Introduction

Top Abstract Introduction Materials Results Discussion References

BCB¹ (CAS: 35691-65-7) is a broad-spectrum microbicidal agent used in a variety of industrial and consumer products. BCB is used to control slime-forming bacterial and fungal growth in industrial water cooling systems, latex paints, metal working fluids, joint cement, and adhesives. In the 1990s, this compound has been used as a preservative in cosmetic products. Several reports have indicated that BCB appears to be a weak skin-sensitizing agent in humans (Van Ginkel and Rundervoort, 1995; Hausen, 1993; Mathias, 1983; Andersen and Rycroft, 1991). The absorption, distribution, metabolism, and elimination of BCB in male Fischer 344 rats were reported by Sauer et al. (1998). They found that no parent compound could be detected in the blood after iv administration of BCB. Its primary metabolite, MGN, however, was detected in blood samples up to 1 hr after iv administration. After 72 hr, a significant portion of the dose was still retained in the blood. This bound fraction in the blood could not be extracted with organic solvents. These results suggest that BCB, or possibly one of its metabolites, covalently binds to constituents of whole blood. Such binding might be a mechanism by which BCB could exert toxicity. The objectives of this study were to elucidate the biotransformation of BCB, characterize its macromolecular binding, and predict the possible toxicological ramifications of such events.

	Materials and Methods

Top Abstract Introduction Materials Results Discussion References

Chemicals. BCB was a gift from the Calgon Corp. (Pittsburgh, PA). Radiolabeled [1,3-¹⁴C]BCB was obtained from Wizard Laboratories (West Sacramento, CA). [1,3-¹⁴C]MGN was synthesized by reacting [1,3-¹⁴C]BCB (90 µCi, 7.0 mg) with GSH (34.7 mg), in 1.5 ml of saline, for 3 hr at 37°C. The purity of both [1,3-¹⁴C]MGN and [1,3-¹⁴C]BCB was 99%, as determined by radiochemical HPLC and GC/MS analysis. MGN, Drabkin's reagent, GSH, bathophenanthrolinedisulfonic acid, 2,4-dinitrofluorobenzene, -glutamyl glutamate, and NEM were obtained from Sigma Chemical Co. (St. Louis, MO). All chemicals and reagents used were of the highest purity commercially available.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Time courses for total, extracted, and bound radioactivity in blood after iv bolus administration of [14C]BCB (8 mg/kg, 120 µC/kg) to male Fischer 344 rats with jugular vein cannulation. Data are expressed as mean ± SD. Blood samples from three individual rats were analyzed in duplicate.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Representative HPLC radiochromatograms of [14C]BCB (4 mg/ml, 60 µCi/ml) and its major metabolite, MGN, incubated for 10 min at 37°C with distilled water, whole blood, or whole blood with NEM (40 mM). BCB and MGN were eluted with retention times of 48.0 and 23.0 min, respectively, and were detected by -RAM Model 2. MGN was identified by its GC/MS electron impact spectrum, in comparison with that for the authentic standard.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Representative GC/MS electron impact spectrum of MGN obtained from an incubation of [14C]BCB with blood. The mass spectrum showed a molecular ion corresponding to m/z 106. The major product ion was at m/z 66, which represents a loss of 40 mass units for the methylenecyanide radical.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Representative HPLC chromatograms for stoichiometric analysis of BCB, MGN, GSH, and GSSG. BCB and MGN were eluted with retention times of 48.0 and 23.0 min, respectively, and were detected by -RAM Model 2. -Glutamyl glutamate [internal standard (I.S.)], GSH, and GSSG were eluted with retention times of 23.6, 25.8, and 29.0 min, respectively, and were monitored by UV detection at 365 nm.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Time course for covalent binding of BCB and MGN, with and without NEM coincubation, in whole blood, isolated erythrocytes, and isolated hemoglobin. Data are expressed as mean ± SD. Samples from four individual rat preparations were analyzed in duplicate. **, p < 0.01, compared with NEM-treated groups.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   In vitro dose-dependent cytotoxicity of BCB in purified erythrocytes, from male Fischer 344 rats, incubated at 37°C for 2 hr. Hemolysis was used as a marker of erythrocyte membrane damage. The calculated EC50 for BCB was approximately 1 mM, coincubation with NEM decreased the EC50 to 0.5 mM, and coincubation with GSH increased the EC50 to >3.0 mM. Data are expressed as mean ± SD. Samples from four individual rat preparations were analyzed in duplicate. *, p < 0.05; **, p < 0.01, compared with the BCB-treated group.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Absorption spectra of hemoglobin (Hb) isolated from male Fischer 344 rats and incubated with different concentrations of BCB (0.75-3.05 mM) at 37°C for 15 min.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Overall mechanism of interaction of BCB with erythrocytes.

In Vivo Animal Studies. Male Fischer 344 rats with jugular vein cannulation were purchased from Hilltop Lab Animals (Scottdale, PA). Animals were housed in a facility with a 12-hr light/dark cycle and temperature maintained at 25°C. Animals were acclimated to this environment for a minimum of 5 days. They were provided food (Teklad 4% Mouse-Rat Diet; Harlan Teklad, Madison, WI) and water ad libitum. [¹⁴C]BCB (80 mg/kg, 120 µCi/kg, 2 ml/kg), in a vehicle of 10% emulphor/15% ethanol/75% saline solution, was administered to male Fischer 344 rats via the indwelling jugular vein cannula. The injection was followed by injection of an equal volume of normal saline solution to flush the cannula. Blood samples (300 µl) were collected at various time points (0, 3, 5, 7, 10, 12, 15, 20, 30, 45, and 60 min and 6, 12, 24, and 48 hr). Immediately after collection, two 25-µl aliquots of blood were assayed for total radioactivity using liquid scintillation counting. Blood samples (150 µl) were also extracted with ethyl acetate and analyzed by HPLC as described below. The remaining portions of the blood samples were centrifuged at 325g for 10 min. Plasma (25 µl) and erythrocytes (25 µl) were assayed for ¹⁴C-equivalent contents by liquid scintillation counting.

Biotransformation of BCB in Whole Blood and Buffered GSH Solution. In an attempt to elucidate the role of free sulfhydryl groups in the biotransformation of BCB, [¹⁴C]BCB (4 mg/ml, 60 µCi/ml, 5 µl) was incubated for various times, at 37°C, with fresh rat blood (150 µl) or phosphate (0.2 M, pH 7.4)-buffered GSH (25 mM) solution, with or without NEM (40 mM). At the end of the incubation, samples were treated as described above. Identification of BCB and its metabolites in in vivo blood samples was performed by HPLC and GC/MS analysis.

Radiochemical HPLC and GC/MS Analysis of BCB and Its Metabolites. The HPLC analysis of BCB and its metabolites used in this study was described in detail by Sauer et al. (1998). Briefly, aliquots of blood (150 µl) or GSH incubation mixtures were extracted immediately with ethyl acetate (300 µl) three times. The organic solvent was pooled, and 150 µl of distilled water was added. Analytes were pushed into the aqueous phase by evaporation of the ethyl acetate by vacuum centrifugation. Separation and quantification of analytes were performed with an HPLC system (SP8800; Spectra Physics, San Jose, CA) equipped with an autosampler (AS 3000; Thermo Separation Products-Spectra Systems, Fremont, CA) and a -RAM Model 2 Detector. (with WinFlow; IN/US Systems, Tampa, FL). The samples (100 µl) were injected onto a Whatman Partisil ODS-2 column (250 × 4.6 mm, 10 µm; Whatman, Hillsboro, OR). Analytes were eluted with a mobile phase of water and acetonitrile (both containing 0.1% acetic acid) at a flow rate of 1 ml/min, with a total run time of 80 min. The mobile phase was maintained at 100% water for 5 min and then changed to 20% acetonitrile in 15 min, to 30% acetonitrile in 20 min, and to 100% water in 10 min, with the final conditions being maintained for 10 min. The column was reequilibrated to the initial conditions for 20 min.

In Vitro BCB and MGN Binding Studies. For these studies, whole blood was separated into plasma and erythrocytes by centrifugation at 325g for 10 min. The erythrocyte fraction was washed with buffered saline (pH 7.4) containing 0.01% CaCl₂, 0.02% KCl, 0.02% KH₂PO₄, 0.0047% MgCl₂, 0.8% NaCl, 0.115% Na₂HPO₄, and 0.1% D-glucose. After three washings, the volume of the packed erythrocytes was adjusted with buffered saline. Hemoglobin was prepared according to the method of Zulstra and Buursma (1987).

In Vitro BCB and MGN Erythrocyte Cytotoxicity. To determine whether BCB and MGN could react with erythrocytes directly and cause cytotoxicity, an in vitro study was performed using rat erythrocytes, according to the method of Rael (L.T. Rael, personal communication). Washed erythrocytes (150 µl) were incubated with various concentrations of BCB or MGN (0.3-3.0 mM) for up to 2 hr at 37°C. Erythrocytes were also coincubated with BCB/MGN and NEM (0.5 mM) or GSH (1.5 mM). After incubation, erythrocytes were centrifuged for 7 sec at 11,750g, and the supernatant (50 µl) was mixed with 3.5 ml of Drabkin's reagent for 15 min. Hemoglobin leakage was used to assess erythrocyte membrane damage and was measured spectrophotometrically (540 nm).

Spectral Effects of BCB and MGN on Purified Hemoglobin. Hemoglobin was isolated and purified from rat blood according the method of Zulstra and Buursma (1987). Freshly isolated hemoglobin (150 µl) was incubated with various concentrations of BCB or MGN (0.75-3.05 mM) at 37°C for 15 min. After incubation, sample absorbance was scanned from 450 to 750 nm using a Beckman DU-7 spectrophotometer, with a normal bandwidth of 0.1 nm and light path of 1.0 cm.

Statistical Analysis of Data. Experiments were performed at least three times, and sample size refers to the number of animals. Data are presented as the arithmetic mean ± standard deviation. Statistical analysis of the data for more than two groups was conducted by repeated-measures one-way analysis of variance followed by the Newman-Keuls test. The paired Student test was used to analyze the data for two groups. Differences were considered to be significant at p < 0.05.

	Results

Top Abstract Introduction Materials Results Discussion References

In Vivo Animal Studies. After iv administration of [¹⁴C]BCB, HPLC analysis of blood samples revealed MGN as the single detectable analyte in the blood. No parent compound was detected in the blood even at the earliest time point of 3 min. The binding percentage of ¹⁴C-equivalents in the blood increased gradually over time and consistently represented >5% of the total dose after 1 hr (fig. 1). After 48 hr, approximately 12% of the dose was covalently associated with the blood. The plasma/erythrocyte disposition ratio for covalently bound ¹⁴C-equivalents was 1:4.

Biotransformation of BCB in Whole Blood and Buffered GSH Solution. After 30 sec of incubation in whole blood at 37°C, BCB was completely biotransformed to a single metabolite (fig. 2). The electron impact mass spectrum of the metabolite showed a molecular ion at m/z 106 (fig. 3). The HPLC retention time and mass spectra of the authentic standard were identical to those of the metabolite isolated from blood and buffered GSH solution, which was positively identified as MGN. The conversion of BCB to MGN in the whole blood and buffered GSH solutions was blocked by coincubation with NEM (40 mM).

In Vitro BCB and MGN Binding Studies. The percentage of binding for BCB and MGN incubated with whole blood, erythrocytes, and hemoglobin is shown in table 2. As the incubation time was increased, the amounts of BCB and MGN bound to whole blood and erythrocytes increased progressively. Coincubation of whole blood and erythrocyte preparations with NEM resulted in significantly decreased binding ratios for both BCB and MGN (fig. 5). Both BCB and MGN showed covalent binding (approximately 30%) to purified hemoglobin. As in the other preparations, binding to hemoglobin could be inhibited with NEM.

In Vitro BCB and MGN Erythrocyte Cytotoxicity. Dose-related hemolysis of erythrocytes was observed when BCB (0.19-3 mM) was incubated with purified erythrocytes (EC₅₀ = 1.0 ± 0.3 mM). The addition of NEM to the incubation mixture caused increased BCB-induced hemolysis (EC₅₀ = 0.3 ± 0.2 mM). In contrast, hemolysis of erythrocytes was greatly decreased when GSH (1.5 mM) was added to the samples. The EC₅₀ of BCB-induced hemolysis of erythrocytes incubated with GSH was >3 mM, which was significantly higher than that for BCB incubated with erythrocytes alone (fig. 6). However, MGN at concentrations as high as 30 mM did not induce erythrocyte hemolysis.

Spectral Effects of BCB and MGN on Purified Hemoglobin. Purified hemoglobin from isolated rat erythrocytes showed a typical hemoglobin spectrum, with _max values of 540 and 574 nm (fig. 7). However, the intensity of the absorbance spectrum declined with increasing concentrations of BCB (0.75-3.1 mM). With MGN, there was no alteration of the hemoglobin spectrum at any of the concentrations tested.

	Discussion

Top Abstract Introduction Materials Results Discussion References

After iv administration, BCB is labile, being rapidly debrominated to form MGN. A progressively larger amount of ¹⁴C-equivalents become bound to blood constituents. Furthermore, approximately 80% of the bound ¹⁴C-equivalents were associated with the erythrocyte fraction of the blood and could not be displaced by extraction with various solvents. These initial observations raise two questions, as follows: what is the mechanism of the rapid conversion of BCB to MGN and what is the chemical entity (BCB or its metabolites) that binds to erythrocytes? To answer these questions, further studies focused on characterizing the biotransformation of BCB and its ability to covalently bind to macromolecules.

Sauer et al. (1998) showed that whole blood debrominates BCB. This rapid conversion by whole blood is likely responsible for the absence of detectable BCB after iv administration. Furthermore, it was speculated that this conversion is mediated by free sulfhydryls, analogously to a metabolic scheme described by Livesey et al. (1982). To further characterize BCB biotransformation, BCB was incubated with plasma, erythrocytes, or whole blood obtained from male Fischer 344 rats. The results showed that BCB was converted to MGN by erythrocytes and whole blood in <30 sec. However, conversion by plasma was much slower than that by erythrocytes or whole blood. Erythrocytes, as well as plasma, contain high concentrations of both protein and nonprotein sulfhydryl groups, which can interact with xenobiotics (erythrocytes  plasma) (Tirumalai et al., 1996; Connor and Schroit, 1990). To investigate the possible role of sulfhydryl groups in the conversion of BCB to MGN, buffered GSH solutions were incubated with BCB. As in incubations with blood, BCB was readily debrominated and converted to MGN. Furthermore, the addition of NEM, a sulfhydryl-alkylating agent, to whole blood and GSH solutions blocked the conversion of BCB to MGN. The reactivity of MGN toward sulfhydryl groups is further illustrated by the urinary metabolites of BCB. The primary urinary metabolite of BCB results from the conjugation of MGN with GSH and is a mercapturic acid conjugate of MGN [N-acetyl-S-(2,4-dicyanobutane)-L-cysteine] (Sauer et al., 1998). Thus, the presence of sulfhydryl-containing molecules is required for the conversion of BCB to MGN, as well as the further metabolism of MGN.

The role of sulfhydryls in the biotransformation of BCB was further elucidated when it was determined that two molecules of GSH were required to convert one molecule of BCB to MGN. During the reaction, the two GSH molecules are oxidized to form GSSG. This evidence strongly suggests that there is a quantitative relationship between the conversion of BCB to MGN and the formation of GSSG from GSH. Indeed, as speculated by Sauer et al. (1998), the debromination of BCB is mediated by a free sulfhydryl-dependent biotransformation pathway. The nucleophilic attack by GSH does not result in the formation of a stable GSH conjugate. Instead, two GSH molecules are oxidized to GSSG. This reaction mechanism is similar to that previously described by Livesey et al. (1982), in which the nucleophilic attack of GSH on a dihaloalkyl substrate results in S-(-haloalkyl)-GSH formation; subsequent attack of a second thiol on the sulfur atom of the conjugate yields GSSG, two halide ions, and an alkene product.

Incubation of BCB or MGN with whole blood or isolated erythrocytes resulted in progressive binding, in which 75-90% of the ¹⁴C-equivalents were bound after 8 hr of incubation. BCB and MGN showed nearly identical kinetic binding profiles. These data, along with the biotransformation data discussed above, suggest that MGN is responsible for the observed macromolecular binding. In addition, there may be specific binding sites or targets for MGN in whole blood and on erythrocyte membranes. NEM, which has been shown to prevent the conversion of BCB to MGN, also reduces the amount of macromolecular binding of both BCB and MGN in whole blood and erythrocyte incubations. Thus, in the case of BCB, it is clear that NEM inhibits biotransformation and only a limited amount of binding occurs. However, for MGN, sulfhydryls are not involved in further metabolism but NEM inhibits binding apparently by competing for common binding sites. Therefore, we conclude that MGN is the reactive species responsible for macromolecular binding and that free sulfhydryl groups are the targets for MGN binding.

Another significant observation was that BCB incubation with erythrocytes resulted in hemolysis, in a dose-related fashion. After 2 hr of incubation, the cytotoxicity curve was fitted to a sigmoidal profile and an EC₅₀ of 1.0 ± 0.3 mM was calculated. Additionally, this dose-response relationship could be altered by coincubation with NEM. Although a 0.5 mM concentration of NEM did not cause hemolysis during the 2-hr incubation period, a synergistic hemolytic response was observed with NEM in combination with BCB. In contrast, coincubation with GSH protected erythrocytes from BCB-induced hemolysis. As previously discussed, the debromination of BCB is mediated by the oxidation and consumption of free sulfhydryls, including GSH. Other xenobiotics that interact with sulfhydryl groups can cause significant physiological or biochemical changes in erythrocytes. For example, several investigators have reported that the oxidation of membrane-associated sulfhydryl groups by diamide and 2,3-diphosphoglycerate results in erythrocyte membrane rigidity and enhanced permeability to hydrophilic nonelectrolytes and ions (Klonk and Deuticke, 1992; Chasis and Mohandas, 1986). These changes were specific to the oxidative modifications of sulfhydryl groups and were reversible upon reduction back to the free sulfhydryls. Lauriault and O'Brien (1991) also reported that disulfiram-induced hemolysis was markedly enhanced in GSH-depleted erythrocytes. This rapid hemolysis involves the oxidation of plasma membrane-associated protein sulfhydryl groups in lysis. It is of interest to note that MGN does not cause hemolysis of erythrocytes even at a concentration that is 10-fold higher than the calculated EC₅₀ for BCB. Thus, it appears that BCB-induced hemolysis results from the oxidation of sulfhydryls, as well as the consumption of GSH, and not from macromolecular binding (fig. 8). However, it is important to note that hemolysis was not observed in the in vivo studies at the doses used in previous studies (data not shown). The explanation for this difference likely involves the concentrations of GSH available to the erythrocytes in vivo, compared with the limited amounts present in in vitro incubations, which can be readily consumed during the conversion of BCB to MGN.

Both BCB and MGN covalently bind to purified hemoglobin, at approximately the same levels (30%). However, unlike the other in vitro preparations, hemoglobin does not catalyze the conversion of BCB to MGN (data not shown). In this case, it appears that BCB binding is not mediated through MGN but, instead, BCB itself binds hemoglobin. Although the binding associated with both of these compounds could be reduced with NEM, there appear to be significant differences in the binding of BCB and MGN to hemoglobin. This difference is evident in a comparison of the abilities of BCB and MGN to alter the spectrum of hemoglobin. With increasing concentrations of BCB, the absorbance spectrum for hemoglobin was gradually reduced, whereas there were no spectral changes after incubation with MGN at any of the concentrations tested.

In summary, BCB is rapidly debrominated to form MGN in the blood, via a free sulfhydryl-dependent biotransformation pathway. This metabolic route results in the consumption of two GSH molecules for each molecule of BCB converted to MGN. This free sulfhydryl-exhausting pathway also results in the loss of erythrocyte viability, mediated by oxidation of free sulfhydryls in the erythrocyte membrane. BCB and its metabolite MGN appear to be the reactive species responsible for macromolecular binding. Although systemic exposure of rats to [¹⁴C]BCB resulted in significant binding of ¹⁴C-equivalents to erythrocytes and other blood constituents, there was no overt in vivo cytotoxicity observed with this acute exposure. This may be the result of the immediate biotransformation of the limited amount of BCB in the systemic environment. However, the oxidative stress induced by BCB in erythrocytes cannot be considered unimportant, when it is viewed as an initiating event in a toxic interaction with another toxicant, which also consumes sulfhydryls or requires GSH for detoxification. Furthermore, BCB acting as a systemic sensitizing agent, thus mediating a hypersensitivity response, remains a possible mechanism of toxicity and warrants further study.