Abstract
The significantly higher event-free survival rates of Down syndrome (DS) children with acute myeloid leukemia compared with non-DS children is linked to increased sensitivity of DS myeloblasts to 1-β-d-arabinofuranosylcytosine (ara-C) and the enhanced metabolism of ara-C to ara-C triphosphate (J. W. Taub et al., Blood, 87: 3395–3403, 1996). The cystathionine-β-synthase (CBS) gene (localized to chromosome 21q22.3) may have downstream effects on reduced folate and S-adenosylmethionine pathways; ara-C metabolism and folate pools are linked by the known synergistic effect of sequential methotrexate and ara-C therapy. We have shown that relative CBS transcripts were significantly higher in DS compared with non-DS myeloblasts, and CBS transcript levels correlated with in vitro ara-C sensitivity (J. W. Taub et al.,Blood, 94: 1393–1400, 1999). A leukemia cell line model to study the relationship of the CBS gene and ara-C metabolism/sensitivity was developed by transfecting CBS-null CCRF-CEM cells with the CBS cDNA. CBS-transfected cells were a median 15-fold more sensitive in vitro to ara-C compared with wild-type cells and generated 8.5-fold higher [3H]ara-C triphosphate levels after in vitro incubation with[3H]ara-C. Severe combined immunodeficient mice implanted with CBS-transfected CEM cells demonstrated greater responsiveness to therapy, reflected in significantly prolonged survivals after ara-C administration compared with mice implanted with wild-type cells and treated with the same dosage schedule. The transfected cells also demonstrated increased in vitro and in vivo sensitivity to gemcitabine. Deoxycytidine kinase (dCK)activity was approximately 22-fold higher in transfected CEM cells compared with wild-type cells. However, levels of dCK transcripts on Northern blots and protein levels on Western blots were nearly identical between CBS-transfected and wild-type cells. Collectively,these results suggest a posttranscriptional regulation of dCK in CBS-overexpressing cells that contributes to increased ara-C phosphorylation and drug activity. Further elucidating the mechanisms of increased sensitivity of DS cells to ara-C related to the CBS gene may lead to the application of these novel approaches to acute myeloid leukemia therapy for non-DS patients.
INTRODUCTION
DS3children have a 10–20-fold increased risk of developing ALL and AML compared with non-DS children (1, 2). The treatment of DS children with ALL has been complicated by significant toxicity after MTX therapy including myelosuppression, infections, and mucositis(3, 4, 5, 6, 7). On the basis of this observation, it has been speculated that DS individuals have alterations in cellular folate metabolism that enhance the antifolate effects of MTX. The systemic toxicity of DS children with AML treated with ara-C-based protocols has varied with studies reporting that DS patients developed both similar(8) and excessive grades of toxicity compared with non-DS patients (9). A universal observation of the treatment of DS AML patients, however, is that DS patients have significantly EFS rates of >70% (8, 10, 11, 12, 13, 14) compared with ∼37% for non-DS AML patients (15), which indicates that DS myeloblasts have enhanced sensitivity to drugs used in AML therapy. We have reported that DS myeloblasts are more sensitive in vitro to ara-C and daunorubicin compared with non-DS myeloblasts and generate higher intracellular levels of the active intracellular ara-C metabolite ara-CTP, after in vitro incubation with[3H]ara-C compared with non-DS cells. This suggests a biochemical basis for the high EFS rates of DS AML patients(16, 17, 18).
CBS (EC 4.2.1.22), a pyridoxal 5′-phosphate-dependent enzyme (gene localized to chromosome 21q22.3), is involved in the transsulfuration pathway which catalyzes the condensation of serine and homocysteine to form cystathionine (19). Changes in CBS activity in DS cells has been proposed to alter folate metabolism by “trapping”5-methyl tetrahydrofolate, leading to increased MTX toxicity(20). ara-C metabolism and folate pools are linked by the known synergistic effect of sequential MTX and ara-C therapy,leading to lowering of dCTP pools and a greater generation of ara-CTP (21, 22). On this basis, we proposed that increased CBS activity may also enhance ara-C metabolism in DS cells(8, 16, 17). In an analysis of myeloblasts obtained from newly diagnosed pediatric AML patients, we found that relative CBS transcripts detected by quantitative RT-PCR were a median 12-fold higher in DS myeloblasts compared with non-DS myeloblasts, and that relative CBS transcripts correlated with both in vitro ara-C sensitivity and the generation of ara-CTP in myeloblasts from both the DS and the non-DS patients (18).
In this study, we extended these patient studies to a model system by transfecting CBS-null CCRF-CEM cells with the CBS cDNA. This permits the study of the relationships between CBS gene expression and ara-C metabolism/sensitivity as a model of DS blast cells, thus eliminating the confounding effect of other chromosome 21-localized genes.
MATERIALS AND METHODS
Chemicals.
[5-3H]-cytosine-β-d-arabinofuranoside(22 Ci/mmol), [14C]-l-serine (171 mCi/mmol) and [2, 8-3H]-2′-deoxyadenosine 5′-triphosphate (15 Ci/mmol) were obtained from Moravek Biochemicals(Brea, CA). Unlabeled 1-β-d-arabinofuranosylcytosine,daunorubicin, and MTX were obtained from Sigma Chemical Co. (St. Louis,MO). Gemcitabine (2′2′-difluoro-2′-deoxycytidine) was obtained commercially from the Eli Lilly and Co. (Indianapolis, IN). Tissue culture reagents and supplies were purchased from assorted vendors. PCR primers were purchased form Genosys Biotechnologies, Inc. (The Woodlands, TX).
Cell Culture.
The CCRF-CEM, K562, KG-1, HL-60, and U937 leukemia cell lines were obtained from the American Type Culture Collection (Rockville, MD). Cells were maintained in RPMI 1640 containing 10% heat-inactivated calf serum, 100 units/ml penicillin and 100 μg/ml streptomycin in a humidified atmosphere at 37° in the presence of 5%CO2/95% air.
Gene Transfection.
The hCBS cDNA was kindly provided by Dr. W. Kruger(23). After digestion with the restriction enzyme EcoRI, the 2.2-kb hCBS insert was directionally ligated into the pcDNA3 expression vector (Invitrogen, San Diego, CA). The resulting recombinant plasmid pcDNA3-CBS was linearized with EcoRI and transfected into CCRF-CEM cells (20 μg of DNA/5 × 106 cells) by electroporation (300 V, 500 μf) using the Electroporator II(Invitrogen). Forty-eight h after transfection, the cells were plated at 20,000 cells/dish in soft agarose (0.35%) containing 1 mg/ml G418(Sigma). Stably transfected cells were selected after 3–4 weeks,expanded, and screened for CBS transcripts and activity by RT-PCR,Northern analysis, and enzyme assays.
Gene Transcript Analysis by RT-PCR.
Northern Blot Analysis.
Total RNA from each cell line (20 μg) was electrophoresed on 0.9%agarose gels containing 2.2 m formaldehyde and 1×4-morpholinepropanesulfonic acid buffer. The gel was equilibrated in 10× SSC and capillary-transferred to GeneScreen Plus membrane (NEN, Boston, MA) in 10× SSC; the membrane was baked at 80°C under vacuum for 1.5 h. Membranes were prehybridized in QuickHyb solution (Stratagene, La Jolla, CA) for 15 min, then hybridized for 1 h with the addition of[32P]dCTP-labeled dCK cDNA insert labeled by random priming. Nonspecific hybridization was removed by washing membranes in 2× SSC, 0.1% SDS at 42°C and, finally, in 0.1× SSC,0.1% SDS at 60°C. Densitometry of the autoradiograms was performed on a Molecular Dynamics Storm 860 fluorescence and radioactivity imaging system and Image Quant software.
CBS Assay.
This radioisotopic assay measures the amount of cystathionine generated by a reaction of homocysteine and serine catalyzed by CBS(25). The cells suspended in a lysis buffer [50 mm Tris-HCl (pH 8.3), 1 mm mercaptoethanol, 1 mm pyridoxal phosphate, and proteolytic inhibitors] were extracted by freeze-thaw lysis (three times). The CBS assay was performed in a total volume of 100 μl containing 10 mm[14C]serine, 15 mm homocysteine,650 μm pyridoxal phosphate, 2.5 mm EDTA, 100μg protein extract, 1 mm propargylglycine to inhibit cystathionase, 0.5 mg/ml BSA, and 150 mm Tris-HCl (pH 8.3). All of the constituents were equilibrated at 37°C for 5 min followed by the addition of homocysteine to start the reaction. The reaction mixture was incubated for 4 h at 37°C and then spotted onto Whatman 3 M chromatography paper and the[14C]cystathionine product separated from the substrate [14C]serine by ascending chromatography in isopropanol:formic acid:water (70:10:20). After drying, 0.2% ninhydrin reagent was used to detect the marker serine and l-cystathionine (Rf,∼0.25). Radioactive compounds were detected by cutting the chromatogram into 1-cm strips that were solubilized in water and counted directly after the addition of scintillation fluid. Radioactivity from an enzyme-free blank was subtracted to determine the amount of enzymatically formed[14C]cystathionine generated. Enzymatic activity was expressed as pmol of cystathionine formed per mg protein per h at 37°C. Proteins were assayed by a modification of the Lowry assay.
In Vitro Drug Cytotoxicity Assay.
For the determination of cytotoxicity, cells were cultured in complete medium with dialyzed FCS in 24-well culture dishes at a density of 50,000 cells/ml of media. Cells were cultured continuously with a range of concentrations of ara-C, daunorubicin, gemcitabine, or MTX at 37°,and the cell numbers were counted after 4 days with a Coulter counter(Coulter Electronics, Hialeah, FL). The IC50 values were calculated as the concentration of drug necessary to inhibit 50% growth compared with control cells grown in the absence of drug.
ara-C Incubations and Measurement of ara-CTP.
SCID Mice Xenografts and Drug Sensitivity.
Four-week-old immunodeficient Fox Chase ICR SCID mice[Tac:Icr:Ha(ICR)-scidfDf] (male or female) were obtained from Taconic (Germantown, NY) and maintained in specific pathogen-free conditions in microisolator cages stored in laminar flow racks. The mice were fed sterile water and autoclaved food pellets ad libitum.
The CCRF-CEM cell lines (wild-type and transfected) were maintained in RPMI 1640 with heat-inactivated 10% FCS,penicillin/streptomycin, and G418 (transfected lines) prior to the experiments. Initially, mice were implanted s.c. with 1 × 107 cells in serum-free RPMI into each flank. The incidence and growth rate of the tumors were calculated from serial measurement of the length and width of the tumors by Vernier calipers (26, 27).
After the development of palpable tumors (∼1500 mg), the mice were killed by cervical neck dislocation, and small fragments of tumors(∼30 mg), were implanted s.c. bilaterally into the flanks of a second group of mice using a 12-gauge trocar. The mice resumed normal activity after the procedure. Other tumor fragments were dissected and made into single-cell suspensions; the leukemia phenotype was confirmed by cytometry and the stability of expression of the transfected genes assessed by RT-PCR analysis.
Antileukemic drug therapy was initiated 14 days after the implantation of the mice with tumors, with all of the drugs being administered in 0.2 ml of RPMI 1640 i.p.; ara-C was administered daily for 8 days(total dose, 328 mg/kg); and gemcitabine was administered daily for 10 days (total dose, 25 mg/kg). Treated and untreated (control) mice were randomly selected for therapy; each treatment group consisted of four to five mice. After drug therapy, the mice were observed daily to assess changes in tumor size, weight changes, and drug side effects.
Mice were killed in any given trial if the tumor size reached 1500 mg, and no tumor was allowed to exceed 10% of the mouse’s body weight. The assessment of antitumor activity of ara-C and gemcitabine was based on the determination of tumor weight calculated as follows:tumor weight (mg) = (A × B)2/2, where A and B represented the tumor width (mm) and length (mm),respectively.
This protocol was approved by the Wayne State University Animal Investigation committee.
Methylation-specific PCR.
Genomic DNA was isolated from leukemia cell lines using the Puregene System (Gentra System, Minneapolis, MN), and Tri Reagent was used to isolate genomic DNA from clinical AML specimens (18). Bisulfite modification of DNA was performed to convert all unmethylated cytosines to uracil, whereas methylated cytosines would remain unmodified (28). Genomic DNA was initially linearized by shearing through a fine needle followed by alkali denaturation. The denatured DNA (4 μg) was incubated in a total volume of 1.2 ml with 3.1 m sodium bisulfite/0.5 mm hydroquinone (pH 5.0) for 20 h at 50°C and was purified with the QIAquick PCR purification kit (Qiagen, Valencia, CA). CpG Ware primer design software (Intergen, Purchase, NY) was used to modify the dCK DNA sequence (GenBank accession no. L07485) and PCR primers were designed using Oligo 5 software to amplify bisulfite-modified DNA,which would discriminate between methylated and unmethylated DNA. The sense primers were: for dCK1, 5′-TGTTTGGGTGTTTGGTTGTTTGGGGTAGAG-3′ (nt 183–213) and for dCK2, 5′-CACAACACCCCAACCTTACATCCCACATT-3′ (nt 654–683; modified). The antisense primers were: for dCK 3,5′-TTTAGAGTTGGTTGAGAAAGATGGGTAGTT-3′ (nt 102–132); and for dCK 4,5′-CAACACCCTCAAACCTCTAAAATCC-3′ (nt 556–581; modified). PCR amplification was performed with 10 μl of bisulfite-modified DNA in a total volume of 50 μl containing 10 mm Tris-HCl (pH 8.3),1.5 mm MgCl2, 50 mm KCl,5% DMSO, 200 μm each dNTP, 1.25 units of Taq DNA polymerase, and 0.2 μm each dCK primer. PCR amplification was denatured for 2 min at 94°C followed by 32 cycles of 1 min at 94°C, 1 min at 50°C, and 2 min at 72°C, followed by a final extension for 10 min. To determine the methylation status of the dCK promoter, the PCR products were subcloned into the pGEM T-easy vector(Promega, Madison, WI) using the TA cloning kit (Invitrogen) and were sequenced using the BigDye Terminator Cycle Sequencing Ready Reaction kit (PE Applied Biosystems, Foster City, CA).
Western Blot Analysis.
Total protein was isolated from cell lines using the lysis buffer[20 mmol/liter Tris-HCl, 0.9% NaCl (pH 7.6), 0.1% Triton X-100, 1 mmol/l phenylmethylsulphonyl fluoride and 0.01% leupeptin], and the protein concentration of the lysates was determined by the Bio-Rad protein assay (Bio-Rad, Richmond, CA). Forty-μg aliquots of each lysate were fractionated on a 7.5% polyacrylamide gel with SDS and electroblotted onto a nitrocellulose membrane. The blot was blocked overnight at 4°C in TTBS [(Tween Tris-based saline with 0.3% Tween 20 (pH 7.5)] containing 8% fat-free dried milk powder and was then incubated with an anti-sera dCK antibody (kindly provided by Dr. D. Shewach), diluted 1:1000 in TTBS containing 1% fat-free dried milk powder for 2 h at room temperature. The blot was washed with TTBS and incubated with a second antibody (goat antirabbit IgG linked to horseradish peroxidase conjugate, diluted 1:5000 in TTBS-1% milk powder) for 1 h at room temperature, and detected by enhanced chemiluminescence with the SuperSignal CL-HRP Substrate System kit.(Pierce, Rockford, IL).
dCK Enzyme Activity.
The dCK assay measures the ara-C phosphorylation ability of a cell-free extract (29). Cells (1 × 107) were extracted in 50 mm Tris by freeze-thaw lysis (three times). The reaction mixture (120 μl)consisted of 20 μl of cell free supernatant, 10 mm ATP,10 mm MgCl2, 50 mm Tris(pH 7.6), 15 mm NaF, 1 mm tetrahydrouridine (to block cytidine deaminase activity) and an ATP regenerating system consisting of 15 mm phosphoenol pyruvate, 3.6 units of myokinase, and 1.4 units of pyruvate kinase. The reaction was started with the addition of 5 μm[3H]ara-C and incubated at 37° for 30 min. The reaction was terminated by immersion of the reaction mixture in boiling water for 1 min followed by the spotting of 50 μl of the mixture onto DE-81 filter paper. The paper was washed with ice-cold 1 mm ammonium formate followed by additional washes with 70%ethanol. The ara-C nts were eluted off the filter paper in scintillation vials containing 0.1 m HCl and 0.2 m KCl, and the radioactivity was measured and quantitated as pmol/mg protein/min.
Deoxynucleotide Pool Assay.
Cultured cells (1 × 107) were extracted with ice-cold perchloric acid and the endogenous dCTP pools were measured by a DNA polymerase assay as described previously(16).
Statistical Analysis.
Differences in drug sensitivity and parameters of drug metabolism were analyzed by the paired t test. The comparison of survival of the SCID mice groups was analyzed by the Kaplan-Meier method with the Mantel-Cox (log-rank) test using StatView statistical software(Berkeley, CA).
RESULTS
To better determine the relationship of CBS gene expression/enzyme activity to the enhanced sensitivity of DS myeloblasts to ara-C, a leukemia cell line model was developed by transfecting cell lines (with a CBS-negative background) with the CBS cDNA. The human growth factor-independent leukemia cell lines CCRF-CEM,K562, U937, HL-60, and KG-1 were initially screened for CBS expression by RT-PCR. Only for the CCRF-CEM cell line could no CBS transcripts be detected by either RT-PCR or Northern analysis. On this basis,this CBS-null cell line was selected for CBS transfection experiments and subsequent pharmacological and metabolic studies.
Several G418-resistant cell clones were identified after transfection of the CCRF-CEM cell line by electroporation; two of the cell line clones, designated CEM-CBS 21 and CEM-CBS 28, were further characterized for our studies. Both of these clones had CBS transcripts identified by RT-PCR (Fig. 1). CBS enzyme studies revealed enzyme activities 6- and 10-fold higher in the CBS 21 (5.99 pmol/mg protein/h) and CBS 28 (10.6 pmol/mg protein/h) cell lines compared with wild-type (0.47 pmol/mg protein/h)or mock-transfected (0.82 pmol/mg protein/h) CCRF-CEM cells (CEM cells transfected with the pcDNA3 expression vector without the CBS cDNA insert).
In Vitro Drug Sensitivity of Transfected Leukemia Cell Lines.
The CBS 21 and CBS 28 lines were ∼15-fold more sensitive by growth inhibition assay in vitro to ara-C compared with the wild-type and mock-transfected cell lines (ara-C IC50, 6.2 nm and 6.8 nm versus 130.2 nm and 90.9 nm; P = 0.02 and 0.01, respectively (Fig. 2,A). The sensitivity of the cell lines to gemcitabine(2′,2′-difluorodeoxycytidine)—a representative nucleoside agent that undergoes phosphorylation via dCK in a fashion similar to that of ara-C—was also determined. The CBS 21 and CBS 28 lines demonstrated ∼18-fold more sensitivity to gemcitabine, compared with that of the wild-type and mock-transfected CEM lines (Fig. 2 B). There were no significant differences in the percentage of cells in S phase to account for the differences in drug sensitivities (CEM, 49.1%; CBS-MOCK, 51.7%; CBS 21, 40.4%; CBS 28,45.8%). Furthermore, this increased sensitivity was not observed with agents not activated by dCK, because there was no increased sensitivity of the CBS 21 and CBS 28 cell lines to either MTX or daunorubicin compared with the wild-type or mock-transfected cell lines (MTX IC50, 16.1 nm and 16.9 nm versus 19.8 nm and 20.3 nm,respectively; daunorubicin IC50, 14.6 nm and 15.7 nmversus 15.3 nm and 15.6 nm, respectively).
Intracellular Metabolism of Ara-C in Transfected Leukemia Cell Lines.
The increased ara-C sensitivities of the CBS-transfected cell lines,CBS 21 and CBS 28, were accompanied by significantly increased generation in vitro of [3H]ara-CTP after 3-h incubations with [3H]ara-C compared with that in the wild-type CEM line (1919.0 ± 52.7 and 1821.3 ± 95.9 pmol/mg protein versus225.3 ± 59.2 pmol/mg protein, respectively; P = 0.005; Fig. 3 A). There were no differences in endogenous dCTP pools between the wild-type CEM cells (0.08 pmol/μg protein) and the CBS 21 and CBS 28 sublines (0.072 and 0.067 pmol/μg protein, respectively).
dCK Activity and Expression.
The greater level of [3H]ara-CTP synthesis in the transfected lines could reflect increased activity of dCK. dCK enzyme activity was significantly increased (∼22-fold higher) in the CBS-transfected cell lines compared with the CBS-null cells (Fig. 3,B). To determine whether the increased dCK activity in CBS-transfected cells was related to elevated levels of dCK expression,possibly attributable to decreased CpG methylation in the dCK promoter,dCK transcripts from total RNA were measured by Northern analysis. There were no significant differences in the levels of dCK transcripts between the CBS-transfected and CBS-null CEM cells to account for the differences in dCK activity (Fig. 4, upper panel). Likewise, there were no differences in methylation of CpG islands in the dCK promoter, determined by PCR amplification of bisulfite-modified DNA. Western blot analysis using a dCK antibody did not demonstrate increased levels of immunoreactive dCK protein in the CBS-transfected CEM cells compared with that in the CBS-null CEM cells (Fig. 4, lower panel).
SCID Mice Xenografts and Drug Sensitivity.
A SCID mouse model was developed to examine whether the CBS-transfected CEM subline, CBS 28, also exhibited increased in vivo ara-C sensitivity, similar to our in vitro results. Both the CEM and CBS 28 cell lines grew readily as palpable tumors after s.c. injection of the cells. There was stable expression of the transfected CBS gene in tumor fragments of mice implanted with the CBS 28 cells. Likewise, the flow-cytometric leukemia phenotype of the wild-type CEM and CEM subline, CBS 28, remained unchanged during passage of leukemia cells between mice.
At the time of initiation of drug therapy (day 14), there were no significant differences in the sizes of palpable tumors between mice implanted with either the wild-type or the CBS 28 cells in both the untreated and drug-treated groups. All of the mice implanted with either wild-type or CBS-transfected cells and treated with RPMI medium alone, readily developed palpable tumors and were killed once the tumors reached 1500 mg in size. There was no significant difference in survival in days, in mice implanted with wild-type CEM cells and treated with either ara-C or gemcitabine (median, 32 versus 37 days (ara-C) and 35 days (gemcitabine); P = 0.37; Fig. 5,A). In contrast, there was a significant difference in survival in mice implanted with the CBS 28 cell line, with no evidence of palpable tumors after treatment with either ara-C or gemcitabine compared with the untreated control group (P < 0.0001); this group was considered cured (Fig. 5 B).
DISCUSSION
The treatment outcome for AML in children has lagged behind the progress accomplished in the treatment and cure of other major forms of childhood cancer including ALL (15). ara-C remains the single most effective agent for AML therapy, and dose intensification of ara-C seems to have improved AML remission rates (30). The recognition that DS children with AML who are treated predominantly with high-dose ara-C protocols have the highest EFS rates of AML patients (8, 10, 11, 14) has led us to explore the pharmacological and molecular basis for this observation, which may potentially lead to improvements in the therapy of AML (16, 17, 18).
Our previous studies indicated a basis of the high EFS rates of DS AML patients; DS myeloblasts are more sensitive in vitro to ara-C and generate higher intracellular ara-CTP levels (16, 17). We also found that relative transcripts, detected by RT-PCR, of the chromosome 21-localized gene, CBS, were significantly higher in DS myeloblasts compared with non-DS myeloblasts; and relative CBS transcripts correlated with both in vitro ara-C sensitivity and generation of ara-CTP in myeloblasts from both the DS and non-DS patients (18).
In this study we demonstrated that the transfection of wild-type CCRF-CEM leukemia cells (which have very low CBS enzyme activity and lack CBS transcripts on Northern and RT-PCR analysis) with the CBS cDNA resulted in a 15-fold increased sensitivity to ara-C and a 8.5-fold increased generation of [3H]ara-CTP after in vitro incubation with [3H]ara-C in CBS- transfected cell clones compared with wild-type cells. A similar increased sensitivity of CBS-transfected cells to the deoxycytidine analogue, gemcitabine, suggests a common mechanism,likely attributable to increased dCK activity. SCID mice that were implanted with CBS-transfected CEM cells had significantly greater lengths of survival after ara-C and gemcitabine therapy compared with the identical therapy in mice implanted with wild-type CEM cells(median, 150 versus 37 days; P < 0.0001), which suggests the applicability of our DS model in vivo.
Increased CBS activity in DS cells has been proposed as a mechanism that accounts for the significant MTX toxicity of DS patients(20). Although our results suggest that the increased MTX sensitivity of DS patients is not directly linked to the CBSgene alone in our CBS-transfected cell line model, the combined effects of increased gene expression and protein activity of CBS and the reduced folate carrier (RFC; the intracellular transport protein of reduced folates including MTX whose gene is localized to 21q22; Ref.31), may be linked to the MTX sensitivity in DS cells. Notably, the CBS-transfected cells did not demonstrate increased sensitivity to daunorubicin, which indicated that there are other mechanisms that account for the increased in vitrodaunorubicin sensitivity of DS myeloblasts (17, 18).
Our hypothesis for the increased ara-C sensitivity of DS myeloblasts was based on a key role for the CBS enzyme interacting with both reduced folate pools and AdoMet/methylation pathways (16, 30). Increased CBS activity is associated with the low plasma homocysteine levels in DS individuals and the absence of atheroma in DS patients at autopsy (32, 33). By contrast, the genetic disorder, homocystinuria, characterized by mental retardation, lens dislocation, skeletal abnormalities, thromboembolism, and early onset atherosclerosis, is attributable to CBS deficiency, and is associated with elevated homocysteine and methionine levels (19). Increased CBS activity associated with decreased homocysteine and AdoMet levels (34, 35), can result in altered allosteric regulation of the MTHF reductase enzyme and altered gene expression via methylation including the dCK gene. We have previously proposed that increased CBS activity results in enhanced ara-C metabolism and sensitivity by the following mechanism: (a) decreased inhibition of MTHF reductase secondary to lower AdoMet levels,leading to increased diversion of MTHF pools to 5-methyl tetrahydrofolate; (b) reduced synthesis of deoxythymidylate and dTTP resulting from lower MTHF pools; and (c) decreased dCTP pools attributable to reduced inhibition of deoxycytidylate deaminase via dTTP; lower dCTP pools would result in increased phosphorylation of ara-C and less competitive binding between dCTP and ara-CTP to DNA (8, 16, 17).
Consistent with this hypothesis, dCK activity was significantly increased in CBS-transfected cells. The levels of dCK transcripts on Northern blots were nearly identical between CBS-transfected and wild-type cells; and DNA sequencing of PCR products, amplified from bisulfite-modified DNA, showed no differences in methylation status of the dCK promoter, as reported by others (36, 37). Thus,the increased dCK activity is not attributable to increased dCK expression via differences in methylation of the dCK promoter. There was, as well, no significant difference in levels of dCK protein identified by Western blot analysis between the wild-type and CBS-transfected CEM cells. Collectively, these results suggest a posttranscriptional regulation of dCK in CBS-overexpressing cells that contributes to increased ara-C phosphorylation and drug activity. Although the mechanism for this effect is uncertain and is under active investigation, this may, in part, be related to the intracellular localization of the dCK protein (38, 39) or, possibly,effects of dCK phosphorylation (40) or catalytic activity. Although dCTP can allosterically regulate dCK, as noted above, we did not detect differences in dCTP pools between CBS-transfected and wild-type CEM cells; this may reflect the compartmentalization of intracellular dCTP pools (41).
Can CBS expression be altered in leukemia cells in patients to mimic our DS cell line model to improve the effectiveness of ara-C therapy for AML? Translating our in vitro/ex vivo model to the clinic is clearly hampered in vivo by difficulties in selectively targeting leukemia cells for gene transfer because of leukemia being a systemic disease (42). Selective targeting of leukemia cells in vivo has been accomplished using monoclonal antibodies that recognize leukemia-specific surface antigens (e.g., anti-CD 33 monoclonal antibody therapy for AML; Refs. 43 and 44). On this basis,the conjugation of a CBS cDNA/expression vector system complex(45) to an anti-CD 33 monoclonal antibody could potentially result in systemic CBS gene transfer into leukemia cells, thus “sensitizing” myeloblasts prior to ara-C therapy.
Our results demonstrating dramatic increases in both ara-C sensitivity and the metabolism of ara-C to ara-CTP in leukemia cell lines transfected with the CBS cDNA, along with increased ara-C sensitivity of transfected cell lines implanted in SCID mice, provide compelling evidence as to the relationship of CBS gene expression and ara-C metabolism in our model. This mechanism is likely a major factor that accounts for the increased chemotherapy sensitivity of pediatric DS AML patients and provides the first molecular basis for the high cure-rate of DS children with AML. Although our model was developed in a T-cell ALL cell line, the development of a CBS-transfected AML cell line would more closely be a model of DS and AML. Studies in progress are currently developing these cell line models. The further identification of the molecular mechanisms of chemotherapy sensitivity of DS AML patients and the development of new therapeutic approaches based on these mechanisms may lead to significant improvements in the treatment and cure of AML.
Analysis of CBS expression in CCRF-CEM wild-type and sublines. Upper panel, RT-PCR analysis of CBS transcripts in CBS-null CCRF-CEM (wild-type) and CEM sublines(CBS-21, CBS-28) transfected with the CBS cDNA. In the lower panel, the transcript levels are normalized to 18S RNA transcript levels.
Analysis of CBS expression in CCRF-CEM wild-type and sublines. Upper panel, RT-PCR analysis of CBS transcripts in CBS-null CCRF-CEM (wild-type) and CEM sublines(CBS-21, CBS-28) transfected with the CBS cDNA. In the lower panel, the transcript levels are normalized to 18S RNA transcript levels.
In vitro drug sensitivity of the CCRF-CEM wild-type and sublines CEM-MOCK, CBS 21 and CBS 28. Drug sensitivity was measured by the percentage of growth inhibition of the cell lines incubated for 4 days with either ara-C (A) or gemcitabine (B). The data represent the mean of five(ara-C) and two (gemcitabine) separate experiments. The CBS-MOCK line was developed by transfecting CCRF-CEM cells with the pcDNA3 expression vector without the CBS cDNA insert.
In vitro drug sensitivity of the CCRF-CEM wild-type and sublines CEM-MOCK, CBS 21 and CBS 28. Drug sensitivity was measured by the percentage of growth inhibition of the cell lines incubated for 4 days with either ara-C (A) or gemcitabine (B). The data represent the mean of five(ara-C) and two (gemcitabine) separate experiments. The CBS-MOCK line was developed by transfecting CCRF-CEM cells with the pcDNA3 expression vector without the CBS cDNA insert.
In vitro ara-C metabolism in CCRF-CEM wild-type and sublines CBS 21 and CBS 28. In A, after incubation of the CCRF-CEM (wild-type) cell line and CBS-transfected sublines CBS 21 and CBS 28 with 5 μm[3H]ara-C for 3 h, [3H]ara-CTP generation was assessed by high-performance liquid chromatography. The data represent the mean of three or four separate incubations per cell line; bars, SEs. B, the endogenous dCK activity of the CEM cell lines. The data represent the mean of six separate experiments; bars, SEs.
In vitro ara-C metabolism in CCRF-CEM wild-type and sublines CBS 21 and CBS 28. In A, after incubation of the CCRF-CEM (wild-type) cell line and CBS-transfected sublines CBS 21 and CBS 28 with 5 μm[3H]ara-C for 3 h, [3H]ara-CTP generation was assessed by high-performance liquid chromatography. The data represent the mean of three or four separate incubations per cell line; bars, SEs. B, the endogenous dCK activity of the CEM cell lines. The data represent the mean of six separate experiments; bars, SEs.
Analysis of dCK expression and protein levels in CCRF-CEM wild-type and sublines. In the upper panel, total RNA(20 μg) from the CCRF-CEM (wild-type), and CEM sublines CBS 21 and CBS 28 were electrophoresed, blotted onto a nylon membrane, and hybridized with the [32P]dCTP-labeled dCK cDNA as described in “Materials and Methods.” The ethidium bromide-stained gel represents equal loading of total RNA. In the lower panel, total proteins (40 μg) from the CCRF-CEM and CEM sublines CBS 21 and CBS 28 were fractionated on a 7.5% polyacrylamide gel with SDS and electroblotted onto nitrocellulose membranes. Immunoreactive dCK protein was detected with anti-sera dCK antibody and an enhanced chemiluminescence kit.
Analysis of dCK expression and protein levels in CCRF-CEM wild-type and sublines. In the upper panel, total RNA(20 μg) from the CCRF-CEM (wild-type), and CEM sublines CBS 21 and CBS 28 were electrophoresed, blotted onto a nylon membrane, and hybridized with the [32P]dCTP-labeled dCK cDNA as described in “Materials and Methods.” The ethidium bromide-stained gel represents equal loading of total RNA. In the lower panel, total proteins (40 μg) from the CCRF-CEM and CEM sublines CBS 21 and CBS 28 were fractionated on a 7.5% polyacrylamide gel with SDS and electroblotted onto nitrocellulose membranes. Immunoreactive dCK protein was detected with anti-sera dCK antibody and an enhanced chemiluminescence kit.
In vivo antileukemic activity of ara-C and gemcitabine against the CCRF-CEM wild-type (A)and subline CBS 28 (B) in a SCID mouse model. SCID mice were implanted bilaterally with leukemic tumor fragments of either the CCRF-CEM (wild-type) or CBS 28 cell lines and were randomly selected to receive placebo therapy (control) or drug therapy, with each treatment condition consisting of 5 mice. All of the treatments were started on day 14 after implantation of the tumor fragments, with all drugs being injected i.p. in 0.2 ml of RPMI medium; ara-C was administered daily for 8 days (total dose, 328 mg/kg), gemcitabine was administered daily for 10 days (total dose, 25 mg/kg), and the control groups were given RPMI medium alone. Mice were euthanized in any given trial if the tumor size reached 1500 mg, although no tumor was allowed to exceed 10% of the mouse’s body weight. The results shown are the cumulative survival of mice implanted with tumors in relation to the initiation of drug therapy.
In vivo antileukemic activity of ara-C and gemcitabine against the CCRF-CEM wild-type (A)and subline CBS 28 (B) in a SCID mouse model. SCID mice were implanted bilaterally with leukemic tumor fragments of either the CCRF-CEM (wild-type) or CBS 28 cell lines and were randomly selected to receive placebo therapy (control) or drug therapy, with each treatment condition consisting of 5 mice. All of the treatments were started on day 14 after implantation of the tumor fragments, with all drugs being injected i.p. in 0.2 ml of RPMI medium; ara-C was administered daily for 8 days (total dose, 328 mg/kg), gemcitabine was administered daily for 10 days (total dose, 25 mg/kg), and the control groups were given RPMI medium alone. Mice were euthanized in any given trial if the tumor size reached 1500 mg, although no tumor was allowed to exceed 10% of the mouse’s body weight. The results shown are the cumulative survival of mice implanted with tumors in relation to the initiation of drug therapy.
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.
Supported in part by The Leukemia and Lymphoma Society Research Translational Grant (6203-98), Children’s Leukemia Foundation of Michigan, Children’s Research Center of Michigan, Art Gagnon Memorial Fund, BenePro Corporation, Litvak Foundation, and Leukemia Research, Life, Inc. (Detroit, MI).
The abbreviations used are: DS, Down syndrome; SCID, severe combined immunodeficient; EFS, event-free survival; AML, acute myeloid leukemia; ara-C,1-β-d-arabinofuranosylcytosine; ara-CTP, ara-C triphosphate; AdoMet, S-adenosylmethionine; ALL, acute lymphoblastic leukemia; MTX, methotrexate; CBS;cystathionine-β-synthase; RT, reverse transcription; dCK,deoxycytidine kinase; nt, nucleotide; MTHF,5,10-methylenetetrahydrofolate.
Acknowledgments
We thank Dr. Donna Shewach (University of Michigan, Ann Arbor,MI) for providing the dCK antibody and Steven Buck for the flow cytometric analysis of the cell lines.