[CANCER RESEARCH 49, 554-559, February 1, 1989] Chlorambucil Pharmacokinetics and DNA Binding in Chronic Lymphocytic Leukemia Lymphocytes1 Bruce B. Bank2, Denise Kanganis, Leonard F. Liebes, and Robert Silber3 Department of Medicine, New York University Medical Center, New York, New York 10016 ABSTRACT damaging these macromolecules (10). Hill, using tritiated CLB, demonstrated the presence of radioactivity in DNA from Yosh- Chlorambucil (CLB) uptake by chronic lymphocytic leukemia lympho ida ascites sarcoma cells exposed to the labeled drug (9). Data cytes was studied using a radiometrie and a newly developed high- performance liquid chromatography assay. CLB labeled with 14Cin either supporting this concept are also available for cisplatinum, mel phalan, the nitrosoureas, mitomycin C, and busulfan (11-15). the chloroethyl group or phenyl ring was used with identical results. Drug accumulation by the cells was found to peak at 30 s, was independent We used CLB labeled either in the phenyl ring or chloroethyl of temperature, and was proportional to medium CLB concentration over groups to investigate drug-DNA binding in CLL lymphocytes. a wide range. Efflux from cells loaded with CLB and resuspended in Gel filtration and HPLC analysis of DNA modified by enzyme drug-free medium was nearly complete at 30 s. The metabolic inhibitors digestion or neutral thermal hydrolysis was carried out to 2-deoxyglucose and NaN3, the nitrogen mustard transport inhibitor hem- investigate the nature of drug-DNA binding. icholinium-3, and another alkylating agent, melphalan, had no effect on drug uptake. We conclude that CLB enters and exits chronic lymphocytic leukemia lymphocytes by simple diffusion. Cells from 17 patients with MATERIALS AND METHODS all stages of chronic lymphocytic leukemia were studied including three Lymphocyte Isolation with CLB-resistant disease, and no heterogeneity was found in the peak cell-associated CLB content or in metabolite pattern on high-performance Blood from 20 patients who gave informed consent was collected liquid chromatography. These findings make it unlikely that transport or into heparinized syringes. All but three of these patients were untreated cellular drug metabolism are factors in drug resistance. for at least 1 year or had never received therapy. The three treated Drug-DNA binding was found to be temperature-sensitive and in patients were included because they had clinically resistant disease. creased with time of incubation. Gel filtration of DNA before and after Lymphocytes were isolated using Ficoll-Hypaque (Pharmacia, Pisca- enzymatic digestion indicated the presence of drug-DNA adducts. High- taway, NJ) centrifugation (16). Cell counts and sizing were done on a performance liquid chromatography analysis of digested DNA and DNA Coulter ZBI Counter and Channelyzer (Coulter Electronics, Hialeah, treated by neutral thermal hydrolysis suggested the presence of multiple FL). All patients had monoclonal B-cells as determined by a panel of adducts. Most of the radioactivity was found as purine adducts. Studies markers including B-l, Leu-1, and T-l 1 antisera. with CLB labeled at two different sites revealed the presence of the Cell Incubation. Cells (lOVml) were incubated in RPMI1640 medium phenyl group and ethyl chains in the adducts. A survey of patients showed (GIBCO, Grand Island, NY) at pH 7.4 containing human serum increased drug-DNA binding in cells from patients with clinical CLB albumin (Pentex, Miles Laboratories, Kankakee, IL) at a concentration resistance. of 0.4 mg/ml. Cells were counted and their viability tested by erythrosin B exclusion before incubation and at each time point. Cell viability exceeded 90% in all experiments. INTRODUCTION Drug Uptake and Efflux. CLB, melphalan, sodium azide, and hemi- CLB4 is an alkylating agent widely used in the treatment of cholinium-3 were purchased from Sigma, St. Louis, MO. [14C]CLB, patients with CLL. The drug has been shown to cause a reduc labeled in the phenyl ring, or in the chloroethyl group, was obtained from Amersham, Arlington Heights, IL, and from the National Cancer tion in the circulating lymphocyte count and total mass of Institute. The drug was purified before use by HPLC with an isocratic lymphoid tissue, while ameliorating symptoms in a majority of elution using solvent B (see below). Radiochemical purity after this previously untreated patients (1-4). Although CLB has been in procedure was found to be 90-95%. CLB was dissolved in dimethyl use for over 30 years, factors underlying drug action and resist sulfoxide, and was added to the cell suspension avoiding an organic ance are incompletely understood. solvent concentration of more than 0.1%. Uptake was usually studied Previous reports by Begleiter and Goldenberg using LSI78Y at a 330 fiM drug concentration (100 Mg/m')- Most experiments were (5) and by Hill, Harrap, and coworkers who investigated Yosh- performed at this high concentration to optimize the HPLC signal, but ida ascites sarcoma cell lines and CLL lymphocytes (6-9) experiments performed with 33 pM CLB showed similar results. So indicated that CLB uptake occurs by simple passive diffusion. dium azide was used at 10 HIM,and 2-deoxyglucose at a 5 HIMconcen tration. Melphalan stock was made as described (17). Hemicholinium- The present study combines a radiometrie with a newly devel 3 was dissolved directly in the medium to a final concentration of 100 oped HPLC technique, to further investigate this process in MM.Cells were preincubated with inhibitors for 15 min at 37°C. CLL lymphocytes. To measure drug efflux, cells were "loaded" by incubating with CLB Many authors attribute the chemotherapeutic action of al for 30 s, centrifugea at 1,100 x g for 1 min and then transferred to kylating agents to their ability to bind to DNA in target cells, drug-free medium. Quantitation of Cell-associated |14C|CLB. A modification of the Received 8/8/88; revised 10/17/88; accepted 10/27/88. The costs of publication of this article were defrayed in part by the payment method described by /.welling et al. (18) was used. Aliquots (0.3 ml) of of page charges. This article must therefore be hereby marked advertisement in cell suspension were layered on 0.4 ml of Versilube F-50 silicone oil accordance with 18 U.S.C. Section 1734 solely to indicate this fact. (General Electric, Waterford, NY) in 1.5 ml conical plastic Eppendorf ' This work was supported by a grant from the NIH (CAI 1655) and by a grant tubes; after centrifugation at 4,400 x g for 1 min, the supernatant was from the Harry Winston Research Foundation. 2An American Cancer Society Fellow. Current address: Department of Med decanted and the inside of the tube blotted dry. The tip was then severed icine, UMDNJ—New Jersey Medical School, 185 South Orange Avenue, New and placed overnight in a counting vial containing 2 ml of 0.3 N KOH. ark, NJ 07103. Radioactivity was measured in a Beckman LS 7000 scintillation counter 3To whom requests for reprints should be addressed, at Department of following the addition of 15 ml of acidified Aquassure (New England Medicine, New York University Medical Center, 560 First Avenue, New York, Nuclear, Boston, MA). We determined medium trapping with [3H]- NY 10016. 4 The abbreviations used are: CLB, Chlorambucil; CLL, chronic lymphocytic inulin to be 0.6 Ml/sample; this volume was corrected for in all calcu leukemia; HPLC, high-performance liquid chromatography. lations. 554 Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1989 American Association for Cancer Research. CLB UPTAKE AND DNA BINDING IN CLL LYMPHOCYTES Extraction of CLB for HPLC. The extraction procedure is a modifi patient with clinical CLB resistance was similar to that obtained cation of the one described for plasma by Ahmed et al. (19). The cell with cells from untreated patients. suspension (0.3 ml) was layered on 0.4-ml Versilube F-50 in 3-ml Pyrex As shown in Table 1, 2-deoxyglucose, sodium azide, hemi- conical tubes and centrifuged at 4,000 x g for 1 min. The supernatant cholinium-3, and melphalan, had no effect on the maximal was decanted, the inside of the tube blotted dry, and 1 ml of water amount of cell-associated drug. These data suggest that CLB added. The cells were dispersed by agitation on a Vortex Genie (Sci entific Industries, Bohemia, NY) and allowed to stand for IS min at influx occurs via passive diffusion. This conclusion is further 4°C;1.5 ml of acetonitrile containing 0.2% acetic acid was then added, supported by the linear and unsaturable drug accumulation over the mixture agitated again and frozen on dry ice. The tubes were then centrifuged at 4,000 x g for 5 min. One ml of the organic phase was transferred to glass test tubes and taken to dryness in a Speed-Vac concentrator (Savant Instruments, Hicksville, NY). Each sample was redissolved in 80 n\ of solvent B (see below), of which 50 ^1was injected into the HPLC apparatus. Control experiments showed that recovery of CLB standards was between 95 and 105%. Medium trapping using [3H]inulin was quantified at approximately 2 ^I/sample, accounting for 10-20% of the peak cell-associated CLB. The technique described above yielded organic, aqueous and solid phases. In some experiments, this method was applied to cells incubated with radioactive drug, and a sample of each phase removed for scintil 0.5 30 lation counting, to assess the partitioning of label in the cells. Time (minutes) In selected experiments following centrifugation of the cell suspen Fig. 1. Effect of temperature on CLB concentration in medium. Cells were sion through oil, 200 ¡Aofmedium supernatant were extracted with incubated in RPMI 1640 with 0.4 mg/ml human serum albumin. The CLB 400 p\ of acetonitrile/0.2% acetic acid, using the same technique as for concentration was measured by HPLC and expressed as a percentage of the initial value, at 37'C (•)and at 4°C(O).