Accelerated Decomposition of 4-Hydroxycyclophosphamide by Human Serum Albumin1

(CANCER RESEARCH 47, 1505-1508, March 15. 1987] Accelerated Decomposition of 4-Hydroxycyclophosphamide by Human Serum Albumin1 Chul-Hoon Kwon, Karen Maddison, Lynn LoCastro, and Richard F. Borch2 Department of Pharmacology and the Cancer Center, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642 ABSTRACT Sladek and coworkers have reported a shorter half-life in plasma compared to aqueous buffer for metabolites capable of Cyclophosphamide, a widely used anticancer agent, requires initial generating acrolein (21). We have confirmed this for 4-OHCP metabolic activation to 4-hydroxycyclophosphamide (4-OHCP) to elicit and for its 4-alkylthio-substituted derivative, mafosfamide its activity. The rate of decomposition of <â€¢;v-4-<)IK I' was much faster in plasma than in buffer at pH 7.4. This plasma activity was not affected (ASTA Z 7557), using NMR methods (22). However, the basis by treatment with acid (pH 1.3) or heat ((>(!"( for 30 min). The activity for the increased rate of activation in plasma is not understood. was retained in the macromolecular fraction (>10,000) but not in the The objective of this investigation was to identify the catalytic filtrate. Serum albumin was identified as the catalyst for the elimination species in plasma responsible for this increase and to determine step that generates phosphoramide mustard from aldophosphamide; al the mechanistic details of the process. bumin had no effect on the rate of ring opening of cw-4-OHCP to aldophosphamide. This catalytic activity was dependent on serum albu MATERIALS AND METHODS min concentration and independent of pH over the range of 6.5 to 7.5, in contrast to the buffer-catalyzed reaction. The catalytic rate constants *â€ž, Drugs and Chemicals. c/Ã®-4-Hydroperoxycyclophosphamide was pre (pH 7.4, 37Â°C)for phosphate buffer, human serum albumin, and bovine pared as described elsewhere (4, 6). 3.4-Dihydroxybutyl-yV,/V-bis(2- serum albumin were 1.13, 285, and 83 M~'min~', respectively. Pretreat chloroethyOphosphorodiamidate was prepared as described elsewhere ment of eis-4-OHCP with serum albumin resulted in a time-dependent (4). Sodium phosphate, bovine serum albumin (Fraction V, 98.5% decrease in cytotoxic activity against LI 210 tumor cells in vitro. These albumin), human serum albumin (Fraction V, 98% albumin, essentially data suggest that the albumin-catalyzed reaction of cis-4-OIICP in globulin free), crystalline human serum albumin (>99.5% albumin, plasma represents an important pathway for the transformation of cyclo- globulin free), and human serum albumin solution (99.0% albumin, 1% phosphamide metabolites and further emphasize the importance of con globulin, aseptically prepared) were purchased from Sigma Chemical sidering phosphoramide mustard generated extracellularly versus intra- Co., St. Louis, MO. Other organic reagents and solvents were purchased cellularly and the respective contributions of extracellular and intracel- from Aldrich Chemical Co., Milwaukee, WI. Centricon microconcen- lular phosphoramide mustard to Cyclophosphamide cytotoxicity in vivo. trators, used to separate the M, > 10.000 plasma fraction, were pur chased from Amicon Co., Danvers, MA. Human plasma was obtained from the blood bank. Strong Memorial Hospital, University of Roch INTRODUCTION ester, Rochester, NY, and was frozen at -20Â°Cuntil use. Cyclophosphamide, a widely used antineoplastic and immu- NMR Studies. Unless otherwise noted, NMR studies were carried out as described previously (6). 3IP NMR spectra were recorded on an nosuppressive agent, is of great interest due to its relatively IBM WP-270-SY instrument equipped with an 1BM-VSP multinuclear high oncotoxic specificity (1-3) and its complexity in the acti probe tuned for 109.368 MHz using 10-mm sample tubes, a 5000-H?. vation process (Fig. 1). Cyclophosphamide is a prodrug, and its spectral width, a 10 /<spulse width, a 0.8-s pulse repetition time, and activation is initiated by hepatic mixed-function oxidase-cata- 64 scans. Broad-band gated proton decoupling was used. Temperature lyzed C4-hydroxylation. The resulting 4-OHCP' undergoes ring was maintained at 37 Â±2Â°Cusingan IBM VT 1000 variable temperature opening to Aldo, followed by generation of cytotoxic PDA and unit. Peak assignments for all intermediates were based on the published acrolein by ^-elimination. In aqueous buffer, c/s-4-OHCP es data (4, 6). All 31PNMR kinetics was conducted at 37 Â±2"C.Solutions tablishes a pseudoequilibrium mixture consisting of cis-4- of CH-4-OHCP were prepared immediately prior to use by dissolving OHCP, Ãrans-4-OHCP, Aldo, and its hydrate in the approxi the appropriate quantity of c/Ã-4-hydroperoxycyclophosphamide in 50 mate ratio of 48:33:5:14 (4-6). Although the ring opening to 100 ft\ of methanol and treating this solution with 4 equivalents of reaction is an acid-catalyzed process (6), the ,>'diminution is dimethyl sulfide at 37Â°C.The appropriate buffer (prewarmed to 37Â°C) was then added to give a final c/i-4-OHCP concentration of 30 to 50 subject to general base catalysis (7), and activation may also be catalyzed by 3'-5' exonucleases (8-11). There has been a mM (methanol content, <5% of final volume). A typical buffer mixture consisted of 1.7 ml of 100 mM phosphate buffer and 0.3 ml of D2O. controversy over the identity of the circulating metabolite(s) The appropriate quantity of albumin or separated plasma fraction, that enters cells and ultimately exerts cytotoxic activity. Con prepared from 1.6 ml of plasma, was dissolved in 1.7 ml of 100 mM siderable evidence supports the hypothesis that HCP/Aldo phosphate buffer, and this solution substituted for the above phosphate represents the transport form of CP, and that generation of buffer for comparison studies. A typical plasma mixture consisted of PDA occurs within cells sensitive to this drug (1, 2, 12-16). 1.6 ml of human plasma, 0.17 ml of l M phosphate buffer, and 0.23 ml However, other data suggest that the contribution of extracel of I).() The sample was then introduced into the preequilibrated lular PDA may also be important, especially when the higher spectrometer probe, and spectra were acquired at varying intervals. AUC value for PDA is taken into account (17-20). Time points for each spectrum were assigned at the midpoint of data acquisition. The free induction decay spectra were stored on disk and Received 7/21/86; revised 10/23/86; accepted 12/5/86. subsequently processed by exponential multiplication with 2 Hz of line The costs of publication of this article were defrayed in part by the payment broadening, and relative concentrations of intermediates were deter of page charges. This article must therefore be hereby marked advertisement in mined from the peak heights of their respective phosphorus resonances. accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 'This work was supported by Grants CA 34619 and CA 11198 from the Determination of Protein Concentration. Human plasma protein con National Cancer Institute. Department of Health and Human Services, and this centration was measured by using Coomassie Blue G dye-binding assay support is gratefully acknowledged. (23). 2To whom requests for reprints should be addressed. In Vitro Cytotoxic Activity. A soft agar colony-forming assay (24) 1The abbreviations used are: 4-OHCP, 4-hydrooxycyclophosphamide; CP, Cyclophosphamide; Aldo, aldophosphamide; PDA, phosphoramide mustard; was used and modified where necessary. Cultured mouse LI210 cells HSA, human serum albumin; BSA, bovine serum albumin; HCP, hydroxycyclo- were obtained from EG&G Mason Research Institute, Tumor Bank, phosphamide; AUC, area under curve. Worchester, MA. 4-Hydroperoxycyclophosphamide (0.02 mmol, 5.86 1505 Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1987 American Association for Cancer Research. ALBUMIN-CATALYZED DECOMPOSITION OF ACTIVATED CYCLOPHOSPHAMIDE mg) was treated with 2.2 equivalents of sodium thiosulfate (0.044 mmol, 100 100r 10.92 mg) in 0.2 ml of prewarmed acetate buffer (100 mM, pH 5.25, 37Â°C).AVortex mixer was used to facilitate dissolution. The solution was allowed to stand at 37Â°Cfor 5 min to allow complete reduction to CM-4-OHCP, and Fischer's medium (0.84 ml) (Gibco Lab., Grand Island, NY) was added. This solution was immediately sterilized by passage through a 0.22-nm Millipore filter, and 0.96 ml of 10% aspetically prepared human serum albumin was added. The same vol ume of Fischer's medium was substituted for the albumin solution in the control. This gave a 2-ml final volume of stock solution containing 10 mM 4-OHCP and 48 mg/ml of albumin, which was maintained at 37'C. LI210 cells in exponential growth were suspended in Fischer's medium to give 10-ml volumes at a final density of 2 to 3 x 10s cells/ ml after addition of drug solution. After incubating the stock solutions for 0, 15, 30, and 60 min, appropriate volumes of the drug/albumin or 1O 2O 3O 4O SO X) 20 3O 40 SO drug/Fischer's medium stock solution (5 to 150 ^1) were transferred to TIME (MIN) TIME (MIN) the cell suspensions. The drug-cell suspensions were then incubated for Fig. 2. Reaction of c/s-4-hydroxyc>clophosphamide in phosphate buffer (pH l h at 37'C. The cells were washed 3 times with 3 ml of supplemented 7.4, 100 mM, 37Â°C)(A) and plasma (pH 7.4, 37'C) (B). Points were measured Fischer's medium (containing 10% horse serum) and then resuspended from 31P NMR line intensities; the solid lines represent the best-fit values calculated by the Simplex algorithm. Ã€,c/s-4-OHCP; â€¢,rrans-4-OHCP; â€¢Aldo in 5 ml of fresh medium. A 1-ml portion was used to determine the cell and its hydrate; x.

Load more