Purpose: Lysophosphatidic acid acyltransferase (LPAAT)-β catalyzes the conversion of lysophosphatidic acid to phosphatidic acid, an essential component of several signaling pathways, including the Ras/mitogen-activated protein kinase pathway. Inhibition of LPAAT-β induces growth arrest and apoptosis in cancer cell lines, implicating LPAAT-β as a potential drug target in neoplasia.

Experimental Design: In this study, we investigated the effects of CT32228, a specific LPAAT-β inhibitor, on BCR-ABL-transformed cell lines and primary cells from patients with chronic myelogenous leukemia.

Results: CT32228 had antiproliferative activity against BCR-ABL-positive cell lines in the nanomolar dose range, evidenced by cell cycle arrest in G2-M and induction of apoptosis. Treatment of K562 cells with CT32228 led to inhibition of extracellular signal-regulated kinase 1/2 phosphorylation, consistent with inhibition of mitogen-activated protein kinase signaling. Importantly, CT32228 was highly active in cell lines resistant to the Bcr-Abl kinase inhibitor imatinib. Combination of CT32228 with imatinib produced additive inhibition of proliferation in cell lines with residual sensitivity toward imatinib. In short-term cultures in the absence of growth factors, CT32228 preferentially inhibited the growth of granulocyte-macrophage colony-forming units from chronic myelogenous leukemia patients compared with healthy controls.

Conclusion: These data establish LPAAT-β as a potential drug target for the treatment of BCR-ABL-positive leukemias.

Chronic myelogenous leukemia (CML) is a hematopoietic stem cell disorder caused by the oncogenic fusion protein Bcr-Abl (1), a chimeric tyrosine kinase resulting from the (9;22) chromosomal translocation (2). The constitutively active Bcr-Abl kinase leads to tyrosine phosphorylation of substrates, thereby stimulating signaling pathways that promote cell growth and survival (3, 4). Leukemogenesis driven by Bcr-Abl depends on activation of the Ras/mitogen-activated protein kinase (MAPK) pathway (1, 5). Active Ras can bind and target the cytoplasmic Raf kinase to the cell membrane, where a kinase cascade involving MAPK kinase 1 (MAPKK1 or MEK1) and MAPK1/2 [extracellular signal-regulated kinase (ERK) 1/2] is initiated (6, 7). Raf activation requires phosphatidic acid, a lipid cofactor derived from the acetylation of lysophosphatidic acid by lysophosphatidic acid acyltransferase (LPAAT; ref. 8). LPAATs are an isoenzyme family, with the β isoenzyme (LPAAT-β) showing limited expression in normal tissues but high expression in a variety of tumor cell lines (9). The fact that LPAAT-β expression is tumor specific and that overexpressed LPAAT-β contributes to transformation in vitro have implicated this enzyme as a potential drug target for malignant diseases (9).

CML can effectively be treated with the Bcr-Abl-directed kinase inhibitor imatinib (Gleevec), with excellent response rates in early-stage CML (10). However, patients in advanced CML relapse with high frequency (11). Moreover, imatinib fails to induce a complete cytogenetic response in ∼15% of patients in chronic phase CML, and only a minority of patients achieve complete molecular remission as detected by highly sensitive PCR assays, a situation referred to as disease persistence at the molecular level (12). Several mechanisms causing primary resistance have been proposed, including quiescence of leukemic progenitor cells (13), drug efflux (14), mutations in the imatinib-binding kinase domain (15), or the activation of alternative pathways, including MAPK (16). It is thought that the latter may be compensating for inactivation of the Bcr-Abl kinase.

CT32228, an aryldiaminotriazine, is a small-molecule isoenzyme-specific inhibitor of LPAAT-β. The antiproliferative IC50 in a screen of multiple solid tumor cell lines was in the low nanomolar range (17). CT32228 and related compounds are highly antiproliferative and proapoptotic in multiple myeloma cells (18). Biochemical studies showed inhibition of Ras/Raf/MAPK and phosphatidylinositol 3-kinase/Akt pathways leading to cell cycle arrest and apoptosis (17). However, little is known about the antileukemic potency of LPAAT-β inhibitors in leukemia.

Given the involvement of MAPK signaling in the pathogenesis of CML and its potential role in imatinib resistance, we decided to evaluate the effects of CT32228 in inhibiting LPAAT-β activity with CT32228 in BCR-ABL-positive cell lines and primary CML cells. Moreover, we explored the activity of this small molecule in cell lines resistant to imatinib.

Cell culture and drug preparation. MO7e (19), a human megakaryoblastic cell line, was grown in RPMI 1640 with 20% heat-inactivated fetal bovine serum (Life Technologies, Carlsbad, CA), 2% l-glutamine (Life Technologies), 1% penicillin/streptomycin, and 10 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF; Immunex, Seattle, WA). 32D (20), a murine myeloid hematopoietic cell line, was grown in RPMI 1640 with 10% FBS, 2% l-glutamine, 1% penicillin/streptomycin, and 15% WEHI-conditioned medium as the source of interleukin-3 (IL-3). All other cell lines were grown in RPMI 1640 with 10% fetal bovine serum, 2% l-glutamine, and 1% penicillin/streptomycin: the CML blast crisis cell lines KCL-22 (21), Lama84 (22), and K562 (23) as well as the BCR-ABL-transduced cell lines 32Dp210, BaF/BCR-ABL, BaF/BCR-ABLY253F, BaF/BCR-ABLT315I, BaF/BCR-ABLM351T, and BaF/BCR-ABLH396R (24). The imatinib-resistant sublines AR230-r1 and BaF/BCR-ABL-r1 have been described previously (25); they maintain viability in the continuous presence of 1 μmol/L imatinib due to overexpression of Bcr-Abl.

Imatinib was prepared as a 10 mmol/L stock solution in sterile PBS and stored at −20°C. CT32228 was prepared as a 10 mmol/L stock solution in DMSO and stored at −80°C.

Colony-forming assays. Colony-forming assays in the presence of 50 ng/mL recombinant human stem cell factor, 10 ng/mL recombinant human GM-CSF, and 10 ng/mL recombinant human IL-3 with 3 units/mL recombinant human erythropoietin [for culture of blast-forming unit erythroid (BFU-E)] or without erythropoietin [for culture of granulocyte-macrophage colony-forming unit (CFU-GM)] were done as described previously (26). Normal bone marrow was purchased commercially (AllCells, LLC, Berkeley, CA). Patient samples were collected from patients with newly diagnosed chronic phase CML before treatment with imatinib. Cytoreductive therapy at the time of bone marrow harvest consisted of hydroxyurea only. Approval was obtained from the Oregon Health and Science University Institutional Review Board (Portland, OR) for these studies. Informed consent was provided according to the Declaration of Helsinki. Ficoll-Hypaque-separated cells (Amersham Pharmacia, Uppsala, Sweden) were plated at 5 × 104/mL to 1 × 105/mL density with CT32228 at graded concentrations. Short-term treatment was done in RPMI 1640 containing 10% fetal bovine serum in either the presence or absence of cytokines (IL-3, GM-CSF, SCF, and +/−EPO) with graded concentrations of CT32228. Cytokine concentrations were as given for the semisolid medium protocol (see above). After 3 days, cells were washed and subsequently plated in duplicate in methylcellulose medium as described above. After a 2-week incubation at 37°C, BFU-E and CFU-GM colonies were counted. Results were calculated as the percentage of control. Statistical significance of differences was derived using the t test (SPSS software, SPSS, Inc., Chicago, IL).

Measurement of cell proliferation. The tetrazolium-based 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) or 3-(4-5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium salt (MTS) assays were done as described previously (26). Exponentially growing cells were plated in triplicate or quadruplicate at 5 × 103 per well and exposed to escalating doses of each drug independently or in combination. Cell proliferation was measured with MTT-based (Sigma, St. Louis, MO) or MTS-based viability assay (CellTiter 96 Aqueous One Solution Reagent, Promega, Madison, WI) at 24-hour intervals under exponential growth conditions. Means and SDs generated from three to four independent experiments are reported as the percentage of growth versus control at 72 hours. The IC50 and IC80 values were derived manually from the dose-response curve generated by Microsoft Excel.

Combination studies were designed according to Chou and Talalay (27): cell lines were incubated with increasing doses of drug dilutions or combinations of drug dilutions. Calcusyn software (Biosoft, Cambridge, United Kingdom) was used to calculate the combination index at different levels of growth inhibition as a quantitative measure of the degree of drug interaction. Combination index values greater than 1 indicate antagonism, values equal to 1 indicate additivity, and values lower than 1 indicate synergy. Combined drugs were used at fixed molar ratios to accommodate software requirements. Equitoxic drug doses that produced ∼50% of growth inhibition in single-agent experiments were chosen to determine an appropriate fixed molar ratio of 2 combined drugs.

Immunoblotting. K562 cells were cultured with increasing concentrations of CT32228 or with 0.01% DMSO as solvent control for up to 72 hours, and aliquots of cultures were harvested at certain intervals. For immunoblot analysis of Bcr-Abl, ERK and phosphorylated ERK, total phosphotyrosine, and Akt and phosphorylated Akt, the cells were washed twice with cold PBS and lysed in 1% NP40, 150 mmol/L NaCl, 20 mmol/L Tris [tris(hydroxylmethyl)aminomethane (pH 8.0)], 10% glycerol, and 1 mmol/L EDTA containing 1 mmol/L phenylmethylsulfonyl fluoride, 1 mmol/L orthovanadate, and 10 μg/mL aprotinin. The Abl antibody (clone 24-11) was from Santa Cruz Biotechnology (Santa Cruz, CA); the phosphotyrosine (4G10) was from Upstate Biotechnology (Lake Placid, NY); ERK, phosphorylated ERK, Akt, and phosphorylated Akt antibodies were from Cell Signaling Technology (Beverly, MA); and caspase-3, caspase-8, and caspase-9 antibodies were from BD Biosciences (Palo Alto, CA). The antibody detecting poly(ADP-ribose) polymerase (PARP) cleavage products was purchased from Upstate Biotechnology. Immunoblots were probed with a secondary horseradish peroxidase–conjugated antibody (Promega) and developed with the enhanced chemiluminescence kit (Pierce Biotechnology, Rockford, IL). Autoradiographs were scanned with a LumiImager (Boehringer Mannheim, Mannheim, Germany).

Cell cycle analysis. K562 cells were incubated in the presence of CT32228 or DMSO only for up to 48 hours. Cells were fixed and stained according to the manufacturer's instructions and analyzed on a Guava Technologies Personal Cell Analysis instrument equipped with Cytosoft software (Guava Technologies, Hayward, CA). The relative percentages of cells in G1, S, or G2-M phase were calculated from FL-2 histograms using ModFit LT software.

Detection of apoptosis and activated caspase-3. K562 cells were cultured at 105/mL and incubated in the presence of different concentrations of CT32228 or with DMSO alone, as a control, for 72 hours. The cells were harvested, stained with Alexa Fluor 488 Annexin V and propidium iodide (PI; Molecular Probes, Eugene, OR) according to the manufacturer's instructions, and analyzed on a Becton Dickinson (San Jose, CA) FACSAria flow cytometer. Results based on two independent triplicate experiments are reported as averages ± SE. For flow cytometric detection of activated caspase-3, K562 cells were cultured at 105/mL and incubated in the presence of different concentrations of CT32228 or with DMSO alone, as a control. The cells were harvested and analyzed at 24, 48, and 72 hours according to the manufacturer's instructions (Caspase-3 Detection kit, Calbiochem, San Diego CA). Results based on three independent experiments are reported as averages ± SE.

Additionally, cytospin slides of the respective samples were analyzed by Giemsa staining (28).

CT32228 has antiproliferative activity in BCR-ABL-positive and BCR-ABL-negative hematopoietic cell lines. MTS proliferation assays in the presence of increasing concentrations of CT32228 were done comparing three growth factor-dependent cell lines (BaF/3, 32D, and MO7e) with their growth factor-independent derivatives engineered to express Bcr-Abl (BaF/Bcr-Abl 3p210, 32Dp210, and MO7p210). Growth inhibition by CT32228 was independent of BCR-ABL status (Fig. 1), with IC50 values in the low nanomolar dose range (50-105 nmol/L; Table 1). This suggested that CT32228 affects pathways common to IL-3 (BaF/3 and 32D) or GM-CSF (MO7e) and Bcr-Abl. Consistent with this, IL-3 (WEHI-conditioned medium) failed to rescue BaF/Bcr-Abl 3p210 cells from the effects of LPAAT-β inhibition, although we observed a small but reproducible reduction of CT32228 sensitivity in the presence of IL-3 (Table 1A). Cell lines derived from patients with CML blast crisis (AR230-s, K562, and KCL-22) showed IC50 values between 23 and 90 nmol/L in the same range as the values observed in the cell lines engineered to express Bcr-Abl.

Fig. 1.

Antiproliferative activity of CT32228 analyzed by the MTS assay. Murine (32D and BaF/3) and human (Mo7e) hematopoietic cells were transduced with p210BCR-ABL and compared with their parental lines. Percentage inhibition of untreated control (Y axis) by increasing concentrations of CT32228 (X axis). Points, mean of three independent experiments.

Fig. 1.

Antiproliferative activity of CT32228 analyzed by the MTS assay. Murine (32D and BaF/3) and human (Mo7e) hematopoietic cells were transduced with p210BCR-ABL and compared with their parental lines. Percentage inhibition of untreated control (Y axis) by increasing concentrations of CT32228 (X axis). Points, mean of three independent experiments.

Close modal
Table 1.

Inhibitory concentrations of CT32228 in proliferation assays

A. Antiproliferative activity of CT32228 against a panel of hematopoietic cell lines
Cell lineIC50 (nmol/L)IC80 (nmol/L)
32D 60 95 
32DBCR-ABL 80 112 
Mo7e 100 150 
Mo7eBCR-ABL 105 165 
AR230-s 23 35 
AR230-r1 85 108 
BaF/3 88 125 
BaF/BCR-ABL 85 110 
BaF/BCR-ABL (WEHI) 105 >250 
BaF/BCR-ABL-r1 90 112 
BaF/BCR-ABLY253F 92 135 
BaF/BCR-ABLT315I 85 122 
BaF/BCR-ABLM351T 82 120 
BaF/BCR-ABLH396R 89 125 
K562 56 115 
KCL-22 90 225 
Lama84 62 120 
     
B. Antiproliferative activity of CT32228 against primary hematopoietic progenitor cells.
 
    
 IC50 (nmol/L)
 
 IC80 (nmol/L)
 
 
 Normal
 
CML
 
Normal
 
CML
 
CFU-GM, plus GF 190 240 410 590 
CFU-GM, no GF 170 70 >2,000 110 
A. Antiproliferative activity of CT32228 against a panel of hematopoietic cell lines
Cell lineIC50 (nmol/L)IC80 (nmol/L)
32D 60 95 
32DBCR-ABL 80 112 
Mo7e 100 150 
Mo7eBCR-ABL 105 165 
AR230-s 23 35 
AR230-r1 85 108 
BaF/3 88 125 
BaF/BCR-ABL 85 110 
BaF/BCR-ABL (WEHI) 105 >250 
BaF/BCR-ABL-r1 90 112 
BaF/BCR-ABLY253F 92 135 
BaF/BCR-ABLT315I 85 122 
BaF/BCR-ABLM351T 82 120 
BaF/BCR-ABLH396R 89 125 
K562 56 115 
KCL-22 90 225 
Lama84 62 120 
     
B. Antiproliferative activity of CT32228 against primary hematopoietic progenitor cells.
 
    
 IC50 (nmol/L)
 
 IC80 (nmol/L)
 
 
 Normal
 
CML
 
Normal
 
CML
 
CFU-GM, plus GF 190 240 410 590 
CFU-GM, no GF 170 70 >2,000 110 

NOTE: (A) Given are drug concentrations that induce 50 (IC50) or 80% (IC80) growth inhibition compared with untreated control as derived from MTS/MTT assays. Given are means of three independent experiments. (B) Given are drug concentrations that induce 50 (IC50) or 80% (IC80) growth inhibition of CFU-GM compared with untreated control as derived from colony-forming assays. Given are means of five independent patient samples (see Fig. 2). Conditions with and without growth factors are listed (plus GF and no GF).

The fact that the sensitivity of cell lines to CT32228 was independent of their BCR-ABL status suggested that Bcr-Abl expression did not influence LPAAT-β activity. Consistent with this, we detected comparable levels of LPAAT-β activity in BaF/Bcr-Abl 3p210, MO7p210 cells, and their parental lines (data not shown).

Effects of CT32228 on primary hematopoietic cells. Next, we investigated the effects of CT32228 on primary hematopoietic cells from CML patients and healthy bone marrow donors. In the first set of experiments, we assessed colony formation by mononuclear cells grown in semisolid culture in the continuous presence of CT32228. The compound reduced formation of CFU-GM and BFU-E colonies by normal mononuclear cells in a dose-dependent manner, with IC50 and IC80 values 2 to 10 times higher than for cell lines. However, there was no significant difference between CML patients and healthy donors (Fig. 2A; Table 1B). Because previous studies had shown that inhibition of LPAAT-β activity may be more toxic to proliferating cells than to resting cells (9), we sought to exploit the known capacity of CML progenitor cells to enter the cell cycle in the absence of cytokines (29). Mononuclear cells from CML patients and healthy individuals were cultured in graded concentrations of CT32228, with or without added cytokines. After 72 hours, the cells were plated in semisolid medium supplemented with cytokines, and CFU-GM and BFU-E were counted after 14 days. At 0.125 μmol/L CT32228, the lowest concentration used in these experiments, BFU-E and CFU-GM derived from CML patients were reduced to 18% and 10% of controls, respectively, whereas colony formation by normal MNC was much less affected (∼60% of controls; P = 0.0001). Moreover, whereas there was practically complete inhibition of colony formation by CML cells at concentrations above 0.5 μmol/L, inhibition of normal colonies plateaued at ∼70% at concentrations of 0.5 μmol/L, with no increase in toxicity at higher concentrations (Fig. 2B; Table 1B).

Fig. 2.

CT32228 selectively inhibits colony formation of CML progenitors that grow in the absence of growth factors. Mononuclear cells derived from patients in chronic phase CML (•; untreated or on hydroxyurea, n = 5) or from healthy donors (○; n = 5) were assayed in semisolid medium in the continuous presence of increasing concentrations of CT32228 (A) or after short-term drug exposure in liquid medium without growth factors and subsequent culture in growth factor containing semisolid medium without drug (B). Depending on growth factors added, CFU-GM (left) and BFU-E (right) were counted. Statistical analysis using the t test revealed a significant differential growth-inhibitory activity of CT32228 against CML progenitors that grow in the absence of growth factors (B). Data points represent mean of duplicate analysis and are given as percentage growth inhibition versus untreated control (Y axis).

Fig. 2.

CT32228 selectively inhibits colony formation of CML progenitors that grow in the absence of growth factors. Mononuclear cells derived from patients in chronic phase CML (•; untreated or on hydroxyurea, n = 5) or from healthy donors (○; n = 5) were assayed in semisolid medium in the continuous presence of increasing concentrations of CT32228 (A) or after short-term drug exposure in liquid medium without growth factors and subsequent culture in growth factor containing semisolid medium without drug (B). Depending on growth factors added, CFU-GM (left) and BFU-E (right) were counted. Statistical analysis using the t test revealed a significant differential growth-inhibitory activity of CT32228 against CML progenitors that grow in the absence of growth factors (B). Data points represent mean of duplicate analysis and are given as percentage growth inhibition versus untreated control (Y axis).

Close modal

Effects of CT32228 on cell lines resistant to imatinib. Resistance to imatinib as a result of increased Bcr-Abl expression or mutations in the kinase domain is a significant clinical problem (30). We therefore tested the activity of CT32228 against cell lines representing these common mechanisms of imatinib resistance. Our panel included BaF3 cells expressing several Bcr-Abl kinase domain mutants frequently detected in patients (BaF/BCR-ABLY253F, BaF/BCR-ABLT315I, BaF/BCR-ABLM351T, and BaF/BCR-ABLH396R)(31) and sublines with significantly increased levels of Bcr-Abl (AR230-r1 and BaF3/p210-r1)(25). Only AR230-r1 cells were less sensitive to CT32228 than the parental line, whereas all others showed very similar sensitivity (Table 1).

LPAAT-β inhibitors may eventually be combined with imatinib to overcome drug resistance. We therefore tested the combination of imatinib and CT32228 for its activity in BaF/3 cells expressing wild-type Bcr-Abl, the T315I and M351T mutants, as well as a subclone with increased protein expression. We observed additive to synergistic effects in all cell lines with residual sensitivity to imatinib, but not in the T315I mutant, which is completely resistant to the drug (Fig. 3; Table 2). These results suggest that combinations of LPAAT-β inhibitors and imatinib may be clinically beneficial in patients with imatinib resistance, as long as there is residual sensitivity to imatinib. This is consistent with our previous observations (26).

Fig. 3.

Combined antiproliferative activity of CT32228 with imatinib. BCR-ABL-transduced cell lines sensitive (Baf/BCR-ABL) and resistant (Baf/BCR-ABL-r1, Baf/BCR-ABLM351T, and Baf/BCR-ABLT315I) to the Abl kinase inhibitor imatinib were treated with CT32228 in the presence of increasing concentrations of imatinib (colored lines, concentrations). Percentage inhibition of untreated control (Y axis) by increasing concentrations of CT32228 (X axis). Points, mean of three independent experiments.

Fig. 3.

Combined antiproliferative activity of CT32228 with imatinib. BCR-ABL-transduced cell lines sensitive (Baf/BCR-ABL) and resistant (Baf/BCR-ABL-r1, Baf/BCR-ABLM351T, and Baf/BCR-ABLT315I) to the Abl kinase inhibitor imatinib were treated with CT32228 in the presence of increasing concentrations of imatinib (colored lines, concentrations). Percentage inhibition of untreated control (Y axis) by increasing concentrations of CT32228 (X axis). Points, mean of three independent experiments.

Close modal
Table 2.

Median effect analysis of BCR-ABL-positive cell lines sensitive and resistant to the Abl kinase inhibitor imatinib treated with the combination of imatinib and CT32228

Cell lineDose ratio
IC80IC90
Imatinib:CT32228
Baf/BCR-ABL 2:1 0.97 (0.02) 0.82 (0.05) 
Baf/BCR-ABLM351T 10:1 0.83 (0.79-0.86) 0.73 (0.66-0.79) 
Baf/BCR-ABLT315I 50:1 NA NA 
Baf/BCR-ABL-r1 25:1 0.92 (0.78-1.07) 0.80 (0.62-0.99) 
K562 5:1 1.09 (1.03-1.16) 1.01 (0.94-1.08) 
32Dp210 2:1 0.96 (0.16) 0.87 (0.14) 
Cell lineDose ratio
IC80IC90
Imatinib:CT32228
Baf/BCR-ABL 2:1 0.97 (0.02) 0.82 (0.05) 
Baf/BCR-ABLM351T 10:1 0.83 (0.79-0.86) 0.73 (0.66-0.79) 
Baf/BCR-ABLT315I 50:1 NA NA 
Baf/BCR-ABL-r1 25:1 0.92 (0.78-1.07) 0.80 (0.62-0.99) 
K562 5:1 1.09 (1.03-1.16) 1.01 (0.94-1.08) 
32Dp210 2:1 0.96 (0.16) 0.87 (0.14) 

NOTE: Indicated cell lines were treated with a fixed ratio of increasing concentrations of either drug and analyzed by the MTS assay. Data analyis was done by the Calcusyn software. Only cells with significant growth modulation by either drug alone can be analyzed (all except Baf/BCR-ABLT315I, which is completely resistant to imatinib). Combination index values are given at different levels of growth inhibition (IC80 and IC90) and indicate antagonism (>1), additivity (1), or synergy (<1) of combined treatment. Reported values represent mean of two or three independent experiments. The range of individual results (two experiments) or the SD (three experiments) is indicated in the parenthesis.

Abbreviation: NA, not analyzed.

CT32228 induces a G2-M arrest and apoptosis. To investigate the cellular mechanisms leading to growth inhibition in the presence of CT32228, we treated K562 cells with CT32228 for 1, 6, 12, and 24 hours and did cell cycle analysis by PI staining. For these experiments, a dose of 115 nmol/L was chosen, corresponding to the IC80 in the MTS assays (Table 1). As shown in Fig. 4, K562 cells started to accumulate in the G2-M phase after 6 hours of incubation. At 24 hours, ∼50% of the cells were in G2-M, and an increase of cells in the sub-G1 (hypodiploid) compartment was evident, indicating cell death (Fig. 4). Giemsa staining of K562 cells treated with 115 nmol/L CT32228 for 24 hours showed morphologic evidence of mitotic catastrophe as well as apoptosis (Fig. 5A). To verify specificity of the cell cycle effect, this experiment was also done using another CML cell line (Lama84). As seen with K562 cells, treatment with CT32228 leads to G2-M arrest of Lama84 cells with similar kinetics shown with K562 (data not shown).

Fig. 4.

Cell cycle analysis of K562 cells after treatment with CT32228 and staining with PI. K562 cells were treated with 115 nmol/L CT32228 (IC80 in proliferation assay) and harvested at indicated time points. Histograms show one of three representative results of the statistical analysis done by the ModFit software. It should be noted that the Guava cytometer does time-limited analysis as opposed to count-limited analysis causing total count variability at different time points.

Fig. 4.

Cell cycle analysis of K562 cells after treatment with CT32228 and staining with PI. K562 cells were treated with 115 nmol/L CT32228 (IC80 in proliferation assay) and harvested at indicated time points. Histograms show one of three representative results of the statistical analysis done by the ModFit software. It should be noted that the Guava cytometer does time-limited analysis as opposed to count-limited analysis causing total count variability at different time points.

Close modal
Fig. 5.

Mitotic catastrophe and induction of apoptosis in K562 cells treated with CT32228. A, K562 cells were treated or not with 115 nmol/L CT32228 for 24 hours. Photographs of Giemsa-stained cell smears of untreated (control) and treated (CT32228) cells. Characteristics of mitotic catastrophe and apoptosis are clearly visible (bottom). At 56 nmol/L, only slight changes in cell morphology were seen (data not shown). B, K562 cells were treated with 115 nmol/L CT32228 up to 72 hours and analyzed by flow cytometry after staining with Annexin V/PI (○) and anti–active caspase-3 (•). After 24 hours, a significant proportion of cells underwent apoptosis indicated by Annexin V–positive (PI negative) cells. Activation of caspase-3 was detected at lower level and was consistently visible after 48 hours. Points, mean of percentage positive cells; bars, SD. C, whole-cell lysates of K562 cells untreated or treated with 115 nmol/L CT32228 were analyzed by immunoblotting for cleavage of PARP. Arrows, intact (116 kDa) and cleaved (86 kDa) PARP indicative for activation of apoptosis signaling downstream of the caspase cascade. Representative immunoblot of three independent experiments.

Fig. 5.

Mitotic catastrophe and induction of apoptosis in K562 cells treated with CT32228. A, K562 cells were treated or not with 115 nmol/L CT32228 for 24 hours. Photographs of Giemsa-stained cell smears of untreated (control) and treated (CT32228) cells. Characteristics of mitotic catastrophe and apoptosis are clearly visible (bottom). At 56 nmol/L, only slight changes in cell morphology were seen (data not shown). B, K562 cells were treated with 115 nmol/L CT32228 up to 72 hours and analyzed by flow cytometry after staining with Annexin V/PI (○) and anti–active caspase-3 (•). After 24 hours, a significant proportion of cells underwent apoptosis indicated by Annexin V–positive (PI negative) cells. Activation of caspase-3 was detected at lower level and was consistently visible after 48 hours. Points, mean of percentage positive cells; bars, SD. C, whole-cell lysates of K562 cells untreated or treated with 115 nmol/L CT32228 were analyzed by immunoblotting for cleavage of PARP. Arrows, intact (116 kDa) and cleaved (86 kDa) PARP indicative for activation of apoptosis signaling downstream of the caspase cascade. Representative immunoblot of three independent experiments.

Close modal

Because the hierarchical activation of caspases by proteolytic cleavage is a key event in the extrinsic and intrinsic pathways of apoptosis, we analyzed caspase activation in cellular lysates prepared from K562 exposed to graded concentrations of CT32228. Caspase-3, caspase-8, and caspase-9 were detected by immunoblotting with specific antibodies. However, in multiple repeat experiments and despite using multiple different antibodies, we were unable to detect significant cleavage of any of these caspases on immunoblots even at time points when a considerable proportion of cells were apoptotic as assessed by Annexin V/PI staining (data not shown). Hideshima et al. reported that the levels of cleaved caspase-3 in multiple myeloma cells treated with a CT32228 analogue are relatively low, suggesting that our failure to detect cleavage in the K562 cells was related to the limited sensitivity of the immunoblots. We therefore analyzed caspase-3 activation by fluorescence-activated cell sorting. Cells were exposed to 115 nmol/L CT32228 for up to 72 hours and analyzed daily after staining with Annexin V/PI or anti–active caspase-3 (Fig. 5B). After 24 hours, the Annexin V-positive/PI-negative cell fraction was ∼12%, whereas caspase-3 activation became detectable after 48 hours. Immunoblot detection of cleaved PARP, a signaling marker for caspase-dependent apoptosis and, as recently recognized (32), caspase-independent apoptosis, was positive after 24 hours (Fig. 5C).

CT32228 inhibits MAPK signaling in BCR-ABL-positive cells. To exclude any nonspecific effects of CT32228 on Bcr-Abl kinase activity or protein levels, we did immunoblots on lysates from K562 cells treated with graded concentrations of CT32228 for up to 48 hours. As expected, CT32228 did not affect Bcr-Abl protein levels or phosphorylation (Fig. 6). We next examined effects of CT32228 on pathways downstream of Bcr-Abl. CT32228 has been shown to inhibit translocation of Raf1 to the plasma membrane (9). We reasoned that this might interfere with signal transduction from Ras to MAPK. Consistent with this hypothesis, a significant reduction of ERK phosphorylation was detectable at 24 hours and was more pronounced with higher concentrations of CT32228 (Fig. 6). At 48 hours, phosphorylation was partially restored in cells treated with 56 nmol/L CT32228, the concentration corresponding to the IC50 in cell proliferation assays. In contrast to ERK, phosphorylation of Akt was not altered by treatment with CT32228, suggesting that this compound does not affect signaling via the phosphatidylinositol 3-kinase pathway in BCR-ABL-positive cells (Fig. 6).

Fig. 6.

LPAAT-β inhibition by CT32228 does not affect expression or activation of BCR-ABL or AKT but effectively blocks ERK phosphorylation. Whole-cell lysates of K562 cells treated at indicated concentrations of CT32228 for up to 48 hours were analyzed by immunoblotting using antibodies detecting tyrosine phosphorylated proteins (α p-Tyr), Abl and BCR-ABL (α Abl), phosphorylated (α p-Akt) and total Akt (α Akt), and phosphorylated ERK (α p-ERK) and total ERK (α ERK). About BCR-ABL and Akt, no consistent modulation of expression or activation compared with the untreated control sample was observed. Beginning at 12 hours and more pronounced after 24 hours, low- and high-dose treatment render ERK dephosphorylated indicative for strong inhibition of the Ras pathway. Representative immunoblots (sequentially probed and stripped) of two independent experiments.

Fig. 6.

LPAAT-β inhibition by CT32228 does not affect expression or activation of BCR-ABL or AKT but effectively blocks ERK phosphorylation. Whole-cell lysates of K562 cells treated at indicated concentrations of CT32228 for up to 48 hours were analyzed by immunoblotting using antibodies detecting tyrosine phosphorylated proteins (α p-Tyr), Abl and BCR-ABL (α Abl), phosphorylated (α p-Akt) and total Akt (α Akt), and phosphorylated ERK (α p-ERK) and total ERK (α ERK). About BCR-ABL and Akt, no consistent modulation of expression or activation compared with the untreated control sample was observed. Beginning at 12 hours and more pronounced after 24 hours, low- and high-dose treatment render ERK dephosphorylated indicative for strong inhibition of the Ras pathway. Representative immunoblots (sequentially probed and stripped) of two independent experiments.

Close modal

To assess the potential of LPAAT-β inhibition as a treatment of BCR-ABL-positive leukemias, we tested the effects of CT32228, a specific inhibitor of LPAAT-β in a panel of BCR-ABL-positive and BCR-ABL-negative cell lines as well as primary cells from CML patients. We found that CT32228 has strong antiproliferative activity in the nanomolar dose range against all cell lines tested, regardless of their BCR-ABL status. Consistent with this lack of differential activity, LPAAT-β activity was not significantly different between Mo7e and Ba/F3 cells transduced with BCR-ABL and the parental lines, and CT32228 did not have detectable effects on Bcr-Abl expression or tyrosine phosphorylation.

Effective and tolerable treatment of leukemia depends on differential activity of the antileukemic agent, which should preferentially kill leukemic cells while sparing normal tissues from cytotoxic effects. Our experiments using short-term liquid cultures followed by colony-forming assays reveal differential activity of CT32228 in CML cells compared with normal progenitor cells only when cultured in the absence of growth factors. Under these conditions, normal cells were protected, whereas in the presence of cytokines, no differential effect was observed. Of note, at increasing concentrations of CT32228, growth factor-independent CML progenitor colonies were practically eradicated, whereas healthy progenitor colonies plateaued at 70% growth inhibition. The difference seen in the absence of cytokines may reflect the fact that only CML cells are capable of entering the cell cycle in the absence of cytokines and could evidently be exploited for in vitro purging of CML bone marrow (29). Of note, Hideshima et al. (18) reported that a CT32228 analogue had only minor effects on the growth of normal mononuclear cells. Whether a differential effect of CT32228 against CML versus normal progenitor cells will be seen in vivo may ultimately depend on local cytokine concentrations in the bone marrow.

Combinations of CT32228 with imatinib revealed additivity or synergism both in imatinib-sensitive and imatinib-resistant cell lines. However, this was limited to cells with residual sensitivity (i.e., sensitive to dose escalation) to imatinib, whereas no synergism or additivity was seen in the T315I mutant (i.e., insensitive to dose escalation). These data are consistent with our previous observations using combinations of imatinib and decitabine or arsenic trioxide (26). Taken together, the results of the present study show that CT32228 has considerable antileukemic activity as a single agent and in combination with imatinib.

Inhibition of proliferation of K562 cells by CT32228 was associated with blockade of the cell cycle in G2-M (Fig. 4). This observation is in line with data reported by Hideshima et al. (18), who analyzed CT32228-related compounds with anti-LPAAT-β activity in multiple myeloma cells. As early as 24 hours after start of treatment with CT32228, K562 showed morphologic changes consistent with mitotic catastrophe, a finding that is consistent with the cell cycle analysis. The mechanism responsible for the G2-M arrest in the K562 cells is unclear. Hideshima et al. (18) suggested that the G2-M arrest may be triggered through a p53-dependent DNA damage response involving up-regulation of p21. However, as K562 cells lack functional p53 (33), this mechanism cannot account for the G2-M arrest in this cell line. Alternatively, a p53-independent mechanism via p14(ARF) may account for the G2-M arrest in our hands (34). Further experiments will be necessary to determine the underlying mechanism.

FACS analysis with Annexin V/PI-negative staining showed that K562 cells treated with CT32228 die via apoptosis (Fig. 5). Consistent with this, we were able to show cleavage/activation of caspase-3 and cleavage of PARP. It should be noted that the extent of caspase-3 cleavage, detectable by FACS but not on immunoblots, was disproportional and delayed in relation to the significant percentage of apoptotic cells appearing as early as 24 hours after start of treatment. This suggests that a caspase-3-independent mechanism must be responsible for apoptosis induction, at least early after initiating treatment. In line with this observation, cleavage of PARP, a readout molecule for caspase-dependent and caspase-independent apoptosis, was detectable before caspase-3 activation (32).

Because lysophosphatidic acid is required for the activation of Raf1, a key intermediate in the Ras/MAPK pathway, depletion of lysophosphatidic acid by LPAAT-β inhibition was expected to inhibit ERK1/2. Consistent with this, ERK1/2 phosphorylation was reduced in cells exposed to CT32228 (Fig. 6). As activation of MAPK signaling has been reported in response to imatinib treatment as a potential resistance-inducing pathway (16), inhibitors that counteract MAPK activity may be of clinical value to overcome imatinib resistance. In contrast to ERK inhibition, LPAAT-β inhibition did not affect phosphatidylinositol 3-kinase/Akt signaling in K562 cells (Fig. 6), although Coon et al. (17) observed a significant effect on Akt signaling in vascular smooth muscle cells and a minor effect in a lymphoma cell line. The biochemical basis for this apparent cell line specificity remains elusive.

In summary our data implicate LPAAT-β as a therapeutic target in BCR-ABL-positive leukemias. Although the clinical usefulness of CT32228 is limited by its low bioavailability due to hydrophobicity, the development of compounds with more favorable pharmacokinetic properties is ongoing. Given the excellent activity of CT32228 in solid tumor cell lines (9), inhibition of LPAAT-β may be a broadly applicable therapeutic strategy in malignant diseases.

Grant support: Howard Hughes Medical Institute, National Cancer Institute grants, The Leukemia and Lymphoma Society, Buroughs Wellcome Fund, T.J. Martell Foundation, and Doris Duke Charitable Foundation (B.J. Druker). American Society of Hematology Clinical/Translational Research Scholar Award (M.W. Deininger). Postdoctoral grant of the Deutsche Krebshilfe, Dr. Mildred Scheel Stiftung, Germany (P. La Rosée).

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.

Note: P. La Rosée and T. Jia contributed equally to this article.

1
Deininger MW, Goldman JM, Melo JV. The molecular biology of chronic myeloid leukemia.
Blood
2000
;
96
:
3343
–56.
2
Rowley JD. A new consistent chromosome abnormality in chronic myelogenous leukaemia detected by quinacrine fluorescence and Giemsa staining.
Nature
1973
;
243
:
290
–3.
3
Puil L, Liu J, Gish G, et al. Bcr-Abl oncoproteins bind directly to activators of the Ras signalling pathway.
EMBO J
1994
;
13
:
764
–73.
4
Bedi A, Zehnbauer BA, Barber JP, Sharkis SJ, Jones RJ. Inhibition of apoptosis by BCR-ABL in chronic myeloid leukemia.
Blood
1994
;
83
:
2038
–44.
5
Sawyers CL, McLaughlin J, Witte ON. Genetic requirement for Ras in the transformation of fibroblasts and hematopoietic cells by the Bcr-Abl oncogene.
J Exp Med
1995
;
181
:
307
–13.
6
Leevers SJ, Paterson HF, Marshall CJ. Requirement for Ras in Raf activation is overcome by targeting Raf to the plasma membrane.
Nature
1994
;
369
:
411
–4.
7
Stokoe D, Macdonald SG, Cadwallader K, Symons M, Hancock JF. Activation of Raf as a result of recruitment to the plasma membrane.
Science
1994
;
264
:
1463
–7.
8
Abraham E, Bursten S, Shenkar R, et al. Phosphatidic acid signaling mediates lung cytokine expression and lung inflammatory injury after hemorrhage in mice.
J Exp Med
1995
;
181
:
569
–75.
9
Bonham L, Leung DW, White T, et al. Lysophosphatidic acid acyltransferase-β: a novel target for induction of tumour cell apoptosis.
Expert Opin Ther Targets
2003
;
7
:
643
–61.
10
O'Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia.
N Engl J Med
2003
;
348
:
994
–1004.
11
Sawyers CL, Hochhaus A, Feldman E, et al. Imatinib induces hematological and cytogenetic responses in patients with chronic myelogenous leukemia in myeloid blast crisis: results of a phase II study.
Blood
2002
;
99
:
3530
–9.
12
Marin D, Kaeda J, Szydlo R, et al. Monitoring patients in complete cytogenetic remission after treatment of CML in chronic phase with imatinib: patterns of residual leukaemia and prognostic factors for cytogenetic relapse.
Leukemia
2005
;
19
:
507
–12.
13
Graham SM, Jorgensen HG, Allan E, et al. Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro.
Blood
2002
;
99
:
319
–25.
14
Thomas J, Wang L, Clark RE, Pirmohamed M. Active transport of imatinib into and out of cells: implications for drug resistance.
Blood
2004
;
104
:
3739
–45.
15
Chu S, Xu H, Shah NP, et al. Detection of BCR-ABL kinase mutations in CD34+ cells from chronic myelogenous leukemia patients in complete cytogenetic remission on imatinib mesylate treatment.
Blood
2005
;
105
:
2093
–8.
16
Chu S, Holtz M, Gupta M, Bhatia R. BCR/ABL kinase inhibition by imatinib mesylate enhances MAP kinase activity in chronic myelogenous leukemia CD34+ cells.
Blood
2004
;
103
:
3167
–74.
17
Coon M, Ball A, Pound J, et al. Inhibition of lysophosphatidic acid acyltransferase β disrupts proliferative and survival signals in normal cells and induces apoptosis of tumor cells.
Mol Cancer Ther
2003
;
2
:
1067
–78.
18
Hideshima T, Chauhan D, Hayashi T, et al. Antitumor activity of lysophosphatidic acid acyltransferase-β inhibitors, a novel class of agents, in multiple myeloma.
Cancer Res
2003
;
63
:
8428
–36.
19
Avanzi GC, Brizzi MF, Giannotti J, et al. M-07e human leukemic factor-dependent cell line provides a rapid and sensitive bioassay for the human cytokines GM-CSF and IL-3.
J Cell Physiol
1990
;
145
:
458
–64.
20
Greenberger JS, Sakakeeny MA, Humphries RK, Eaves CJ, Eckner RJ. Demonstration of permanent factor-dependent multipotential (erythroid/neutrophil/basophil) hematopoietic progenitor cell lines.
Proc Natl Acad Sci U S A
1983
;
80
:
2931
–5.
21
Kubonishi I, Miyoshi I. Establishment of a Ph1 chromosome-positive cell line from chronic myelogenous leukemia in blast crisis.
Int J Cell Cloning
1983
;
1
:
105
–17.
22
Seigneurin D, Champelovier P, Mouchiroud G, et al. Human chronic myeloid leukemic cell line with positive Philadelphia chromosome exhibits megakaryocytic and erythroid characteristics.
Exp Hematol
1987
;
15
:
822
–32.
23
Lozzio CB, Lozzio BB. Human chronic myelogenous leukemia cell-line with positive Philadelphia chromosome.
Blood
1975
;
45
:
321
–34.
24
La Rosée P, Corbin AS, Stoffregen EP, Deininger MW, Druker BJ. Activity of the Bcr-Abl kinase inhibitor PD180970 against clinically relevant Bcr-Abl isoforms that cause resistance to imatinib mesylate (Gleevec, STI571).
Cancer Res
2002
;
62
:
7149
–53.
25
Mahon FX, Deininger MW, Schultheis B, et al. Selection and characterization of BCR-ABL positive cell lines with differential sensitivity to the tyrosine kinase inhibitor STI571: diverse mechanisms of resistance.
Blood
2000
;
96
:
1070
–9.
26
La Rosée P, Johnson K, Corbin AS, et al. In vitro efficacy of combined treatment depends on the underlying mechanism of resistance in imatinib-resistant Bcr-Abl-positive cell lines.
Blood
2004
;
103
:
208
–15.
27
Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors.
Adv Enzyme Regul
1984
;
22
:
27
–55.
28
Brown AB. Hematology: principles and procedures. Philadelphia: Lea & Febiger; 1993. p. 101.
29
Jonuleit T, Peschel C, Schwab R, et al. Bcr-Abl kinase promotes cell cycle entry of primary myeloid CML cells in the absence of growth factors.
Br J Haematol
1998
;
100
:
295
–303.
30
Hochhaus A, Kreil S, Corbin AS, et al. Molecular and chromosomal mechanisms of resistance to imatinib (STI571) therapy.
Leukemia
2002
;
16
:
2190
–6.
31
Hochhaus A, La Rosée P. Imatinib therapy in chronic myelogenous leukemia: strategies to avoid and overcome resistance.
Leukemia
2004
;
18
:
1321
–31.
32
Kang YH, Yi MJ, Kim MJ, et al. Caspase-independent cell death by arsenic trioxide in human cervical cancer cells: reactive oxygen species-mediated poly(ADP-ribose) polymerase-1 activation signals apoptosis-inducing factor release from mitochondria.
Cancer Res
2004
;
64
:
8960
–7.
33
Law JC, Ritke MK, Yalowich JC, Leder GH, Ferrell RE. Mutational inactivation of the p53 gene in the human erythroid leukemic K562 cell line.
Leuk Res
1993
;
17
:
1045
–50.
34
Normand G, Hemmati PG, Verdoodt B, et al. p14ARF induces G2 cell cycle arrest in p53- and p21-deficient cells by down-regulating p34cdc2 kinase activity.
J Biol Chem
2005
;
280
:
7118
–30.