Introduction

Top Abstract Introduction Materials and methods Results Discussion References

Cyclophosphamide (CY)¹ is one of the most frequently used alkylating agents in the treatment of malignancy and in preparative regimens for bone marrow transplantation. It is a prodrug. At therapeutic concentrations of CY, it is metabolized by CYP2C9 and CYP3A4 to form 4-hydroxycyclophosphamide (HCY) or by unidentified cytochrome P-450 isoform(s) to form deschloroethylcyclophosphamide and chloroacetaldehyde (Ren et al., 1997). Deschloroethylcyclophosphamide has no antitumor effect. Chloroacetaldehyde is a potent pulmonary toxin but is formed in relatively low abundance from CY. HCY is the major active circulating metabolite. HCY enters cells and decomposes (through its tautomer aldophosphamide) to phosphoramide mustard (PM) and acrolein. PM is a bifunctional alkylator of DNA, the ultimate cytotoxic metabolite of cyclophosphamide. Alternatively, HCY is detoxified to 4-ketocyclophosphamide by cytochrome P-450 and carboxyethylphosphoramide mustard (CEPM) by aldehyde dehydrogenase 1 (ALDH1) (Dockham et al., 1992). The formation of CEPM from HCY appears to be the most important detoxifying pathway of HCY (Sladek, 1994).

In bone marrow transplantation, CY (60 mg/kg) is usually administered once a day by i.v. infusion for 2 consecutive days primarily to facilitate engraftment of donor cells, although an antitumor effect may also be obtained. In a pharmacokinetic study in bone marrow transplantation patients, we found that the area under the plasma concentration time curve of HCY increased 54.7% (P < .002) from day 1 to day 2 due to an increased formation clearance of HCY from CY and a decreased elimination clearance of HCY. The decreased elimination clearance of HCY apparently was caused by decreased ALDH1 activity from day 1 to day 2 as measured ex vivo in patient erythrocytes (Ren et al., 1998). ALDH1 accounts for 95% of the human liver ALDH activity forming CEPM from HCY (Dockham et al., 1992) and is the only ALDH isoform present in human erythrocytes (Mathewson and Record, 1986).

The objective of this investigation was to elucidate the mechanism of the decrease in ALDH1 activity after CY administration.

	Materials and Methods

Top Abstract Introduction Materials and methods Results Discussion References

Materials. Indole-3-acetaldehyde (IAL), indole-3-acetic acid (IAA), and NAD were purchased from Sigma Chemical Company (St. Louis, MO). Acrolein was purchased from Aldrich Chemical (Milwaukee, WI). 3-Aminophenol and hydroxylamine hydrochloride were purchased from Fluka Chemical (Ronkonkoma, NY). CEPM and PM were generous gifts from ASTA Medica AG (Frankfurt, Germany). 4-Hydroperoxycyclophosphamide was prepared in our laboratory by published methods (Takamizawa et al., 1975). 4-Hydroperoxycyclophosphamide (5 mM) was reduced to HCY by adding sodium thiosulfate (25 mM) and allowing the mixture to stand on ice for 1 h immediately before incubation. Isoelectric focusing gel and buffers (pH 3-10) were purchased from Novex (San Diego, CA).

Incubations. Human liver cytosol was used to screen potential inhibitors of ALDH. Human livers were obtained from the human liver bank in the Departments of Pharmaceutics and Medicinal Chemistry at the University of Washington (Seattle). Liver was homogenized in 100 mM potassium phosphate buffer (pH 7.4) and centrifuged at 10,000g, 4°C for 30 min. The supernatant was filtered through six-ply surgical gauze and centrifuged at 100,000g, 4°C for 60 min. The resulting supernatant cytosolic fraction was stored at 70°C until use. Protein concentration was determined with Bio-Rad (Oakland, CA) protein assay reagent, with bovine serum albumin as the standard (Bradford, 1976).

Analysis of Acrolein. The concentrations of acrolein in incubations containing HCY (total volume, 0.5 ml) were measured by derivatization with the addition of 200 µl of 5 mg/ml 3-aminophenol, 5 mg/ml hydroxylamine hydrochloride, and 10% (w/v) ferrous sulfate in 2.5 M sulfuric acid, followed by the addition of 200 µl of 16.8% perchloric acid. The sample was sealed, vortexed, and centrifuged for 5 min. The supernatant was transferred to a glass injection vial, capped, and heated at 100°C for 25 min. After the sample was cooled to room temperature, 20 µl of the 16.8% perchloric acid was injected into a Hewlett-Packard 1050 series high-performace liquid chromatography system equipped with a Rainin Microsorb C₁₈ column and a fluorescence detector (excitation, 350 nm; emission, 515 nm). The mobile phase was 10% acetonitrile and 90% 50 mM ammonium phosphate buffer (pH 2.5), delivered at a rate of 1 ml/min. The retention time of 7-hydroxyquinoline (derivatization product of acrolein) was 2.5 min and the run time was 8 min. The concentration of acrolein was quantified by peak height.

Data Analysis. The inhibition of IAA formation by acrolein in ALDH1 prepared from human erythrocytes was first analyzed by a Lineweaver-Burk plot. A competitive inhibition model (Segel, 1975) thus was found to be the best model to fit these data: The value of K_i was estimated by fitting this model to the untransformed IAA formation rate and IAL concentration data with the WinNonlin program (Scientific Consulting, Inc., Cary, NC). A weight of V^-1 was used in the iterative fitting process.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

The inhibition of ALDH activity was studied in human liver cytosol incubations with IAL as the probe substrate. HCY (5 and 10 µM) caused a preincubation time- and concentration-dependent decrease of IAA formation rate (Fig. 1). The addition of NAD to the preincubation of human liver cytosol with HCY did not further decrease IAA formation rate (Table 1).

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Time-dependent loss of ALDH activity by HCY in human liver cytosol-144. Human liver cytosol, NAD (0.5 mM), and HCY (0, 5, or 10 µM) were preincubated at 37°C for 0, 2, or 5 min, and the reaction was initiated with the addition of IAL (6 µM) and another aliquot of NAD (0.5 mM) and stopped after 5 min.

Among the metabolites of CY tested (at the maximal observed clinical plasma concentration and 10 times that concentration) for inhibition of ALDH activity, only acrolein showed a pronounced effect, whereas CEPM had a modest effect and no significant inhibition was observed with PM or chloroacetaldehyde (Table 1). Thus, the mechanism of ALDH inhibition by acrolein was investigated.

The effect of acrolein on the formation of CEPM from HCY was first investigated in human liver cytosol incubations. The percentage of inhibition of CEPM formation was 35.4 ± 13.8% and 85.3 ± 4.1% (n = 3), respectively, when 10 or 50 µM acrolein was added exogenously to the incubations compared with no acrolein added exogenously (Table 2). The actual mean acrolein concentrations at the beginning and the end of the incubations were 5, 15, and 55 µM in the control incubation, and in the incubations with the addition of 10 and 50 µM acrolein exogenously, respectively.(Acrolein is formed from the chemical decomposition of HCY during the incubation.)

Because ALDH1 is the major ALDH isoform responsible for the formation of CEPM from HCY, the mechanism of ALDH1 inhibition by acrolein was sought in incubations with ALDH1 prepared from human erythrocytes. Only one band at isoelectric point (pI) 5.2 was detected by isoelectric focusing and staining for ALDH activity in the ALDH1 prepared from human erythrocytes (Fig. 2). This pI is identical with the pI of human liver cytosolic ALDH1. Human liver homogenate and cytosol showed multiple bands, corresponding to various ALDH isoforms. The Lineweaver-Burk plot of IAA formation rate versus IAL concentration at various concentrations of acrolein and the replot of the slope indicated competitive inhibition (Fig. 3). A competitive inhibition model as described in Materials and Methods was fit to the untransformed IAA formation and IAL concentration data. The model-predicted versus the observed IAA formation rate as a function of IAL concentration at various concentrations of acrolein is shown in Fig. 4. The estimated value of K_i was 0.646 µM. Preincubation of acrolein with the ALDH1 preparation for 10 min did not increase the degree of inhibition compared with no preincubation. The reversibility of the inhibition was examined by dialysis. The IAA formation rate (at 1.5 µM IAL) in the presence of 6 µM acrolein was 19.7% of the control activity before dialysis, and it returned to 94.4% of the control activity after dialysis.

View larger version (59K): [in this window] [in a new window]   Fig. 2.   Isoelectric focusing analysis. Isoelectric focusing analysis of human liver homogenate (lane 3) and cytosol (lane 2), and ALDH1 prepared from human erythrocytes (lane 4). Lane 1 is pI standards stained with Brilliant Blue R.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Lineweaver-Burk plot for inhibition of IAA formation by acrolein and replot of the slope of the Lineweaver-Burk plot. (Upper) Lineweaver-Burk plot for inhibition of IAA formation by acrolein in ALDH1 prepared from human erythrocytes. (Lower) Replot of the slope of the Lineweaver-Burk plot at various concentrations of acrolein.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Model predicted versus observed IAA formation rate as a function of IAL concentration at various concentrations of acrolein in ALDH1 prepared from human erythrocytes. The model used was a competitive inhibition model as described in Materials and Methods. The estimated value of Ki was 0.646 µM.

The percentage of inhibition of ALDH1 activity in vivo by acrolein after CY administration was calculated based on the in vitro K_m and K_i for ALDH1, and the peak blood concentrations of HCY and acrolein in patients. Acrolein was estimated to inhibit ALDH1 activity by 85, 88, and 91% on days 1, 2, and 3 (24 h after the dose on day 2) of CY administration, respectively.

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

The major finding of this investigation was that HCY inhibited ALDH activity in human liver cytosol incubations mainly through its degradation product acrolein. CEPM caused some inhibition, but only at very high concentrations. The other CY metabolites evaluated, PM and chloroacetaldehyde, did not show any significant inhibition of ALDH activity even at 10 times their respective maximal clinical plasma concentrations. The inhibition of erythrocyte ALDH1 activity by acrolein was best fit by a competitive inhibition model, independent of preincubation time and reversible by dialysis.

ALDH activity declined with HCY preincubation time in a first-order manner. The addition of the ALDH cofactor NAD did not increase the inhibition by HCY. These data indicated that HCY inhibited ALDH in a time-dependent manner but independent of ALDH catalytic activity. The most likely explanation was that a chemical degradation product of HCY inhibited ALDH. We therefore examined PM and acrolein as inhibitors and compared them to CEPM. We also included chloroacetaldehyde, a very minor metabolite of CY (Ren et al., 1997) because, as an aldehyde, it might inhibit ALDH. Acrolein was the major inhibitor of ALDH, although CEPM showed some inhibition at very high concentrations.

Because ALDH1 has been shown to be the major ALDH isoform for the formation of CEPM from HCY (Dockham et al., 1992), the inhibition mechanism of ALDH1 by acrolein was investigated in ALDH1 prepared from human erythrocytes. Erythrocyte ALDH1 has been demonstrated to be structurally identical with the liver cytosolic ALDH1 and to have the same biochemical and kinetic characteristics as the latter (Helander, 1993). Our isoelectric focusing analysis confirmed that the ALDH1 prepared from human erythrocytes contained only ALDH1, not any other ALDH isoforms, in agreement with previous reports (Agarwal et al., 1983, Sugata et al., 1988).

Acrolein has been shown to be an inhibitor of rat liver mitochondrial and cytosolic ALDH (Mitchell and Petersen, 1988). It has also been shown to be an inhibitor of ALDH1 and ALDH2 purified from human liver (Ferencz-Biro and Pietruszko, 1984). However, the mechanism of inhibition has not been reported for human ALDH1. We found acrolein to be a competitive inhibitor for ALDH1 prepared from human erythrocytes.

Preincubation of human erythrocyte ALDH1 and acrolein for up to 10 min at 37°C did not increase the inhibition of ALDH1 activity, which indicated that the inhibition by acrolein was direct. Dialysis for 6 h at 4°C completely restored ALDH1 activity in ALDH1 preincubated with acrolein, demonstrating that the inhibition was reversible.

It is not practical to directly study the inhibition mechanism of acrolein on CEPM formation with HCY as substrate because acrolein is formed chemically from HCY. However, we were able to show that the formation rate of CEPM was decreased when acrolein was added exogenously to the incubations.

The inhibition of ALDH1 by acrolein in vivo in patients receiving CY was estimated with in vivo substrate and inhibitor concentrations and in vitro K_i obtained with IAL as a substrate. Acrolein appears to be a potent inhibitor of ALDH1 activity at therapeutic concentrations of HCY and acrolein after administration of CY. The inhibition by acrolein was more pronounced on days 2 and 3 than on day 1 (the calculated uninhibited ALDH1 activity was 79% and 60% on days 2 and 3 compared with day 1), which was consistent with the lower elimination clearance of HCY on day 2 than on day 1 of CY administration as we have observed in a clinical study (Ren et al., 1998). Our estimate of the degree of in vivo inhibition of ALDH1 by acrolein is based on the concentration of acrolein measured in blood. Acrolein is very reactive and readily forms conjugates with thiols. It has been shown that acrolein-thiol conjugates readily release acrolein (Alarcon, 1976, Ramu et al., 1996). Thus, our estimate of circulating acrolein concentration may be high. Nonetheless, the data strongly implicate acrolein as a pharmacokinetically significant inhibitor of ALDH1 in patients receiving high-dose CY.

In summary, we found that HCY inhibited ALDH activity through its degradation product acrolein in human liver cytosol incubations. The inhibition of ALDH1 activity by acrolein in incubations with ALDH1 prepared from human erythrocytes was competitive. The inhibition was independent of preincubation time and reversible by dialysis. Acrolein appears to be a potent inhibitor of ALDH1 activity at clinical concentrations of HCY and acrolein.