Apoptosis Triggered by 1-O-Octadecyl-2-O-methyl-rac-glycero-3-phosphocholine Is Prevented by Increased Expression of CTP:Phosphocholine Cytidylyltransferase*

A HeLa cell line was constructed for the regulation of CTP:phosphocholine cytidylyltransferase (CCT) expression via a tetracycline-responsive promoter to test the role of CCT in apoptosis triggered by exposure of cells to the antineoplastic phospholipid 1-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine (ET-18-OCH3). Basal CCT expression in the engineered HeLa cell line was the same as in control HeLa cells lines, and CCT activity and protein were elevated 25-fold following 48 h of induction with doxycycline. Increased CCT expression prevented ET-18-OCH3-induced apoptosis. Acylation of exogenous lysophosphatidylcholine circumvented the requirement for CCT activity by providing an alternate route to phosphatidylcholine, and heightened CCT expression and lysophosphatidylcholine supplementation were equally effective in reversing the cytotoxic effect of ET-18-OCH3. Neither CCT overexpression nor lysophosphatidylcholine supplementation allowed the HeLa cells to proliferate in the presence of ET-18-OCH3, indicating that the cytostatic property of ET-18-OCH3 was independent of its effect on membrane phospholipid synthesis. These data provide compelling genetic evidence to support the conclusion that the interruption of phosphatidylcholine synthesis at the CCT step by ET-18-OCH3 is the primary physiological imbalance that accounts for the cytotoxic action of the drug.

ET-18-OCH 3 1 is a nonmetabolizable analog of LPC and belongs to the first generation of ether lipids tested as growth inhibitors (1). These compounds do not directly target DNA, and numerous studies have demonstrated a selective cytotoxic action of ET-18-OCH 3 against transformed cells in whole animals and tissue culture (2)(3)(4)(5)(6)(7)(8). Recent work has established that the cytotoxic effect of ET-18-OCH 3 is due to the ability of the antineoplastic phospholipids to induce apoptosis in sensitive cells (9 -12). A plethora of biological processes have been suggested as primary targets for the antineoplastic etherlinked phospholipids (for reviews, see Refs. [13][14][15][16]. The long list of physiological imbalances includes the inhibition of phosphatidylinositol phospholipase C and calcium movements (17)(18)(19), protein kinase C-regulated functions (20 -24), lysophospholipid metabolism (25,26), and PtdCho synthesis (12,(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37)(38)(39). Although some of the results are contradictory, it is clear that there are multiple targets for ET-18-OCH 3 , and it is not yet possible to distinguish the physiological imbalances that are causative from those that are either derivative or unrelated to the main event. Thus, a major contemporary focus in the field is to identify the critical cellular target(s) that are responsible for the cytotoxic and cytostatic actions of ET-18-OCH 3 .
Our work has focused on the role of the inhibition of PtdCho synthesis in the mechanism of antineoplastic phospholipid action. PtdCho is essential for the survival of cultured cells because it is a major structural building block of biological membranes and the precursor to the other two most abundant membrane phospholipids, phosphatidylethanolamine (40) and sphingomyelin (41). Thus, the cessation of PtdCho synthesis has a global effect on membrane structure and function. CCT catalyzes the formation of CDP-choline and is a key enzyme controlling the PtdCho biosynthetic pathway (42,43). We proposed that the inhibition of PtdCho synthesis was the underlying cause for the cytotoxicity of ET-18-OCH 3 and hexadecylphosphocholine due to their ability to limit the formation of CDP-choline. These compounds effectively mimic LPC, a physiological regulator of CCT activity (12,38,39). Like LPC, both ET-18-OCH 3 and hexadecylphosphocholine reduce the CDPcholine formation in intact cells and inhibit purified CCT in an in vitro assay (12,38,39).
Supplementation of the medium with LPC provides a pathway to PtdCho that is independent of CCT and prevents apoptosis induced by either ET-18-OCH 3 (12) or hexadecylphosphocholine (39), consistent with the idea that CCT inhibition is causative in initiating apoptosis. This idea is corroborated by experiments with a mutant CHO cell line, mutant 58, that has a temperature-sensitive defect in CCT activity (44). The mutant 58 cell line undergoes apoptosis when shifted to the nonpermissive temperature (45), and LPC supplementation rescues the mutant cells from programmed cell death (45,46). LPC supplementation does not allow the continued proliferation of cells in the presence of ET-18-OCH 3 , indicating that the drug also has cytostatic properties that are not related to its interference with PtdCho biosynthesis (12,39). These experiments are completely concordant with the hypothesis that CCT is the critical cellular target for ET-18-OCH 3 -induced apoptosis. However, LPC could possibly rescue cells by competitively reversing the inhibitory effect of ET-18-OCH 3 on another important cellular target in addition to restoring PtdCho synthesis. Thus, additional experiments are required to verify a specific role for CCT in ET-18-OCH 3 action. In this report we provide genetic evidence for the CCT hypothesis by demon-strating that the specific elevation of cellular CCT content confers resistance to ET-18-OCH 3 cytotoxicity.

EXPERIMENTAL PROCEDURES
Materials-Sources of supplies were: CLONTECH, pTRE, and pTK-Hyg vectors, and HeLa Tet-On TM cells; Sigma, doxycycline, protein A-Sepharose, and chromatography standards; Life Technologies, cell culture media; Calbiochem, ET-18-OCH 3 ; Avanti Polar Lipids, lysophosphatidylcholine; American Radiolabeled Chemicals, phospho-[methyl-14 C]choline; Analtech, thin-layer chromatography plates; Amersham, ECL kit for detection of proteins on immunoblots. Molecular biology reagents were purchased from either Promega or New England Biolabs. The CCT cDNA was described previously (47). All other materials were reagent grade or better.
Isolation of a HeLa Cell Line with Regulated CCT Expression-HeLa Tet-On cells from CLONTECH laboratories express the regulatory Tetresponsive transcription factor. CCT cDNA was cloned into the pTRE vector downstream of the Tet-regulated promoter elements. Briefly, XbaI linkers were added to a fragment containing the CCT-coding sequence and ligated into the XbaI site of the pTRE vector. The HeLa Tet-On cells were co-transfected with the pTRE-CCT and pTK-Hyg vectors using a calcium-phosphate DNA precipitation technique (48). The next day, the cells were split 1/10 and seeded into dishes containing the hygromycin selective medium. Several weeks later, individual hygromycin-resistant colonies appeared. These clonal colonies were isolated, expanded, and screened for the induction of CCT activity following the addition of doxycycline. Clone CCT.12 was selected for our experiments from the 32 clones examined because the basal level of CCT activity was identical to control HeLa cells in the absence of inducer, and this level was significantly increased by the addition of doxycycline. A control cell line, CCT.00, was constructed using the same procedure except that the pTRE vector lacked the CCT insert.
CCT Assay-HeLa cells were washed twice with phosphate-buffered saline on ice and harvested by scraping into 1 ml of the same buffer followed by centrifugation. The cell pellets were resuspended in lysis buffer (10 mM HEPES, pH 7.4, 10 mM NaCl, 1 mM EDTA, 2 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 2% aprotinin, 1 g/ml leupeptin, 50 mM NaF, 100 M Na 3 VO 4 ) and sonicated three times for 30 s. The standard CCT activity assay contained an aliquot of the HeLa cell lysate mixed with 125 mM bis-Tris-HCl, pH 6.5, 0.5 Ci of phospho[methyl-14 C]choline, 1 mM phosphocholine, 2 mM CTP, 20 mM MgCl 2 , 50 M PtdCho/18:1 (1:1) in a final volume of 40 l. The incubations were for 10 min at 37°C and were stopped by placing the samples on ice and adding 5 l of 0.5 M EDTA. CDP-[ 14 C]choline formation was determined by thin layer chromatography (49). CCT-specific activity was calculated from a series of assays that were linear with time and protein. Protein was determined according to the method of Bradford (50) using ␥-globulin as a standard.
Immunoprecipitation and Immunoblotting-Cells were washed with cold phosphate-buffered saline, lysed on the plate with 400 l of radioimmune precipitation buffer (0.15 M NaCl, 50 mM Tris, pH 8.0, 1% Nonidet P-40, 0.5% deoxycholic acid, 0.1% SDS, 33 mM NaF, 3.3 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin) on ice for 30 min with occasional rocking. The cell lysates were centrifuged for 10 min at 230 ϫ g, and the protein concentration was determined by the Bradford method. Equivalent amounts of protein were immunoprecipitated with 15 g of affinity-purified anti-peptide antibody. The antibody was raised against the N-terminal CCT sequence used by Watkins and Kent (51) to generate immunoprecipitating CCT antisera. The samples were incubated overnight at 4°C in a rotator. Protein A-Sepharose beads were prewashed with radioimmune precipitation buffer and resuspended at 10% (v/v) in the radioimmune precipitation buffer described above. The suspended protein A-Sepharose beads (100 l) were added to each sample and the mixture was incubated in a rotator at 4°C for 1 h. The beads were collected by centrifugation at 4°C, washed three times with radioimmune precipitation buffer, and resuspended in 50 l of SDS sample buffer (2% SDS, 10% glycerol, 100 mM dithiothreitol, 60 mM Tris, pH 6.8, 0.001% bromphenol blue), boiled for 10 min, and centrifuged at room temperature. The supernatants were loaded on 10% SDS-polyacrylamide gel electrophoresis and electroblotted onto nitrocellulose membranes. The membranes were blocked with 1% dry milk for 1 h, washed with Tris-buffered saline (150 mM NaCl, 10 mM Tris-HCl pH 8.0, 0.1% Triton X-100), and exposed to anti-recombinant CCT antibody (52) diluted 1/200. The blots were washed with Trisbuffered saline and incubated with the secondary antibody conjugated with peroxidase for 1 h. An Amersham ECL detection kit was used to localize the secondary antibody.
Cytotoxicity Studies-Cells were seeded at densities 2.5 ϫ 10 5 cells/ 10-cm tissue culture plate and treated with 2 g/ml doxycycline and/or 50 M LPC 4 h prior to the addition of 4 M ET-18-OCH 3 . At 24-, 48-, and 72-h intervals, the cells were washed twice with phosphate-buffered saline, and the washes were collected. Adherent cells were removed by trypsinization for 1 min. Trypsinization was stopped by addition of 1 ml of complete medium. The trypsinized cells were combined with saline washes, centrifuged, and resuspended in 1-2 ml of complete medium. One drop of trypan blue was added to each sample, and the numbers of total and viable cells were counted.

Regulation of CCT Expression in HeLa
Cells-HeLa cells were selected for this study for two reasons. First, HeLa cells are sensitive to ET-18-OCH 3 and rapidly undergo apoptosis when exposed to the drug (9). Second, a HeLa cell line is available that stably expresses a transcriptional regulator consisting of the DNA binding domain of a mutant Tet repressor fused to the transcriptional activation domain of VP16 (53). This cell line affords tight control over the expression of introduced genes cloned downstream of the promoter elements for the tetracycline-responsive transcriptional repressor of Escherichia coli. Doxycycline binding to the mutant Tet repressor-VP16 fusion protein potently activates transcription of the target gene. This approach allows the same clone to be used to determine the sensitivity of the cells to ET-18-OCH 3 in the presence or absence of CCT overexpression, thus avoiding the inherent problems associated with clonal variation. The CCT.12 HeLa cell line was constructed by co-transfecting HeLa Tet-On cells with the pTRE-CCT and pTK-Hyg vectors. Individual hygromycin-resistant colonies were isolated, expanded, and screened for elevated levels of CCT activity in cell extracts following exposure of the clones to the inducer, doxycycline. Of the 32 hygromycin-resistant clones examined, three showed elevated CCT activity when grown in doxycycline and possessed the characteristics required for our experiments (see below). Clones were also isolated by the same procedure from HeLa Tet-On cells cotransfected with the empty pTRE vector plus the selectable pTK-Hyg vector. One clone (CCT.00) was randomly selected to serve as an experimental control.
Clone CCT.12 was selected from the three candidates because it possessed a basal CCT-specific activity that was the same as the parental HeLa Tet-On cells and the empty vector controls, and because CCT activity was strongly induced by the addition of doxycycline (Fig. 1). A doxycycline dose-response experiment and the time course for induction showed that the highest levels of CCT expression occurred in the presence of 2 g/ml doxycycline for 48 h (not shown). Under these induction conditions, cell-free extracts from clone CCT.12 exhibited a CCT-specific activity that was 25-fold higher than the activity in the uninduced cell extracts (Fig. 1B). The CCT-specific activity of the uninduced CCT.12 cell line was the same as the CCT-specific activity in the control clone, CCT.00 (Fig. 1B). The increased CCT-specific activity correlated with elevated levels of CCT protein in the doxycycline-induced clone CCT.12 (Fig.  1A). The detection of multiple bands in CCT immunoblots (Fig.  1B) was not unexpected since this type of banding pattern was observed previously and is due to the presence of phosphorylated isoforms of the protein (52). We did not observe any differences in HeLa cell morphology or growth rate of the CCT.12 cells grown in the presence or absence of doxycycline. The CCT-specific activity in extracts from the control clone (CCT.00) was the same as in extracts from the parental HeLa Tet-On cells (not shown). CCT-specific activity was not elevated by treatment of the control or parental cells with doxycycline (not shown).
CCT Overexpression Rescued Cells from ET-18-OCH 3 -induced Apoptosis-The role of CCT in the cytotoxic and cyto-static action of ET-18-OCH 3 was directly tested by comparing the sensitivity of the CCT.12 cell line to growth inhibition and cell death in the presence and absence of CCT overexpression induced by doxycycline. Previous work had established that HeLa cells undergo apoptosis when exposed to ET-18-OCH 3 (9), and our HeLa cell populations rapidly lost viability and exhibited condensed nuclei and other morphological features of apoptosis following exposure to ET-18-OCH 3 . We used a series of growth experiments with CCT.12 cells in the presence of doxycycline and ET-18-OCH 3 to test whether the specific overexpression of CCT would prevent programmed cell death. First, we determined whether CCT expression altered the response of HeLa cells to ET-18-OCH 3 -triggered apoptosis by exposing CCT.12 and control (CCT.00) cells to 4 M ET-18-OCH 3 for 48 h either in the presence or absence of 2 g/ml doxycycline (Fig. 2). ET-18-OCH 3 effectively killed the control cells either in the presence or absence of doxycycline (Fig. 2, panels C and E). In sharp contrast, CCT.12 cells were protected from ET-18-OCH 3dependent cell death by the induction of CCT expression (Fig.  2, panels D and F). The gross morphological appearance of the CCT.12 cells treated with doxycycline plus ET-18-OCH 3 was not distinguishable from the appearance of cells that were not treated with ET-18-OCH 3 (Fig. 2). The percentage of the total cells present that were viable in the doxycycline-induced ET-18-OCH 3 -treated sample (Fig. 2, panel F) was the same as the percent viability of the untreated controls. However, the number of viable cells recovered in the doxycycline plus ET-18-OCH 3 -treated group (Fig. 2, panel F) was consistently lower than in the control groups (Fig. 2, panels A and B). This result suggested that, while CCT overexpression prevented apoptosis, the CCT.12 cells did not proliferate in the presence of ET-18-OCH 3 . This conclusion was tested in an experiment where cell growth was measured following ET-18-OCH 3 treatment (Fig.  3). Exponentially growing CCT.12 cells were treated with doxycycline for 5 h prior to the addition of 4 M ET-18-OCH 3 at time 0, and the viable cell number was determined over a 3-day period. Control cells left untreated or treated with 2 g/ml doxycycline continued exponential growth, and cells exposed to 4 M ET-18-OCH 3 alone died rapidly. In contrast, CCT.12 cells treated with doxycycline plus 4 M ET-18-OCH 3 did not proliferate. Despite the inability of the cells to divide, the viability of the population remained as high as the control groups, and dead cells did not accumulate in the culture. This indicated that CCT overexpression rescued the entire cell population from ET-18-OCH 3 and that the relatively constant viable cell count displayed in Fig. 3 did not arise from continued cell growth accompanied by some degree of cell death in the population. These data show that CCT overexpression overrides the cytotoxic effect of ET-18-OCH 3 but cannot overcome the cytostatic action of the drug.
CCT Overexpression and LPC Supplementation Were Not Additive-Our previous experiments demonstrated that LPC supplementation prevented ET-18-OCH 3 cytotoxicity but did not permit cell proliferation (12,39). These data were interpreted to mean that providing an alternate source for PtdCho synthesis via the acylation of exogenous LPC circumvented the requirement for CCT and thus reversed the cytotoxic effects of CCT protein was immunoprecipitated from cell extracts with the Nterminal anti-peptide CCT antibody from line CCT.12 or the empty vector control (Vector). The immunoprecipitates were separated by SDS-polyacrylamide gel electrophoresis, and CCT protein was detected by immunoblotting using an antibody directed against recombinant CCT as described under "Experimental Procedures." The lane on the left-hand side (Sf9CCT) was a control that contained an extract from Sf9 cells infected with a recombinant baculovirus expressing CCT (49). Panel B, HeLa cell lines containing either the CCT expression vector (CCT.12) were treated with 2 g/ml doxycycline for 48 h or left untreated for the same length of time. Duplicate dishes of CCT.12 cells or the empty vector control (Vector) were harvested, and the CCT-specific activity was determined as a function of protein concentration as described under "Experimental Procedures." Doxycycline had no effect on the level of CCT expression or activity in the control cell line (CCT.00) containing the empty vector control (not shown). ET-18-OCH 3 . However, it was possible that LPC had an effect on another cellular target in addition to CCT that was responsible for these results. Comparing the effectiveness of LPC supplementation and CCT expression alone or in combination on ET-18-OCH 3 cytotoxicity (Fig. 4) tested this point. LPC supplementation rescued cells from ET-18-OCH 3 regardless of the presence of doxycycline, and the extent of cell rescue was the same as observed in doxycycline-treated cells. In both cases, the cells remained viable, but did not proliferate. Importantly, LPC supplementation plus doxycycline induction of CCT overexpression was not additive, suggesting that both treatments targeted the same process. These data support the conclusion that LPC supplementation prevents ET-18-OCH 3induced cytotoxicity by supplying an alternate route to PtdCho synthesis that circumvents the CCT step.

DISCUSSION
Our genetic experiments provide compelling support for the conclusion that the interruption of PtdCho synthesis at the CCT step is essential for the cytotoxic activity of ET-18-OCH 3 . The inhibition of PtdCho synthesis is a universal effect of ET-18-OCH 3 on cells (27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37), suggesting that the interference with PtdCho metabolism may be responsible for the biological effects of the drug. Our previous work identified CCT as the ET-18-OCH 3 target in the PtdCho biosynthetic pathway based on the pattern of accumulation of choline-derived pathway intermediates in vivo. ET-18-OCH 3 also was able to inhibit purified CCT activity in vitro by competing for the lipid activator site on the enzyme (12,38,54). The doxycycline-dependent CCT expression system developed in this report demonstrates that specifically increasing the cellular concentration of CCT imparts ET-18-OCH 3 resistance to HeLa cells. These experiments provide a critical genetic test that strongly supports a central role for CCT inhibition in ET-18-OCH 3 -induced apoptosis. Acylation of exogenous LPC is a direct route to PtdCho that circumvents the requirement for cellular CCT activity (45,46) and, accordingly, also prevents the cytotoxic effects of ET-18-OCH 3 (12). The possibility that LPC affects the interaction between ET-18-OCH 3 and a cellular target other than CCT appears untenable in light of the identical effects of LPC supplementation and CCT overexpression on reversing ET-18-OCH 3 cytotoxicity and the absence of an additive effect when the two treatments are used in combination. Taken together, the biochemical, physiological, and genetic data provide convincing evidence that the inhibition of PtdCho synthesis at the CCT step is the underlying cause for the programmed cell death caused by ET-18-OCH 3 .
The antiproliferative property of ET-18-OCH 3 is independent of its suppression of PtdCho biosynthesis. A large number of physiological imbalances are promoted by ET-18-OCH 3 (13)(14)(15)(16), and it is reasonable to assume that the inhibition of one or more of these potential targets is responsible for the cessation of cell growth. Our previous work shows that ET-18-OCH 3 arrests cells primarily in the G 1 and G 2 phases of the cell cycle and that LPC rescue results in G 1 arrest (12). These data suggest that the relevant targets for the cytostatic action of ET-18-OCH 3 will be involved in cell cycle progression. Current research has focused on the ability of ET-18-OCH 3 to inhibit components of signal transduction pathways involved in growth stimulation with the idea that blocking one or more of these events may account for the antiproliferative effects of the drug (17-19, 55, 56). However, we must also consider the possibility that ET-18-OCH 3 activates a biochemical cascade that counteracts the normal transmission of proliferative signals. In this regard, the reports that ET-18-OCH 3 stimulates the expression of immediate early genes such as c-fos and c-junB (57, 58) may be relevant to the activation of a signaling cascade that results in growth arrest. The problem of identifying a definitive cellular target responsible for the cytostatic action of ET-18-OCH 3 within this constellation of candidates will be a major challenge for future research.
The existence of cell lines that are relatively resistant to ET-18-OCH 3 was recognized early in the investigation of the action of this drug (6). Our result that elevated CCT expression bestows ET-18-OCH 3 resistance raises the question of whether or not the differences between sensitive and resistant cell lines can be attributed to variations in the cellular concentrations of CCT. A survey of CCT-specific activities in resistant and sensitive cells is not available, but alterations in the level of CCT expression may not explain the large range of drug sensitivity among cell types. Several recent studies indicate that antineoplastic ether lipids accumulate to a lesser extent in resistant cell lines (59 -63). ET-18-OCH 3 is a substrate for class I and class II P-glycoproteins and other ABC transporters that mediate pleiotropic drug resistance in a wide range of organisms (64). Resistance to the effects of ET-18-OCH 3 in yeast was dependent on overexpression of a functional efflux pump (64). The interaction of ET-18-OCH 3 with all members of this group of proteins suggests that they are not specific transporters for ET-18-OCH 3 or lysophospholipids. Lack of accumulation of effective dosages of ET-18-OCH 3 , together with elevated expression levels of CCT target protein, may each make independent contributions to cellular resistance mechanisms.