MCF-12A cells, which was set to 1. Columns, mean; bars, SD. *, P < 0.05; **, P < 0.01. Abbreviation: ns, not significant compared with MCF-12A.

Close modal

To test for differentiation of the different breast epithelial cell lines and the choline kinase knockdown clones, we probed for galectin-3 protein expression by SDS-PAGE followed by immunoblotting. Galectin-3 is involved in several biological processes (49), including cell adhesion, migration, cell growth, differentiation, apoptosis, tumor progression, and metastasis (43, 44, 50), in human breast epithelial cells. Figure 7A shows that nonmalignant MCF-12A cells expressed almost undetectable levels of galectin-3 compared with malignant MCF-7, MDA-MB-231, and MDA-MB-435 breast cancer cells, where progressively increasing galectin-3 levels matched the decreasing differentiation of the cell lines. Galectin-3 protein expression levels were significantly (P < 0.05, n = 3) higher in all three malignant breast cancer cell lines compared with the nonmalignant MCF-12A HMEC line (Fig. 7A and C). Consistent with the transient transfection studies, galectin-3 expression was significantly lower (P < 0.05, n = 3; Fig. 7B-C), approaching levels in nonmalignant MCF-12A cells, in MDA-MB-231 cells stably expressing U6-shRNA-chk compared with empty vector control cells (Fig. 7B-C).

Figure 7.

A, galectin-3 protein expression levels detected in the nonmalignant MCF-12A HMEC line, the poorly metastatic malignant MCF-7 breast cancer cell line, and the invasive/metastatic malignant breast cancer cell lines MDA-MB-231 and MDA-MB-435. B, galectin-3 protein expression levels in MDA-MB-231 cells containing stably knocked down choline kinase compared with vector control. Actin was probed as a loading control in all blots. C, densitometry of Western blots showed semiquantitative changes relative to nonmalignant MCF-12A cells (n = 3). Relative fold changes in galectin-3 were normalized to the Gal-3 immunoreactive band in nonmalignant MCF-12A cells, which was set to 1. Columns, mean; bars, SD. *, P < 0.05. Abbreviation: ns, not significant compared with MCF-12A.

Figure 7.

A, galectin-3 protein expression levels detected in the nonmalignant MCF-12A HMEC line, the poorly metastatic malignant MCF-7 breast cancer cell line, and the invasive/metastatic malignant breast cancer cell lines MDA-MB-231 and MDA-MB-435. B, galectin-3 protein expression levels in MDA-MB-231 cells containing stably knocked down choline kinase compared with vector control. Actin was probed as a loading control in all blots. C, densitometry of Western blots showed semiquantitative changes relative to nonmalignant MCF-12A cells (n = 3). Relative fold changes in galectin-3 were normalized to the Gal-3 immunoreactive band in nonmalignant MCF-12A cells, which was set to 1. Columns, mean; bars, SD. *, P < 0.05. Abbreviation: ns, not significant compared with MCF-12A.

Close modal

Here we have shown, for the first time, that the use of siRNA specific for choline kinase provides a valuable strategy to specifically and efficiently target choline kinase in breast cancer cells. At least three isoforms of choline kinase are known to exist in mammalian cells, which are encoded by two separate genes: ck-α and ck-β (29). The two functional isoforms CK-α1 and CK-α2 are derived from the same gene ck-α by alternative splicing (29) with homodimeric or heterodimeric forms of choline kinase conferring enzymatic activity (29). Choline kinase can be down-regulated efficiently by RNAi, because the mRNA sequences transcribed from the chk-α and chk-β genes allow for the design of a single siRNA capable of down-regulating both of these choline kinase mRNAs, as was done in our studies. The importance of choline kinase in malignant transformation and progression is apparent from RNAi knockdown of choline kinase increasing differentiation and reducing proliferation of breast cancer cells. These results are also consistent with data from pharmacologic inhibition of choline kinase in breast cancer cells, which resulted in decreased proliferation and tumor xenograft growth (3).

Characterization of a panel of HMEC lines representing nonmalignant MCF-12A cells and malignant MCF-7, MDA-MB-231, and MDA-MB-435 breast cancer cells revealed that changes in choline kinase levels matched the changes in proliferation and differentiation markers in these cells. First, consistent with previous studies (2), we detected increased choline kinase protein expression levels in malignant MCF-7, MDA-MB-231, and MDA-MB-435 breast cancer cell lines compared with the nonmalignant MCF-12A HMEC line. Choline kinase expression levels increased with progressively increasing malignancy in these human breast epithelial cell lines. Second, the expression levels of proliferation markers, such as PCNA (46, 47) and Ki-67 (48), also matched choline kinase expression levels, suggesting a causal relationship between choline kinase levels and proliferation. These data are in excellent agreement with previous studies showing that activation of choline kinase is a critical requirement for proliferation of HMECs (3). Choline kinase was shown to play an essential role in the proliferation of primary HMECs by stimulating DNA synthesis and promoting G1 → S transition of the cell cycle (3). In malignant MCF-7 breast cancer cells, choline kinase activation was required during growth factor stimulation with heregulin to promote DNA synthesis (3). We also showed, for the first time, that choline kinase expression levels matched galectin-3 expression levels in the nonmalignant MCF-12A cell line and malignant MCF-7, MDA-MB-231, and MDA-MB-435 breast cancer cell lines representing different degrees of malignancy. Galectin-3 is involved, among other pathways, in cell adhesion, migration, differentiation, tumor progression, and metastasis (43, 44, 49, 50) in human breast epithelial cells. Collectively, these data support a strong correlation between malignancy and choline kinase expression in breast cancer cells.

Both transient and stable down-regulation of choline kinase expression by RNAi resulted in significantly reduced choline kinase mRNA and protein levels in malignant MDA-MB-231 breast cancer cells but not in the nonmalignant MCF-12A HMEC line. The functional outcome of this down-regulation was evident in the low phosphocholine and total choline-containing compound signals observed in magnetic resonance spectra of cells with transient and stable down-regulation of choline kinase. Phosphocholine itself has been reported to be mitogenic and may be a second messenger or mediator for the mitogenic activity of several growth factors (28, 51). The negligible effect of siRNA-chk-mediated choline kinase down-regulation on choline kinase, phosphocholine, and total choline-containing compound levels in nonmalignant MCF-12A HMECs was most likely due to the low basal choline kinase expression and consequently the low phosphocholine levels in these cells. Changes in phosphocholine were detected in high-resolution extract 1H magnetic resonance spectra, which may not be possible to detect in vivo; total choline-containing compounds on the other hand are easily detected in vivo (52). The reduction in total choline-containing compounds in our in vitro study suggests that 1H MRS imaging techniques (53) can potentially be used to detect the effect of choline kinase inhibition in vivo at the preclinical and clinical level.

Two major phenotypic changes, an increase in differentiation and a reduction of proliferation, were detected following RNAi-mediated down-regulation of choline kinase. Both transient and stable knockdown of choline kinase resulted in an accumulation of triacylglycerol-containing intracellular lipid droplets in malignant MDA-MB-231 cells and nonmalignant MCF-12A cells. This increase in the number of intracellular lipid droplets was more pronounced following transient down-regulation of choline kinase levels in MDA-MB-231 cells using siRNA-chk treatment compared with sustained down-regulation in MDA-MB-231 cells stably expressing the U6-shRNA-chk vector and may be due to cells adapting to sustained choline kinase down-regulation. Cytosolic triacylglycerol-containing lipid droplets are used by the mammary gland for the production of milk (54) and are therefore a marker of differentiation in human breast epithelial cells (41, 42). They are the precursors for milk fat globules, which are a major component of milk secreted by HMECs during lactation (54, 55). Down-regulation of choline kinase increases the production of these triacyglycerols, which are stored in intracellular lipid droplets in breast cancer cells and HMECs. Upon choline kinase down-regulation, 1,2-diacylglycerol may be used in the Kennedy pathway to a much lesser extent because of low choline kinase and phosphocholine levels. This excess 1,2-diacylglycerol in choline kinase knockdown clones may be used for the synthesis of triacylglycerol by acyl-CoA/diacylglycerol acyltransferase (EC 2.3.1.20; refs. 56, 57) and stored in the form of lipid droplets to potentially detoxify cells from excess 1,2-diacylglycerol. Tetraphosphonium-based cationic lipophilic chemotherapeutic agents were also shown to increase lipid droplet formation in breast cancer cells (5860), and a strong correlation was observed between magnetic resonance–detectable triacylglycerol accumulation and cytoplasmic lipid droplets (5860). Another possible explanation of our data may be that triacylglycerol-containing lipid droplets increased following RNAi-mediated choline kinase knockdown to provide cells with a potential reservoir of energy (60). The changes in lipid droplet formation are also consistent with the significant increase in the mixed fatty acid signal (Fmix) at 1.3 ppm following RNAi of choline kinase. Because it is possible to detect signal from mixed fatty acids in proton spectra obtained noninvasively in vivo, monitoring an increase of this mixed fatty acid signal in vivo may be yet another potential approach to detect choline kinase down-regulation. The differentiation of malignant breast cancer cells following RNAi-mediated down-regulation of choline kinase was additionally supported by the reduction of galectin-3 levels in malignant MDA-MB-231 cells following RNAi.

Breast cancer cell proliferation was significantly slower upon down-regulation of choline kinase in malignant MDA-MB-231 cells, which was further supported by the reduction of proliferation markers following RNAi. These results also compare well with recent findings that choline kinase is required for breast tumor progression, and that pharmacologic inhibition of choline kinase reduced tumor growth in an animal xenograft model of breast cancer (3).

In conclusion, specific down-regulation of choline kinase in malignant breast cancer cells induced profound alterations in proliferation and differentiation. The increase in differentiation and decrease of proliferation strongly support the development of therapeutic RNAi strategies targeting choline kinase (31) in breast cancer. Finally and equally importantly, such a therapeutic strategy would lend itself easily to imaging and characterization by magnetic resonance techniques in cells, in preclinical models in vivo, and, potentially, in the clinical setting.

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 Yelena Mironchik, Tomoyo Takagi, and Gary Cromwell for laboratory support.

1
Ramirez de Molina A, Gutierrez R, Ramos MA, et al. Increased choline kinase activity in human breast carcinomas: clinical evidence for a potential novel antitumor strategy.
Oncogene
2002
;
21
:
4317
–22.
2
Ramirez de Molina A, Rodriguez-Gonzalez A, Gutierrez R, et al. Overexpression of choline kinase is a frequent feature in human tumor-derived cell lines and in lung, prostate, and colorectal human cancers.
Biochem Biophys Res Commun
2002
;
296
:
580
–3.
3
de Molina AR, Banez-Coronel M, Gutierrez R, et al. Choline kinase activation is a critical requirement for the proliferation of primary human mammary epithelial cells and breast tumor progression.
Cancer Res
2004
;
64
:
6732
–9.
4
Nakagami K, Uchida T, Ohwada S, et al. Increased choline kinase activity and elevated phosphocholine levels in human colon cancer.
Jpn J Cancer Res
1999
;
90
:
419
–24.
5
Nakagami K, Uchida T, Ohwada S, Koibuchi Y, Morishita Y. Increased choline kinase activity in 1,2-dimethylhydrazine-induced rat colon cancer.
Jpn J Cancer Res
1999
;
90
:
1212
–7.
6
Ratnam S, Kent C. Early increase in choline kinase activity upon induction of the H-ras oncogene in mouse fibroblast cell lines.
Arch Biochem Biophys
1995
;
323
:
313
–22.
7
Teegarden D, Taparowsky EJ, Kent C. Altered phosphatidylcholine metabolism in C3H10T1/2 cells transfected with the Harvey-ras oncogene.
J Biol Chem
1990
;
265
:
6042
–7.
8
Ramirez de Molina A, Penalva V, Lucas L, Lacal JC. Regulation of choline kinase activity by Ras proteins involves Ral-GDS and PI3K.
Oncogene
2002
;
21
:
937
–46.
9
Ronen SM, Jackson LE, Beloueche M, Leach MO. Magnetic resonance detects changes in phosphocholine associated with Ras activation and inhibition in NIH 3T3 cells.
Br J Cancer
2001
;
84
:
691
–6.
10
Hernandez-Alcoceba R, Fernandez F, Lacal JC. In vivo antitumor activity of choline kinase inhibitors: a novel target for anticancer drug discovery.
Cancer Res
1999
;
59
:
3112
–8.
11
Podo F. Tumour phospholipid metabolism.
NMR Biomed
1999
;
12
:
413
–39.
12
Negendank W. Studies of human tumors by MRS: a review.
NMR Biomed
1992
;
5
:
303
–24.
13
Ackerstaff E, Glunde K, Bhujwalla ZM. Choline phospholipid metabolism: a target in cancer cells?
J Cell Biochem
2003
;
90
:
525
–33.
14
Ronen SM, Leach MO. Imaging biochemistry: applications to breast cancer.
Breast Cancer Res
2001
;
3
:
36
–40.
15
Ruiz-Cabello J, Cohen JS. Phospholipid metabolites as indicators of cancer cell function.
NMR Biomed
1992
;
5
:
226
–33.
16
Ting YL, Sherr D, Degani H. Variations in energy and phospholipid metabolism in normal and cancer human mammary epithelial cells.
Anticancer Res
1996
;
16
:
1381
–8.
17
Aboagye EO, Bhujwalla ZM. Malignant transformation alters membrane choline phospholipid metabolism of human mammary epithelial cells.
Cancer Res
1999
;
59
:
80
–4.
18
Glunde K, Jie C, Bhujwalla ZM. Molecular causes of the aberrant choline phospholipid metabolism in breast cancer.
Cancer Res
2004
;
64
:
4270
–6.
19
Ramirez de Molina A, Rodriguez-Gonzalez A, Penalva V, Lucas L, Lacal JC. Inhibition of ChoK is an efficient antitumor strategy for Harvey-, Kirsten-, and N-ras-transformed cells.
Biochem Biophys Res Commun
2001
;
285
:
873
–9.
20
Hernandez-Alcoceba R, Saniger L, Campos J, et al. Choline kinase inhibitors as a novel approach for antiproliferative drug design.
Oncogene
1997
;
15
:
2289
–301.
21
Warden CH, Friedkin M. Regulation of choline kinase activity and phosphatidylcholine biosynthesis by mitogenic growth factors in 3T3 fibroblasts.
J Biol Chem
1985
;
260
:
6006
–11.
22
Vigo C, Vance DE. Effect of diethylstilboestrol on phosphatidylcholine biosynthesis and choline metabolism in the liver of roosters.
Biochem J
1981
;
200
:
321
–6.
23
Pelech SL, Vance DE. Regulation of phosphatidylcholine biosynthesis.
Biochim Biophys Acta
1984
;
779
:
217
–51.
24
Uchida T. Stimulation of phospholipid synthesis in HeLa cells by epidermal growth factor and insulin: activation of choline kinase and glycerophosphate acyltransferase.
Biochim Biophys Acta
1996
;
1304
:
89
–104.
25
Jimenez B, del Peso L, Montaner S, Esteve P, Lacal JC. Generation of phosphorylcholine as an essential event in the activation of Raf-1 and MAP-kinases in growth factors-induced mitogenic stimulation.
J Cell Biochem
1995
;
57
:
141
–9.
26
Ko KW, Cook HW, Vance DE. Reduction of phosphatidylcholine turnover in a Nb 2 lymphoma cell line after prolactin treatment. A novel mechanism for control of phosphatidylcholine levels in cells.
J Biol Chem
1986
;
261
:
7846
–52.
27
Neeman M, Degani H. Early estrogen-induced metabolic changes and their inhibition by actinomycin D and cycloheximide in human breast cancer cells: 31P and 13C NMR studies.
Proc Natl Acad Sci U S A
1989
;
86
:
5585
–9.
28
Chung T, Huang JS, Mukherjee JJ, Crilly KS, Kiss Z. Expression of human choline kinase in NIH 3T3 fibroblasts increases the mitogenic potential of insulin and insulin-like growth factor I.
Cell Signal
2000
;
12
:
279
–88.
29
Aoyama C, Liao H, Ishidate K. Structure and function of choline kinase isoforms in mammalian cells.
Prog Lipid Res
2004
;
43
:
266
–81.
30
Campos JM, Nunez MC, Sanchez RM, et al. Quantitative structure-activity relationships for a series of symmetrical bisquaternary anticancer compounds.
Bioorg Med Chem
2002
;
10
:
2215
–31.
31
Hannon GJ, Rossi JJ. Unlocking the potential of the human genome with RNA interference.
Nature
2004
;
431
:
371
–8.
32
Paddison PJ, Caudy AA, Hannon GJ. Stable suppression of gene expression by RNAi in mammalian cells.
Proc Natl Acad Sci U S A
2002
;
99
:
1443
–8.
33
Paddison PJ, Caudy AA, Bernstein E, Hannon GJ, Conklin DS. Short hairpin RNAs (shRNAs) induce sequence-specific silencing in mammalian cells.
Genes Dev
2002
;
16
:
948
–58.
34
Paine TM, Soule HD, Pauley RJ, Dawson PJ. Characterization of epithelial phenotypes in mortal and immortal human breast cells.
Int J Cancer
1992
;
50
:
463
–73.
35
Soule HD, Vazguez J, Long A, Albert S, Brennan M. A human cell line from a pleural effusion derived from a breast carcinoma.
J Natl Cancer Inst
1973
;
51
:
1409
–16.
36
Glunde K, Guggino SE, Solaiyappan M, Pathak AP, Ichikawa Y, Bhujwalla ZM. Extracellular acidification alters lysosomal trafficking in human breast cancer cells.
Neoplasia
2003
;
5
:
533
–45.
37
Tyagi RK, Azrad A, Degani H, Salomon Y. Simultaneous extraction of cellular lipids and water-soluble metabolites: evaluation by NMR spectroscopy.
Magn Reson Med
1996
;
35
:
194
–200.
38
Greenspan P, Mayer EP, Fowler SD. Nile red: a selective fluorescent stain for intracellular lipid droplets.
J Cell Biol
1985
;
100
:
965
–73.
39
Greenspan P, Fowler SD. Spectrofluorometric studies of the lipid probe, Nile red.
J Lipid Res
1985
;
26
:
781
–9.
40
Fukumoto S, Fujimoto T. Deformation of lipid droplets in fixed samples.
Histochem Cell Biol
2002
;
118
:
423
–8.
41
Munster PN, Srethapakdi M, Moasser MM, Rosen N. Inhibition of heat shock protein 90 function by ansamycins causes the morphological and functional differentiation of breast cancer cells.
Cancer Res
2001
;
61
:
2945
–52.
42
Guilbaud NF, Gas N, Dupont MA, Valette A. Effects of differentiation-inducing agents on maturation of human MCF-7 breast cancer cells.
J Cell Physiol
1990
;
145
:
162
–72.
43
Honjo Y, Nangia-Makker P, Inohara H, Raz A. Down-regulation of galectin-3 suppresses tumorigenicity of human breast carcinoma cells.
Clin Cancer Res
2001
;
7
:
661
–8.
44
Takenaka Y, Fukumori T, Raz A. Galectin-3 and metastasis.
Glycoconj J
2004
;
19
:
543
–9.
45
May GL, Wright LC, Holmes KT, et al. Assignment of methylene proton resonances in NMR spectra of embryonic and transformed cells to plasma membrane triglyceride.
J Biol Chem
1986
;
261
:
3048
–53.
46
Heimann R, Hellman S. Individual characterisation of the metastatic capacity of human breast carcinoma.
Eur J Cancer
2000
;
36
:
1631
–9.
47
Maga G, Hubscher U. Proliferating cell nuclear antigen (PCNA): a dancer with many partners.
J Cell Sci
2003
;
116
:
3051
–60.
48
Scholzen T, Gerdes J. The Ki-67 protein: from the known and the unknown.
J Cell Physiol
2000
;
182
:
311
–22.
49
Perillo NL, Marcus ME, Baum LG. Galectins: versatile modulators of cell adhesion, cell proliferation, and cell death.
J Mol Med
1998
;
76
:
402
–12.
50
Iurisci I, Tinari N, Natoli C, Angelucci D, Cianchetti E, Iacobelli S. Concentrations of galectin-3 in the sera of normal controls and cancer patients.
Clin Cancer Res
2000
;
6
:
1389
–93.
51
Cuadrado A, Carnero A, Dolfi F, Jimenez B, Lacal JC. Phosphorylcholine: a novel second messenger essential for mitogenic activity of growth factors.
Oncogene
1993
;
8
:
2959
–68.
52
Katz-Brull R, Lavin PT, Lenkinski RE. Clinical utility of proton magnetic resonance spectroscopy in characterizing breast lesions.
J Natl Cancer Inst
2002
;
94
:
1197
–203.
53
Coakley FV, Qayyum A, Kurhanewicz J. Magnetic resonance imaging and spectroscopic imaging of prostate cancer.
J Urol
2003
;
170
:
S69
–75; discussion S75–6.
54
Mather IH, Keenan TW. Origin and secretion of milk lipids.
J Mammary Gland Biol Neoplasia
1998
;
3
:
259
–73.
55
Jensen RG, Ferris AM, Lammi-Keefe CJ. Lipids in human milk and infant formulas.
Annu Rev Nutr
1992
;
12
:
417
–41.
56
Bell RM, Coleman RA. Enzymes of glycerolipid synthesis in eukaryotes.
Annu Rev Biochem
1980
;
49
:
459
–87.
57
Lehner R, Kuksis A. Biosynthesis of triacylglycerols.
Prog Lipid Res
1996
;
35
:
169
–201.
58
Delikatny EJ, Roman SK, Hancock R, et al. Tetraphenylphosphonium chloride induced MR-visible lipid accumulation in a malignant human breast cell line.
Int J Cancer
1996
;
67
:
72
–9.
59
Sathasivam N, Brammah S, Wright LC, Delikatny EJ. Inhibition of tetraphenylphosphonium-induced NMR-visible lipid accumulation in human breast cells by chlorpromazine.
Biochim Biophys Acta
2003
;
1633
:
149
–60.
60
Delikatny EJ, Cooper WA, Brammah S, Sathasivam N, Rideout DC. Nuclear magnetic resonance-visible lipids induced by cationic lipophilic chemotherapeutic agents are accompanied by increased lipid droplet formation and damaged mitochondria.
Cancer Res
2002
;
62
:
1394
–400.