Introduction

Top Abstract Introduction Results Discussion References

Cyclophosphamide [cis-(±)-2-(bis(2-chloroethyl)amino)tetrahydro-2-oxide-2H-1,3,2-oxazaphosphorine] is a leading anticancer agent. Activation of the drug requires hydroxylation by the hepatic microsomal cytochrome P-450 system in vivo. The resulting 4-hydroxylated metabolite spontaneously degrades via -cleavage of its aldophosphamide tautomer to phosphoramide mustard (an active anticancer component) and acrolein (a highly electrophilic, ,-unsaturated aldehyde) (Hohorst et al., 1976; Colvin, 1999; Kehrer and Biswal, 2000). Moreover, aldophosphamide is oxidized by aldehyde dehydrogenase to the inactive metabolite carboxyphosphamide (Bunting and Townsend, 1998; Sladek, 1999; Giorgianni et al., 2000). Other minor oxidations include dechloroethylation of a chloroethyl side chain to dechloroethylated metabolite and chloroacetaldehyde (Ludeman, 1999).

The reaction mechanism of phosphoramide mustard [N,N-bis-2-(2-chloroethyl)phosphorodiamidic acid, R-N(CH₂-CH₂-Cl)₂] involves generation of the intermediate phosphoramide aziridinium ion through an intramolecular nucleophilic attack (cyclization reaction) of the nitrogen on the -carbon of a chloroethyl chain (Ludeman, 1999). Cellular thiols (e.g., GSH¹) and other nucleophiles react rapidly with phosphoramide aziridinium ions, producing thioether products (Gamcsik et al., 1999). Moreover, reversible 4-(alkylthio)cyclophosphamide products have been observed (Hohorst et al., 1976; Niemeyer et al., 1984; Kwon et al., 1987; Lee, 1991a,b).

Reactions of thiols with acrolein (CH₂ = CH-CHO), on the other hand, are via nucleophilic addition at the carbon atom, forming stable thioether compounds (e.g., 3-oxopropyl glutathione) (Ramu et al., 1995, 1996).

The 4-hydroxycyclophosphamide, aldophosphamide, and acrolein metabolites can readily cross the cell membrane. In contrast, phosphoramide mustard bears a negative charge at physiologic pH (pK_a ~4.8) and, thus, is relatively membrane impermeable. Phosphoramide mustard is, therefore, primarily generated intracellularly (Boyd et al., 1986).

In vitro systems have employed 4OOH-CP as an activated congener of cyclophosphamide, which spontaneously degrades to the reactive alkylating metabolites (Blomgren and Hallstrom, 1991). The 4OOH-CP compound is used clinically to purge hematopoietic cells prior to autologous bone marrow transplantation.

The sulfhydryl compounds, mesna (HS-CH₂-CH₂SO₃Na), and WR-2721 [amifostine, S-2-(3-aminopropylamino)ethyl phosphorothioic acid, ⁺H₃N-(CH₂)₃-NH₂⁺-(CH₂)₂-S-PO₃H] are commonly used to ameliorate toxicities of cyclophosphamide and platinum-based compounds (Brock et al., 1982; Souid et al., 1999; Sadowitz et al., 2002). The protective mechanism of mesna and WR-1065 [WR-2721 active metabolite, ⁺H₃N-(CH₂)₃-NH₂⁺-(CH₂)₂-SH] involves the thiolate anions that participate in chemical reactions similar to those of GSH (Danehy and Noel, 1960). However, the reaction of mesna with the chloroethyl moieties of phosphoramide mustard is unfavorable (Seitz et al., 1989).

The distribution of WR-1065 and mesna differs markedly. WR-1065 distributes equally between the extra- and intracellular compartments, whereas mesna distributes mostly in the extracellular compartment (Newton et al., 1996; Souid et al., 2001).

The aim of the present study is to investigate the kinetics of reactions that are important for optimizing the clinical use of WR-1065 and mesna with high-dose alkylating agents. We estimated the rate constants for reactions of the protective thiols and GSH with 4OOH-CP and acrolein in a cell-free medium at physiologic pH (~7.5). We also investigated the kinetics of depletion of GSH inside PBMC by 4OOH-CP and acrolein in the presence and absence of mesna and WR-1065. The protective capacity of WR-1065 is compared with that of mesna, to provide a framework for drug thiol use during administration of high-dose alkylating agents.

Materials and Methods

Chemicals. 4OOH-CP (D-18864, MW 293.1) was obtained from ASTA Medica AG (Frankfurt, Germany); mesna (MW 164.18, purchased as 100 mg/ml solution) was obtained from Bristol-Myers Squibb Co. (Princeton, NJ); WR-1065 · 2HCl (MW 207.16) was obtained from US Bioscience (West Conshohocken, PA); Ficoll-Paque from Amersham Biosciences AB (Uppsala, Sweden); methanesulfonic acid (MSA) and tris(hydroxymethyl)-aminomethane (Tris) was purchased from Fluka BioChemika (Ronkonkoma, NY); mBBr was purchased from Molecular Probes Inc. (Eugene, OR); acrolein (90%, or ~15 M solution), GSH, 5,5'-dithio-bis(2-nitrobenzoic acid) (DTNB) and anhydrous Na₂HPO₄ were obtained from Sigma-Aldrich (St. Louis, MO); and RPMI 1640 medium (pH 7.15, without L-glutamine) were purchased from Mediatech (Herndon, VA).

Solutions. Mesna and WR-1065 solutions were prepared in dH₂O and stored at 70°C in small aliquots; their concentrations were determined by titration with DTNB immediately prior to each use (Souid et al., 1998). Solutions of 4OOH-CP and acrolein were made in dH₂O immediately prior to their use in various experimental procedures and kept on ice. The mBBr (0.1 M in acetonitrile) and 2 µM thiol-bimane standard solutions were prepared and stored as described (Souid et al., 1998).

Reactions of GSH, Mesna, and WR-1065 with 4OOH-CP and Acrolein. Reactions were carried out at 22 ± 2°C in flat-bottom glass vials with rapid stirring via a magnetic stirring bar in a total volume of 200 µl containing 20 mM Na₂HPO₄ and the thiol (final pH, ~7.5). Reactions were started by adding the alkylating agent (4OOH-CP or acrolein). At times indicated in the figures, reactions with thiol were terminated by adding 0.8 ml of 5 mM (final concentration) mBBr (diluted in 10 mM Na₂HPO₄ immediately prior to each addition), which reacts with the thiolate anion to form a fluorescent thioether derivative. Dilution of the medium by the addition of mBBr reduces the rate of reaction of thiol with alkylating agent by ~5-fold. The samples were then incubated in the dark at room temperature (RT) for 15 min. Twenty micro-liters of 5 M MSA were added, mixed, and extracted with 1.0 ml H₂O-saturated CH₂Cl₂. Portions of the aqueous layers were loaded on high-pressure liquid chromatography (HPLC) and analyzed as described (Souid et al., 1999). A linear calibration curve, from 0 to 100 pmol of mixed thiol standard was generated with each analytical run (r > 0.99). Peak identification was confirmed by retention time in comparison with authentic standards. Quantification was based on peak area against the appropriate reference standards.

(1)

(2)

View larger version (5K): [in this window] [in a new window]   Scheme 1.   Reaction of a thiolate ion with acrolein.

View larger version (6K): [in this window] [in a new window]   Scheme 2.   Reaction of a thiol with phosphoramide mustard.

View larger version (7K): [in this window] [in a new window]   Scheme 3.   Reaction of a thiol with mBBr to form the bimane (thioether) adduct RS-B.

(3)

1

1

(4)

(5)

(6)

1

(7)

(8)

PBMC Collection. Blood was freshly drawn from healthy volunteers into heparin containing Vacutainer tubes (BD Biosciences, San Jose, CA). PBMC were collected and counted as described (Souid et al., 2001).

Incubation of PBMC with Drugs. In all experiments, ~10⁷ cells per condition were incubated in RPMI medium in a final volume of 1.0 ml at 37°C for 60 min. A control sample with no addition was incubated along with the experimental conditions. Mixing was by rapid inversions. Cells were preincubated with WR-1065 or mesna for 5 min before adding 4OOH-CP or acrolein. At the end of the incubation period, the cells were collected by centrifugation, suspended in 20 mM Tris-MSA (pH 8.0) and 5 mM mBBr (final volume 0.5 ml), and incubated in the dark at RT for 15 min. The amount of GSH in the cellular acid-soluble supernatants was determined using HPLC as described (Souid et al., 1998). Most experiments were carried out in triplicates. Calculations were based on the amount of GSH per reaction volume and dry cellular pellet weight; both methods gave the same results. Similar results were obtained at pH 7.7.

	Results

Top Abstract Introduction Results Discussion References

Kinetics of Thiol Reactions with 4OOH-CP and Acrolein in a Cell-Free Solution. We first explored the rates of reactions of mBBr (500 µM) with GSH, mesna, and WR-1065 (100 µM) in 20 mM Na₂HPO₄, pH 7.5 (final volume, 200 µl). The mixtures were incubated at RT for 15 to 120 s and quenched by lowering the pH with 5 µl of 5 M MSA. The results are given in Fig. 1, where the quantity
(see eq. 3) is plotted. The value of k_2' for the reaction of GSH with mBBr was 7.3 ± 0.5 M¹s¹ (t_1/2, 3.4 min), mesna 6.1 ± 0.3 M¹s¹ (t_1/2, 4.0 min), and WR-1065 9.4 ± 0.9 M¹s¹ (t_1/2, 2.6 min), Fig. 1 (r > 0.98 for all three).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Kinetics of thiol reactions with mBBr. The reactions contained 20 mM Na2HPO4 (pH 7.5), either 100 µM GSH, 100 µM mesna or 100 µM WR-1065, and 500 µM mBBr (final volume 200 µl). The mixtures were incubated at RT for indicated time. At the end of the incubation periods, 5 µl of 5 M MSA were added. The samples were then diluted to 1.0 ml with 10 mM MSA and extracted with 1.0 ml H2O-saturated CH2Cl2. The thiol-bimane derivatives were quantitated on HPLC as described under Materials and Methods. Spontaneous oxidation of the thiols during the 120-s period was <1%. Shown are values of ln Q (see eq. 3) calculated from experimental data for GSH (circles), mesna (squares), and WR-1065 (diamonds). Linear least-square fits to ln Q versus t are shown (solid, long-dashed, and short-dashed lines, respectively); k2 is determined from the slope in each case.

1

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Kinetics of thiol reactions with 4OOH-CP. The reactions were performed at 22 ± 2°C in 20 mM Na2HPO4, pH 7.5 (final volume 200 µl). The GSH reactions contained 100 µM GSH and 50 µM 4OOH-CP, the mesna reactions 200 µM mesna and 100 µM 4OOH-CP, and the WR-1065 reactions 100 µM WR-1065 and 75 µM 4OOH-CP. At the end of the incubation periods, 0.8 ml of mBBr was added (final mBBr concentration 5 mM), and the samples were incubated in the dark at RT for 5 min. Five µl of 5 M MSA were then added, and the samples were diluted to 1.0 ml with 10 mM MSA and extracted with 1.0 ml H2O-saturated CH2Cl2. The supernatants were analyzed on HPLC as described under Materials and Methods. Ln Q (eq. 3) is calculated from the results for GSH (circles), mesna (squares), and WR-1065 (triangles) and fitted to lines (solid, short-dashed, and long-dashed, respectively). The k2 values are derived from the slopes of the lines.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Kinetics of thiol reactions with acrolein. The reactions were performed at 22 ± 2°C in 20 mM Na2HPO4, pH 7.5 (final volume 200 µl). The GSH reactions contained 100 µM GSH and 50 µM acrolein, the mesna reactions 200 µM mesna and 100 µM acrolein, and the WR-1065 reactions 100 µM WR-1065 and 50 µM acrolein. At the end of the incubation periods, 0.8 ml mBBr was added (final mBBr concentration 5 mM), and the samples were incubated in the dark at RT for 5 min. Five µl of 5 M MSA were then added, and the samples were diluted to 1.0 ml with 10 mM MSA and extracted with 1.0 ml H2O-saturated CH2Cl2. The supernatants were analyzed on HPLC as described under Materials and Methods. Values of ln Q (eq. 3) calculated from measurements are plotted (circles for GSH, squares for mesna) and fitted to lines (solid and dashed, respectively) to obtain k2 values. For WR-1065, k2 was too large for measurement (data not shown).

1

1

1

1

1

1

1

1

1

1

1

1

1

1

Rates of the Reversed Chemical Reactions. The rate for removal of thiolates from products formed in the reaction of alkylating agents with thiols (that is, the reverse rate) was investigated by adding 5 mM mBBr to reactions containing 100 to 200 µM thiol (GSH, mesna, or WR-1065) and 50 to 100 µM alkylating agent (acrolein or 4OOH-CP). The mixtures were incubated for 1, 5, or 15 min before the addition of MSA. For all incubation times, there was no change in the amount of free thiol detected by mBBr, implying that, during the longest time interval, the reaction product does not decompose to give free thiol.

Depletion of Cellular GSH by 4OOH-CP and Acrolein. The filled circles in Fig. 4 show [S]/[S]₀, the fraction of PBMC GSH as a function of 4OOH-CP (left panel) and acrolein (right panel) concentrations in the incubation medium. The incubation time was 60 min. Concentrations of 4OOH-CP 10 µM hardly affected cellular GSH. Fitting the results in the left panel to an exponential (dashed curve), we find that 50% cellular GSH depletion occurs for 39 µM 4OOH-CP, 90% for 123 µM, and 95% for 159 µM. Acrolein was more effective than 4OOH-CP at depleting cellular GSH. Fitting the results in the right panel to an exponential (dashed curve) shows that, with acrolein, 50% cellular GSH depletion occurs for 6 µM acrolein, 90% for 20 µM, and 95% for 25 µM. Exponential fits are appropriate if the rate of transfer of alkylating agent across the cell membrane is much faster than the rate of reaction.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Cellular GSH depletion by 4OOH-CP and acrolein. PBMC (~107 cells per condition) were incubated with 0 to 150 µM 4OOH-CP or 0 to 20 µM acrolein at 37°C for 60 min. The cells were then collected by centrifugation, suspended in 20 mM Tris-MSA containing 5 mM mBBr, and incubated at RT for 15 min. The amount of GSH in the cellular acid-soluble supernatants was determined on HPLC. The GSH content in the control sample (cells incubated with no addition) was set as 100%. Concentration of alkylating agent is plotted versus measured fraction GSH recovered. The quantity [S]/[S]0, is the fraction of GSH remaining. Dashed curves are best exponential fits, assuming infinitely fast transfer across cell membrane. Solid curves are fits to complete model, assuming k2 equal to its value for homogeneous solution and finding best value for transfer rate constant.

Kinetics of Cellular GSH Depletion by 4OOH-CP and Acrolein. Figure 5, filled circles, shows PBMC GSH depletion as a function of time with 100 µM 4OOH-CP (left panel) or 20 µM acrolein (right panel) in the incubation medium. If the rate of transfer of alkylating agent across the cell membrane is much faster than the rate of reaction inside the cell, the kinetics becomes pseudo-first-order, and fitting the data to exponentials (dashed curves) is justified. This gives, for 4OOH-CP, a pseudo-first-order rate constant (k₁) of 0.099 ± 0.009 min¹ and t_1/2 = 7.0 ± 0.6 min (r = 0.98). Since [4OOH-CP] = 100 µM, the k₂ for reaction of 4OOH-CP with cellular GSH is 16.5 M¹s¹ (calculated using eq. 7). This is lower than the k₂ for reaction of 4OOH-CP with extracellular GSH, 38 M¹s¹ (Table 1). Acrolein depletes cellular GSH much more quickly than 4OOH-CP; fitting the data for acrolein to exponential decay gives 0.26 ± 0.10 min¹ for the k₁ and t_1/2 = 2.7 ± 1.0 min (r = 0.79). The k₂, assuming [acrolein] = 20 µM, is 220 ± 80 M¹s¹, lower than the value for extracellular GSH, 490 ± 100 M¹s¹ (Table 1).

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Kinetics of cellular GSH depletion by 4OOH-CP (left panel) and acrolein (right panel). PBMC (~107 cells per condition) were incubated with 100 µM 4OOH-CP or 20 µM acrolein at 37°C for various times. At the end of the incubation periods, the reactions were quenched by adding 5 mM mBBr. The cells were then collected by centrifugation, re-suspended in 20 mM Tris-MSA and 5 mM mBBr, and incubated at RT for 15 min. The amount of GSH in the cellular acid-soluble supernatants was determined on HPLC. The GSH content in the control sample (cells incubated at 37°C with no addition) was set as 100%. Filled circles are experimental results. Dashed curves are best exponential fits (assuming very fast transfer of alkylating agents across cell membrane). Solid curves are fits to complete model, assuming k2 equals the value for bulk solution and choosing transfer rate constant to obtain the best fit to the data. The quantity [S]/[S]0 is the fraction of GSH remaining.

Estimation of Transport Rate Across Cell Membrane. The reason that the k₂ values for reactions of 4OOH-CP and acrolein with cellular GSH are smaller than the k₂ values for their reactions with GSH in solution (Table 1) may be that the transport of the alkylating agents across the cell membrane is not extremely fast, or that k₂ is slower inside the cell than in the cell-free medium. However, since the reactants are small molecules that collide by diffusion through solvent, the overall rates leading to product inside and outside the cell should be comparable. If one assumes that k₂ is the same inside and outside the cell, one can estimate the transport rate coefficient. Nevertheless, since the experimental conditions are not identical, these coefficient values should be considered approximates.

(9)

(10)

1

1

10

11

10

11

1

1

1

Cellular GSH Protection by Mesna and WR-1065. Figure 6, left panel, shows PBMC GSH protection by mesna (squares) and WR-1065 (circles) in the presence of 100 µM 4OOH-CP. Figure 6, right panel, shows PBMC GSH protection by mesna (squares) and WR-1065 (circles) in the presence of 20 µM acrolein. The protection (P) is calculated according to

(11)

Measurement of the quantity in the denominator was performed four times, giving 378, 347, 450, and 387 pmol per 50 µl (volume injected into HPLC), i.e., 390 ± 37.5 (mean and root mean square deviation from the mean). If the same 9.6% uncertainty obtains for all measurements of GSH recovered, the uncertainty in each value of P is 13.6%.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Influences of WR-1065 and mesna on GSH depletion by 4-OOH-CP and acrolein. PBMC (~107 cells per condition) were incubated with indicated concentrations of WR-1065 (circles) or mesna (squares) at 37°C for 5 min. Then 100 µM 4OOH-CP (left panel) or 20 µM acrolein (right panel) was added and the incubation continued at 37°C for 60 min. The cells were collected by centrifugation, suspended in 20 mM Tris-MSA containing 5 mM mBBr, and incubated at RT for 15 min. The amount of GSH in the cellular acid-soluble supernatants was determined on HPLC. The GSH content in the control sample (cells incubated with no additions) was set as 100%. The curves are best fits to the S-shaped function of eq. 12 (dashed for mesna, solid for WR-1065).

(12)

Cellular GSH Protection by Intracellular Mesna and WR-1065. The contribution of intracellular drug thiols to cellular GSH protection was also investigated. PBMC (~10⁷ cells per condition) were first incubated (37°C for 30 min) in RPMI medium without or with 300 µM mesna or WR-1065. Cells were then collected by centrifugation, suspended in fresh medium without or with 20 µM acrolein, and re-incubated at 37°C for 60 min. At the end of the incubation period, cells were collected, labeled with mBBr, and analyzed for glutathione-bimane-bimane derivatives as described under Materials and Methods. The percentage of GSH was calculated as 100 times the ratio of amount of GSH found to amount of GSH in the control sample (that is, cells incubated without any addition). The percentage of GSH in the absence of drug thiols was ~6%, whereas the presence of either mesna or WR-1065 raised this to ~40%. Although the k₂ values for WR-1065 and mesna with acrolein are different (Table 1), the long period of the reaction, 60 min, insures that all of the acrolein has been consumed in the alkylation process.

	Discussion

Top Abstract Introduction Results Discussion References

Rate constants were measured for reactions of thiols (GSH, mesna, and WR-1065) with mBBr (Fig. 1). It was concluded that with 5 mM mBBr, the rate of thiol reaction with mBBr is faster than the rate of thiol reaction with 4OOH-CP and acrolein (at the concentrations used in most of our experiments). This means that mBBr at this concentration can be used to quench most thiol reactions with 4OOH-CP and acrolein. If the reaction with mBBr is not fast enough, the term "ln Q" in eq. 3 will not be a linear function of time, as it should be for second-order reactions. Numerical calculations (shown under Results) explained how to treat experimental data when this is the case.

We then examined reactions of GSH, mesna, and WR-1065 with 4OOH-CP and acrolein in Na₂HPO₄ buffer at physiologic pH (~7.5) and 22 ± 2°C (Fig. 2-3). The rate constants obtained from these measurements are given in Table 1. These values are "effective rate constants" since they are calculated using total thiol concentration, whereas thiols react mainly by nucleophilic attack of thiolate anions on the reactive moieties of 4OOH-CP and acrolein. Assuming that thiol ionization is rapid enough so the equilibrium in eq. 1 is maintained, the concentration of thiolate is proportional to thiol concentration. Concentration of available nucleophile is determined by the thiol's pK_a, which is ~8.7 for GSH, ~9.1 for mesna and ~7.7 for WR-1065 (Danehy and Noel, 1960; Whitesides et al., 1977; Shaked et al., 1980; Szajewski and Whitesides, 1980; Newton et al., 1992). Calculating using the equation, pH = pK_a + log (RS/RSH), we find that, at pH 7.5, thiolate anion represents ~6% total GSH, ~2% total mesna, and ~39% total WR-1065. Thus, at pH 7.5, the effective rate constant for WR-1065 is expected to be ~6.5 times greater than that for GSH and ~20 times greater than that for mesna. The results in Table 1 show that the k₂ for WR-1065 reaction with 4OOH-CP is more than 20-fold higher than that for GSH with 4OOH-CP (880 M¹s¹ versus 38 M¹s¹) and about 35-fold higher than that for mesna with 4OOH-CP (880 M¹s¹ versus 25 M¹s¹). The k₂ for WR-1065 reaction with acrolein was too high to measure, but the k₂ for GSH-acrolein (490 M¹s¹) is 0.7 times the k₂ for mesna-acrolein (700 M¹s¹) (Table 1), whereas, from the percentage of thiolates calculated above, the ratio should be 6:2. This implies that factors other than pK_a also determine relative rate constants.

Although the chemical reactions of thiols with alkylating agents are reversible, the rates of the reverse reactions are so slow that they have no effect on the measurements (see Results). This observation (that is, a slow reverse rate) is in agreement with the literature (Esterbauer et al., 1975; Wlodek, 1988; Ramu et al., 1995, 1996).

The alkylating agents are known to produce cellular GSH depletion in vitro as well as in vivo (Gurtoo et al., 1981; Crook et al., 1986; Souid et al., 2001). In the present study, the rates of cellular GSH depletion by acrolein and 4OOH-CP were measured, and the effective rate constants were determined (Fig. 4-5). The measured k₂ for cellular GSH depletion by acrolein was ~13 times larger than that by 4OOH-CP (~220 M¹s¹ versus ~16.5 M¹s¹, see Results), about the same as the ratio of the k₂ for GSH reactions with acrolein and 4OOH-CP in solution (~490 M¹s¹ versus ~38 M¹s¹, Table 1). This suggests that (a) the rate constants for intracellular reaction with GSH are the same as for extracellular reaction and (b) the rate of transport across the cell membrane is fast for both agents.

The concentration of acrolein that depletes cellular GSH by 50% is ~5-fold less than that of 4OOH-CP (~7 µM versus ~35 µM; Fig. 4, right and left panels, respectively). These results are in accord with the previous report in Chinese hamster lung fibroblast cells, showing a total cellular GSH depletion within 30 min incubation at 37°C with either 10 µM acrolein or 100 µM 4OOH-CP (Bunting and Townsend, 1998).

Assuming the rate constants for reactions of thiols with alkylating agents were the same inside as outside cells, we were able to estimate the rate constants (k₁) for the transport across the cell membrane from the data of Figs. 4 and 5. Writing the rate of increase of the alkylating agent concentration inside the cell, d[A]/dt, as k₁{[A]_e  [A]}, we estimated that k₁ for 4OOH-CP was ~0.01 s¹ and k₁ for acrolein was ~0.04 s¹. To completely neglect transport across the cell membrane, the transport rate must be much higher than the reaction rate. To compare these two rates, suppose [A]_e  [A] = 100 µM and k₁ = 0.02 s¹, so the rate of transport is 2 × 10⁶ Ms¹. Then, if k₂ = 100 M¹s¹ and [A] = [thiol] =  [A]_e = 50 µM, the rate of reaction is 2.5 × 10⁷ Ms¹, less than an order of magnitude slower than the transport rate.

That GSH reacts faster with acrolein than with 4OOH-CP (Table 1 and Fig. 5) is due partly to faster transport of acrolein across the cell membrane and partly to larger k₂. The rapidity of the reaction explains the failure to obtain a complete cellular GSH protection even with 25-fold molar excess (i.e., 500 µM) of WR-1065 or mesna (Fig. 6, right panel), whereas a complete cellular GSH protection over 4OOH-CP is obtained with only a 10-fold molar excess of WR-1065 (i.e., 1 mM) (Fig. 6, left panel, circles). Mesna (Fig. 6, left panel, squares) has a more limited cytoprotective capacity, since, in the presence of a 10-fold molar excess of mesna (i.e., 1 mM), cellular GSH was below 50%.

The results in Fig. 6 (left panel) are in accord with the previous study showing that mesna, GSH, and N-acetylcysteine protect Chinese hamster lung fibroblast cells from the toxic effects of 4-OOH-CP (Bunting and Townsend, 1998). In another report, a 100-fold molar excess of mesna ameliorated <50% of the growth inhibitory activity of either 100 µM 4OOH-CP or 100 µM acrolein in various cell types (Blomgren and Hallstrom, 1991). Moreover, binding of cyclophosphamide metabolites to intact cells or cellular components was decreased by only ~50% in the presence of an equal molar concentration of mesna or GSH (Wildenauer and Oehlmann, 1982). The fact that a complete cytoprotection requires significant molar excess of WR-1065 over the alkylating agent (Fig. 6, left panel, circles) may account for the lack of cytoprotection by WR-2721 during very intensive chemotherapy (Shapiro et al., 1998).

Studies of cellular uptake of mesna and WR-1065 demonstrate that significant uptake of mesna does occur but at a level much less than that of WR-1065 (Souid et al., 2001). For both agents, the predominant intracellular form is the free thiol (Souid et al., 2001). The present study shows that cytoprotection by WR-1065 and mesna is additive (Table 2), supporting their clinical use in combination (Souid et al., 1999, 2001).

The cytoprotective capacity of a thiol drug is influenced by its pK_a, its rate of reaction with the alkylating agent, and the rate of its uptake by the cell. The data show a superior protection of cellular GSH by WR-1065. This observation is interpreted in terms of the rate constants derived from kinetic analysis of the measured depletion data. A kinetic model to interpret the protection data is also possible but would require assumed values for several parameters in addition to the measured rate constants.