Stability of Hemoglobin and Albumin Adducts of Benzene Oxide and 1,4-Benzoquinone after Administration of Benzene to F344 Rats

Melissa A. Troester*, Andrew B. Lindstrom{dagger}, Lawrence L. Kupper{ddagger}, Suramya Waidyanatha* and Stephen M. Rappaport*,1

* Department of Environmental Sciences and Engineering, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599–7400; {dagger} National Exposure Research Laboratory, United States Environmental Protection Agency, Research Triangle Park, North Carolina 27711; and {ddagger} Department of Biostatistics, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599–7400

Received June 16, 1999; accepted November 13, 1999


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The stability of cysteinyl adducts of benzene oxide (BO) and mono-S-substituted cysteinyl adducts of 1,4-benzoquinone (1,4-BQ) was investigated in both hemoglobin (Hb) and albumin (Alb) following administration of a single oral dose of 400 mg [U-14C/13C6]benzene/kg body weight to F344 rats. Total radiobound adducts to Hb were stable, as were adducts formed by the reaction of [13C6]BO with cysteinyl residues on Hb. In both cases adduct stability was indicated by zero-order kinetics with decay rates consistent with the lifetime of rat erythrocytes. Hb adducts of 1,4-BQ were not detected, possibly due to the production of multi-S-substituted adducts within the erythrocyte. Regarding Alb binding, total radiobound adducts decayed more rapidly than expected (half-life of 0.4 days), suggesting that uncharacterized benzene metabolites were noncovalently bound or formed unstable adducts with Alb. Although adducts from reactions of BO and 1,4-BQ with Alb both decayed with rates consistent with those of Alb turnover in the rat, the half-life for 1,4-BQ-Alb (2.5 days) was shorter than that for BO-Alb (3.1 days), suggesting some instability of 1,4-BQ-Alb. Assuming similar rates of adduct instability in humans and rats, the 1,4-BQ-Alb adducts would be eliminated with a half-life of approximately 8 days, compared with BO-Alb, which would be expected to turnover with Alb (half-life of approximately 21 days).

Key Words: adduct stability; albumin, benzene; benzene oxide; benzoquinone; biomarker; hemoglobin; protein adducts; protein turnover.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Benzene, a ubiquitous environmental contaminant (Wallace, 1996Go), is a product of combustion, a constituent of petroleum, and an important industrial chemical. Benzene is carcinogenic in animals and humans (Snyder and Kalf, 1994Go), and strong evidence exists that it exerts toxicity through its metabolic intermediates (Smith, 1996Go; Snyder and Hedli, 1996Go). Initial metabolism of benzene involves P450 2E1, NADP(H)-dependent oxidation of benzene to benzene oxide (BO) (Kalf, 1987Go; Lindstrom et al., 1997Go; Snyder and Kalf, 1993). Subsequent transformation of BO yields phenol and polyhydroxylated metabolites, notably catechol (CAT), hydroquinone (HQ), and 1,2,4-trihydroxybenzene. These phenolic metabolites are further oxidized to 1,2-benzoquinone (1,2-BQ), 1,4-benzoquinone (1,4-BQ), and 2-hydroxy-1,4-benzoquinone via the corresponding semiquinones. It is thought that electrophilic metabolites of benzene induce toxicity through either covalent binding to cellular macromolecules (Snyder et al., 1996) or oxidative damage (Hiraku and Kawanishi, 1996Go; Lewis et al., 1988Go). Due to the reactivity of benzene's electrophilic metabolites, recent studies have focused upon protein adducts as biomarkers of exposure (Bechtold et al., 1992aGo, 1992bGo; Lindstrom et al., 1998Go; McDonald et al., 1993Go; McDonald et al., 1994Go; Melikian et al., 1992Go; Waidyanatha et al., 1998Go; Yeowell-O'Connell et al. 1996Go, Yeowell-O'Connell et al., 1998Go).

For purposes of biomonitoring and dosimetry, the ideal protein adduct is chemically stable under biologic conditions, does not influence the stability of the protein, is easily accessible for epidemiologic studies, and captures a significant fraction of exposure history (Skipper and Tannenbaum, 1990Go). Hemoglobin (Hb) and albumin (Alb) are abundant and easily obtained from blood, and are relatively long lived in humans (Hb lifetime ~120 days, Alb half-life ~21 days). Previous studies have established that BO, 1,2-BQ, and 1,4-BQ form cysteinyl adducts with Hb and Alb in a dose-dependent fashion following metabolism of benzene in vivo (Lindstrom et al., 1998Go; McDonald et al., 1993Go; McDonald et al., 1994Go; Waidyanatha et al., 1998Go; Yeowell-O'Connell et al., 1996Go). Alb and Hb adducts of BO (BO-Alb and BO-Hb) have also been detected in workers exposed to high concentrations of benzene (Yeowell-O'Connell et al., 1998Go). However, the chemical stability of adducted benzene metabolites and the influence of these adducts on protein stability have not been reported.

The impact of adduct stability on quantitative relationships between adduct levels and exposure has been described for both Hb and Alb (Granath et al., 1992Go). The constant, predetermined life span of the red blood cell (ter), approximately 60 days in the rat or 120 days in humans (Allison, 1960Go; Schalm et al., 1975Go), leads to zero-order kinetics of adduct removal if Hb adducts are stable. If first-order or mixed zero- and first-order kinetics are observed for removal of Hb adducts, it can be concluded that adducts are chemically unstable or that adducted erythrocytes are removed more rapidly than normal red cells (Fennell et al., 1992Go). Alb, conversely, is subject to first-order kinetics of turnover, with a half-life of approximately 2.5 to 3 days in rats and approximately 21 days in humans (Allison, 1960Go; Peters, 1970Go). Alb adduct instability is manifested by first-order kinetics, with a half-life shorter than that observed for normal Alb (Granath et al., 1992Go). Selection of an appropriate mathematical model for estimating the dose of a benzene metabolite from levels of its adducts depends in part upon the stability characteristics of these adducts.

Because Alb and Hb are found in different physiologic compartments, their particular biochemical environments can affect the formation and stability of adducts. To react with Hb, an electrophile must be sufficiently stable to diffuse out of the metabolizing cell (e.g., the hepatocyte) and into the erythrocyte. Within the red cell, high levels of glutathione (>2 mM) (Mori et al., 1990Go; Srivastava, 1971Go) can compete with proteins for reaction with electrophiles. Sulfhydryl groups from glutathione and proteins can also form mixed disulfide bonds that tend to stabilize Hb and protect reactive amino acids from covalent modification (Srivastava, 1971Go). For electrophiles with more than one reactive site, such as 1,4-BQ, glutathione can also increase the proportion of multi-S-substituted adducts formed (Eckert et al., 1990Go; Lau et al., 1988Go). Because Alb is present in plasma with much lower glutathione concentrations (1–10 µM) (Smith et al., 1996Go) and is not protected by the erythrocyte membrane, it would be expected to form adducts more efficiently than Hb. Furthermore, Alb is synthesized within hepatocytes, thereby facilitating reactions with electrophiles that are too reactive to escape the metabolizing cells.

Given that adducts may exhibit differential stability depending on both the electrophilic species and the protein, the stability of each adduct must be evaluated to ensure proper application of dosimetric models. As stability of Hb and Alb adducts arising from benzene metabolism have not been reported, we investigated the time course of such adducts for 21 days after administration of a single, 400 mg/kg oral dose of [U-14C/13C6]benzene to F344 rats. Radiobinding was used to measure adducts from all benzene metabolites; a gas chromatography-mass spectrometry (GC-MS) assay (Waidyanatha et al., 1998Go) was used to measure cysteinyl adducts formed by the reactive metabolites BO, 1,4-BQ, and 1,2-BQ.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Chemicals.
Ascorbic acid and hydrochloric acid (conc.) were from Fisher Scientific (Pittsburgh, PA). Acetone and hexane were obtained from Mallinckrodt (Paris, KY). [U-14C]Benzene (97% radiochemical purity, specific activity = 124 mCi/mmol, representing one 14C atom per benzene ring) was obtained from Amersham Pharmacia Biotech (Buckinghamshire, England). [13C6]Benzene (99% chemical purity) was obtained from Cambridge Isotope Laboratories (Andover, MA). Methoxyflurane was obtained from Pitman-Moore (Mundelein, IL). Trifluoroacetic anhydride (TFAA) was purchased from Pierce (Rockford, IL) and was distilled once before use. Methanesulfonic acid was purchased from Fluka Chemical Company (Switzerland). Human Hb and Alb for standard curves were purchased from Sigma (St. Louis, MO). S-Phenylcysteine (SPC) and [2H5]SPC were kindly provided by Drs. A. Gold and R. Sangaiah of the University of North Carolina, Chapel Hill. 1,4-BQ-bound N-acetyl-L-cysteine (1,4-BQ-NAC) and deuterium-labeled cysteinyl adducts of 1,4-BQ on Hb and Alb ([2H3]1,4-BQ-Hb and [2H3]1,4-BQAlb) were previously prepared using materials and methods described in Waidyanatha et al. (1998).

Caution:
TFAA reacts violently with water and should only be used to derivatize proteins that are completely dry. Benzene is a carcinogen and [U-14C]benzene is a radioactive carcinogen; both should be handled with caution. All of the procedures below involving volatile compounds should be conducted in a certified laboratory fume hood.

Animals and blood collection.
Forty male F344 rats (175–200 g) were obtained from Charles River Breeding Laboratories (Raleigh, NC) and were housed in polycarbonate cages on a 12-h light/dark cycle for at least 2 weeks before use. Food and water were provided ad libitum. Dosing solutions were prepared so that 1 ml of corn oil solution would deliver, by gavage, 400 mg [U-14C/13C6]benzene/kg body weight and 50 µCi per animal. The activity of the dosing solutions was measured by scintillation counting prior to administration. Four rats received corn oil only (vehicle control) and were sacrificed after 24 h. The remaining 36 rats were divided into groups of four and received a single oral dose of 400 mg/kg body weight [U-14C/13C6]benzene in corn oil. One group was sacrificed at each of the following time points: 4 h, 8 h, 12 h, 16 h, 1 day, 2 days, 7 days, 14 days, and 21 days. Rats were anesthetized with methoxyflurane and blood was removed by direct cardiac puncture into a heparinized syringe. Approximately 6–8 ml of blood were collected from each animal. Blood was stored immediately on ice and plasma and red cell fractions were separated within 2 h.

Isolation and purification of Hb and Alb.
Blood samples were centrifuged at 800 x g for 15 min and the plasma layer was removed by pipet. The red blood cell layer was then washed three times with an equal volume of 0.9% saline. The buffy coat was removed by pipette and an equal volume of deionized water was added to the red blood cells. Samples were frozen at –20°C overnight to allow cell lysis prior to isolation of Hb.

Hb and Alb were isolated according to the procedure described in Rappaport et al. (1993), with some modifications. Briefly, Hb was obtained from thawed, lysed red cells by centrifuging at 30,000 x g at 4°C for 40 min to remove cell membranes and purifying the supernatant by exhaustive dialysis (Spectra-Pore 1, 6000–8000 MWCO) against 4 x 3.5 liters of deionized water at 4°C over 24 h (rather than by Sephadex chromatography). Globin was precipitated by adding the hemolysate to cold acidified acetone (0.1% HCl by volume), washing with ice-cold acetone, and drying to constant weight under vacuum at 37°C.

Alb was isolated from thawed pure plasma by dropwise addition of an equal volume of saturated (NH4)2SO4. This mixture was centrifuged at 900 x g to remove immunoglobulins. The supernatant was dialyzed (Spectra-Pore 1, 12,000 MWCO) against 4 x 3.5 liters of deionized water at 4°C over 24 h, and lyophilized to constant weight.

Measurement of BO and 1,4-BQ adducts with Hb and Alb.
Cysteinyl adducts of BO, 1,2-BQ, and 1,4-BQ with Hb and Alb were assayed with the method of Waidyanatha et al. (1998).2 Briefly, 200 µl of a 50 mg/ml solution of Hb or 100 µl of a 50 mg/ml solution of Alb were added to 200 µl of 100 mM ascorbic acid, and isotopic internal standards were added in the following quantities: 10 pmol [2H5]-SPC, 10 µg of [2H3]1,4-BQ-Alb, or 10 µg of [2H3]1,4-BQ-Hb. Samples were then dried in a vacuum oven at 80°C overnight. To the dried proteins, 750 µl of TFAA and 20 µl of methanesulfonic acid were added and the proteins were incubated at 100°C for 40 min to produce phenyltrifluorothioacetate (PTTA) and O,O',S-tris-trifluoroacetyl-hydroquinone (HQ-S-TFA). The samples were then cooled to room temperature and excess TFAA was removed under a stream of nitrogen. To each sample, 1 ml of hexane and 1 ml of 0.1 M Tris buffer (pH 7.5) were added; the mixture was vortexed for 30 s and then centrifuged. The hexane layer was removed and washed twice with 1 ml of deionized water. Samples were then concentrated to 200 µl and transferred to vials for analysis. Standard curves for BO-Hb and BO-Alb were prepared by adding a range of amounts of SPC to 10-mg aliquots of untreated Hb and Alb (Sigma), which were assayed as described above for the experimental samples. Standard curves for 1,4-BQ adducted to Hb and Alb were prepared by adding a range of amounts of 1,4-BQ-NAC to 1-mg aliquots of untreated Hb and Alb (Sigma) and performing the assay as described above for experimental samples.

GC/NICI-MS analysis.
All GC-MS analyses were conducted in negative-ion chemical ionization (NICI) mode with a Hewlett-Packard 5890 series II plus gas chromatograph equipped with a Hewlett-Packard 5989A MS engine using a DB-5 capillary column (60 m, 0.25 mm id, 0.25 µm phase thickness; J & W Scientific, Inc.); 3-µl samples were injected in the splitless mode with a He carrier gas flow of 1.5 ml/min. Methane at a source pressure of 2 Torr was used as the chemical ionization reagent gas.

Analysis of PTTA.
The injection port and source temperatures were 250 and 100°C, respectively. The oven temperature was held for 3 min at 50°C, then increased at 2.4°C/min to 108°C, then increased at 50°C/min to 250°C, where it was held for 10 min. The molecular ions of PTTA (m/z 206), [2H5]PTTA (m/z 211), and [13C6]PTTA (m/z 212) were monitored using the selected ion monitoring mode.

Analysis of HQ-S-TFA.
The injection port and source temperatures were 250 and 150°C, respectively. The oven temperature was held for 3 min at 75°C, then increased at 8°C/min to 98°C, where it was held for 25 min, then increased at 50°C/min to 250°C, where it was held for 10 min. The molecular ions of HQ-S-TFA (m/z 333) and [2H3]HQ-S-TFA (m/z 336) and [13C6]HQ-S-TFA (m/z 339) were monitored using the selected ion monitoring mode.

Radiobinding.
Small aliquots of [U-14C/13C6]benzene dosing solutions or purified proteins (5 mg Alb or Hb) from [U-14C/13C6]benzene-dosed rats were added to 10 ml of Econoscint scintillant (Fisher Scientific, Pittsburgh, PA) and counted on a Wallace 1409 liquid scintillation analyzer for 5 min. Quenching was not observed and bleaching was not needed.

Estimation of elimination rates.
Adduct elimination rates were estimated from relationships developed by Granath et al. (1992). Briefly, we define A = [RY]/[Y] (in units of nmol/g) as the level of adduct RY per gram of protein (i.e., BO-Y or 1,4-BQ-Y for Y = Hb or Alb) at some time t (d) after acute exposure to benzene. Stable adducts of Hb would be eliminated by zero-order kinetics according to the relationship


where A0 represents the initial adduct level (assuming instantaneous production of adducts at time t = 0) and ter is the erythrocyte lifetime (d). If Hb adducts were stable, then from equation 1, least-squares regression of A(t) on t gives a linear relationship with estimated intercept Â0 and slope . er can then be calculated according to the following relationship (Granath et al., 1992Go):

For unstable adducts, mixed zero- and first-order kinetics of adduct turnover result in departures from the linear relationship between A(t) and t.

Adducts of Alb would be eliminated by first-order kinetics as indicated by


where kalb (d–1) is the first-order rate constant for Alb turnover and k (d–1) is the first-order rate constant for adduct instability. Then, from equation 3, the relationship between ln[A(t)] and t is linear with intercept ln(A0) and slope ß = –(kalb + k). Adduct stability can be evaluated by comparing estimated values of (kalb + k) obtained by least squares regression to literature values for kalb, as k = 0 for stable adducts.

Statistical Analysis
All statistical analyses were conducted using SAS Statistical Software (Cary, NC). Data obtained prior to peak adduct levels (i.e., before 1 day for [13C6]BO-Hb, etc.) were excluded from the regression analysis because they were not relevant to the determination of adduct stability. Linear regressions of A(t) and ln[A(t)] on time (d) were conducted for Hb and Alb data, respectively. With Hb, it was not known a priori whether decay would follow zero-order kinetics, so lack-of-fit tests were conducted for the regressions of A(t) on time to verify that a higher-order regression model was unnecessary (Kleinbaum et al., 1998Go).

Standard errors were estimated for all parameter estimates. The standard errors (SE) associated with er estimates were estimated from the following relationship:



where denotes the estimated variances of, and Côv denotes the estimated covariance between the regression parameter estimates Â/0 and

A Kruskall-Wallis test (a nonparametric analog of one-way ANOVA) was conducted to evaluate whether percentages of [13C6]BO-Hb adducts to total radiobound Hb ([14C]R-Hb) adducts varied significantly with time. Statistically significant differences in these percentages over time would suggest that [13C6]BO-Hb and [14C]R-Hb were decaying at different rates. Spearman correlation coefficients were used to test whether the percentages of [13C6]BO-Alb and [13C6]1,4-BQ-Alb to [14C]R-Alb increased as a function of time. A significantly positive estimated correlation coefficient for the ratio of a specific adduct ([13C6]BO-Alb and [13C6]1,4-BQ-Alb) to total radiobound adducts provides evidence that the total radiobound adducts were decaying more rapidly than the specific adduct.

Confidence interval for the difference between the decay rates of BO-Alb and 1,4-BQ-Alb.
Whether the estimated decay rates for [13C6]BO-Alb and [13C6]1,4-BQ-Alb are statistically different cannot be readily determined by comparing their confidence intervals (Greenland, 1998Go). To best determine an appropriate 95% confidence interval for the difference between the first-order decay rates for BO-Alb and 1,4-BQ-Alb, multiple regression analysis was conducted using the following model:


where Y = ln[A(t)], t is time in days, and Z is a dummy variable defined as 1 for 1,4-BQ-Alb adducts and 0 for BO-Alb adducts. When Z = 0, the equation for the regression of ln(BO-Alb) on days is


When Z = 1, the equation for the regression of ln(1,4-BQ-Alb) on days is


Thus, the regression equations for first-order decay of [13C6]BO-Alb and [13C6]1,4-BQ-Alb are both incorporated into a single model, allowing for different slopes and intercepts. The estimated difference between the decay rates of [13C6]BO-Alb and [13C6]1,4-BQ-Alb is given by 3. Again, only the [13C6]BO-Alb and [13C6]1,4-BQ-Alb data that were relevant to stability (8 h–14 days for BO-Alb and 1 day–14 days for 1,4-BQ-Alb) were included in the analysis.

Model (5) would not be appropriate for use if the adduct levels being compared ([13C6]BO-Alb and [13C6]1,4-BQ-Alb) were significantly correlated. To test whether there was significant correlation between levels of [13C6]BO-Alb and [13C6]1,4-BQ-Alb within animals at a given time point, estimated Spearman correlation coefficients were computed for each group of animals. No significant correlation was observed, so the analysis as described is appropriate.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Hb Adducts
The kinetics of disappearance of adducts from Hb of F344 rats dosed with 400 mg/kg [U-14C/13C6]benzene were based on total radiobinding ([14C]R-Hb) and measurement of the specific adduct ([13C6]BO-Hb). Levels of [13C6]1,4-BQ-Hb were below the detection limit (20 pmol adduct/g Hb) of the assay. However, [13C6]BO-Hb was readily detected, as shown in Figure 1Go. Adduct levels were highest at 1 ([13C6]BO-Hb) or 2 days ([14C]R-Hb) after dosing and then decreased according to zero-order kinetics. From the least-squares regressions of A(t) on time (d), the estimated parameters in Table 1Go were obtained. Regressions of [13C6]BO-Hb or [14C]R-Hb on time yielded significant slope estimates (p = 0.005 and p = 0.001, respectively), and there was no evidence for lack of fit (p > 0.10) in either case. The 95% confidence intervals for ter were 32.0–68.4 days and 33.1–93.5 days, based upon radiobinding and [13C6]BO-Hb, respectively. The percentages of [13C6]BO-Hb to [14C]R-Hb, shown in Table 2Go, were not significantly different among the groups of animals sacrificed at different times (p = 0.530).



View larger version (13K):
[in this window]
[in a new window]
 
FIG. 1. [14C]R-Hb (square) and BO-Hb (circle) in male F344 rats dosed with 400 mg/kg body weight [U-14C/13C6]benzene. Each point represents the mean (± SE) for four animals. Only data that were relevant to the stability estimates are shown. Lines represent the equations obtained from the least squares regression of adducts on time.

 

View this table:
[in this window]
[in a new window]
 
TABLE 1 Estimated Parameters for Regression of Hemoglobin Adduct Level on Time
 

View this table:
[in this window]
[in a new window]
 
TABLE 2 Percentages of Total Hemoglobin ([14C]R-Hb) Adducts Represented by [13C6]BO-Hb after Administration of [U-14C/13C6]Benzene to Male F344 Rats
 
Alb Adducts
Results of the analyses of Alb adducts are illustrated in Figure 2Go as plots of ln[A(t)] on time. In this case, both [13C6]BO-Alb and [13C6]1,4-BQ-Alb were readily quantitated. Measured Alb adducts peaked more rapidly than Hb adducts, with both [14C]R-Alb and [13C6]BO-Alb reaching a maximum at 8 h, and [13C6]1,4-BQ-Alb peaking at 24 h. Note that [14C]R-Alb was below our detection limit (10 nmol adduct/g protein) at day 7 and subsequent time points. From least-squares regression of ln[A(t)] on time, the parameter estimates in Table 3Go were obtained. Figure 2Go demonstrates that the [14C]R-Alb decayed much more rapidly than the metabolite-specific adducts. This is reflected by the percentages of [13C6]BO-Alb and [13C6]1,4-BQ-Alb to [14C]R-Alb increasing as a function of time (Table 4Go), with Spearman correlation coefficients of 0.966 (p = 0.001) and 0.895 (p = 0.001), respectively.



View larger version (12K):
[in this window]
[in a new window]
 
FIG. 2. [14C]R-Alb (open circle), [13C6]1,4-BQ-Alb (square) and [13C6]BO-Alb (filled circle) after dosing male F344 rats with 400 mg/kg body weight [U-14C/13C6]benzene. Each point represents the mean (± SE) for four animals. [14C]R-Alb is only shown until day 2 because radioactivity was at background level at subsequent time points. Only data that were relevant to the stability estimates are shown. Lines represent the equations obtained from the least squares regression of ln(adducts) on time.

 

View this table:
[in this window]
[in a new window]
 
TABLE 3 Estimated Parameters for Regression of the Natural Log of Albumin Adduct Level on Time
 

View this table:
[in this window]
[in a new window]
 
TABLE 4 Percentages of Total Albumin ([14C]R-Alb) Adducts Represented by [13C6]BO-Alb and [13C6]1,4-BQ-Alb after Administration of [U-14C/13C6]Benzene to Male F344 Rats
 
The estimated values of equal to –0.225 d–1 for [13C6]BO-Alb (half-life = 3.08 days) and –0.278 d–1 for [13C6]1,4-BQ-Alb (half-life = 2.49 days) have relatively narrow 95% confidence intervals (–0.249 to –0.201 d–1 and –0.31 to –0.245 d–1, respectively). The estimated difference between the first-order decay rates of BO-Alb and 1,4-BQ-Alb was –0.053 d–1, with 95% confidence interval of –0.093 to –0.014 d–1. Assuming that BO-Alb was stable and its adduct decay rate (k) was zero, then the rate of Alb turnover (kalb) was reflected by loss of [13C6]BO-Alb (i.e., alb = 0.225 d–1). Thus, the adduct decay rate of [13C6]1,4-BQ-Alb (k) can be estimated as 0.278 = (0.225 +), where & = 0.053 d–1.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Results of this study show that [14C]R-Hb and [13C6]BO-Hb are reasonably stable. The elimination of [14C]R-Hb and [13C6]BO-Hb from the blood of F344 rats followed zero-order kinetics (Fig. 1Go), with lifetimes for radiobinding (47 days) and [13C6]BO-Hb (56 days) that were similar to reported values for the lifetime of the rat erythrocyte, (45 < ter < 65 d) (Allison, 1960Go; Schalm et al., 1975Go). In addition, the percentage of [13C6]BO-Hb to total adducts was constant over time, at about 35% (Table 2Go), indicating that the kinetic behavior of [13C6]BO-Hb paralleled that of the total (uncharacterized) Hb adducts. The percentage of [13C6]BO-Hb to total Hb adducts is consistent with previous studies where rats received a single dose of benzene and BO-Hb contributed 22–32% of total binding (Bechtold et al., 1992aGo; McDonald et al., 1994Go; Melikian et al, 1992Go; Yeowell-O'Connell et al., 1996Go).

Our inability to measure mono-S-substituted BQ adducts of Hb (1,4-BQ-Hb) is also consistent with previous work from our laboratory (Waidyanatha et al., 1998Go). We suspect that the absence of 1,4-BQ-Hb is a result of the high reactivity of BQs, particularly with sulfhydryl groups. We speculate that the BQs either formed adducts with plasma proteins prior to entering the erythrocyte, or alternatively, that the mono-S-substituted quinone adducts were themselves further modified to multi-S-substituted products by reactions with glutathione and other nucleophiles in the erythrocyte (Eckert et al., 1990Go; Lau et al., 1988Go; Waidyanatha et al., 1998Go). As two previous studies indicated that 1,4-BQ did react with Hb in vivo, the latter explanation seems more likely. [Using a Raney nickel assay to measure the sum of mono- and multi-S-substituted BQ adducts, McDonald et al. (1994) found low levels of [13C6]1,4-BQ-Hb in rats 24 h after administration of 200 and 400 mg [U-14C/13C6] benzene/kg. At higher doses of benzene (800 mg/kg body weight), Melikian et al. (1992) attributed 37% of total binding to mono-S-substituted 1,4-BQ-Hb in F344 rats.]

In contrast to the radiobound Hb adducts of benzene metabolites, the uncharacterized Alb adducts were quite unstable. The half-life for first-order decay of [14C]R-Alb (0.4 days) was markedly more rapid than the 2.5- to 3-day half-life of Alb reported in the literature (Allison, 1960Go; Peters, 1970Go). In fact, the decay of [14C]R-Alb in Figure 2Go appears to follow biphasic kinetics, with rapidly decaying adducts or noncovalently bound species dominating the early portion of the curve, and more stable adducts influencing the latter part.

The half-lives for the metabolite-specific adducts [13C6]BO-Alb and [13C6]1,4-BQ-Alb (3.1 days and 2.5 days, respectively) both fell within the range of normal values for Alb turnover in the rat (2.5–3 days), though the half-life of [13C6]1,4-BQ-Alb was shorter than that of [13C6]BO-Alb. This suggests some instability of [13C6]1,4-BQ-Alb, perhaps due to the formation of multi-S-substituted adducts of [13C6]1,4-BQ-Alb (Eckert et al., 1990Go; Lau et al., 1988Go; Waidyanatha et al., 1998Go). Still, both [13C6]BO-Alb and [13C6]1,4-BQ-Alb were stable relative to the uncharacterized benzene-derived adducts, as illustrated by the apparent increasing percentage of [13C6]BO-Alb and [13C6]1,4-BQ-Alb with time (Table 4Go).

Our results draw attention to problems with the use of radiobinding to generalize about the stability of individual adducts. The Alb data shown in Figure 2Go point to a clear distinction between the uncharacterized and unstable ([14C]R-Alb) benzene adducts that contributed most of the binding and the more stable metabolite-specific adducts ([13C6]BO-Alb and [13C6]1,4-BQ-Alb), which represented less than 3% and less than 6% of the total binding, respectively.

Previous investigations of protein-adduct decay have shown that the stability of adducts depended upon a number of factors, including the binding species, the amino acid that was modified, and the particular protein. For example, radiobound Hb adducts were relatively stable (zero-order decay kinetics) in rats dosed with certain radioactive chemicals [i.e., butadiene (Sun et al., 1989Go) and fluoranthene (Gorelick et al., 1989Go)], whereas with other chemicals the Hb adducts were unstable (first-order decay kinetics) [i.e., o-toluidine (DeBord et al., 1992Go), 4,4'-methylene-bis(2-chloroaniline) (Cheever et al., 1990Go), naphthalene (Cho et al., 1994Go), and benzo(a)pyrene (Viau et al., 1993Go)]. Furthermore, some of these studies in which Hb adducts were unstable provided evidence of stable Alb adducts [o-toluidine (DeBord et al., 1992Go), naphthalene (Cho et al., 1994Go) and benzo(a)pyrene (Viau et al., 1993Go)]. Finally, in mice dosed with styrene, Hb adducts of styrene-7,8-oxide bound to N-terminal valine were observed to be stable, while those bound to carboxylic acids were unstable (Osterman-Golkar et al., 1995Go).

Our results suggest that cysteinyl adducts of BO with both Hb and Alb are stable in vivo and suitable for biomonitoring. However, the adduct of [13C6]1,4-BQ with Hb could not be detected and that with Alb indicated moderate instability that will limit its utility for human studies. We estimated the rate of instability of the 1,4-BQ-Alb adduct in the rat to be k = 0.053 d–1. Assuming that this rate is the same in humans in vivo, then the overall rate of adduct loss in the human, where kalb = 0.033 d–1 (21-day half-life), would be (kalb + k) = (0.033 + 0.053) d–1 = 0.086 d–1. Thus, the half-life of 1,4-BQ-Alb in humans would be approximately 8 days compared with 21 days for the stable BO-Alb.


    ACKNOWLEDGMENTS
 
The authors thank Karen Yeowell-O'Connell and Wendy McKelvey for helpful discussions and Patricia Upton for assistance with some of the animal experiments. This work was supported by the National Institute of Environmental Health Sciences through grant P42ES05948 and training grant 5-T32-ES07018. The work described in this document has been funded in part by the U.S. Environmental Protection Agency. It has been subjected to Agency review and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.


    NOTES
 
1 To whom correspondence should be addressed. Fax: (919) 966–4711. E-mail: stephen_rappaport{at}unc.edu. Back

2 Because 1,2-BQ adducts could not be detected in hemoglobin and were close to the detection limit in albumin, the stability of 1,2-BQ adducts could not be studied. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Allison, A. C. (1960). Turnovers of erythrocytes and plasma proteins in mammals. Nature (London) 188, 37–40.

Bechtold, W. E., Sun, J. D., Birnbaum, L. S., Yin, S. N., Li, G. L., Kasicki, S., Lucier, G., and Henderson, R. F. (1992a). S-phenylcysteine formation in hemoglobin as a biological exposure index to benzene. Arch. Toxicol. 66, 303–309.[ISI][Medline]

Bechtold, W. E., Willis, J. K., Sun, J. D., Griffith, W. C., and Reddy, T. V. (1992b). Biological markers of exposure to benzene: S-phenylcysteine in albumin. Carcinogenesis 13, 1217–1220.[Abstract]

Cheever, K. L., Richards, D. E., Weigel, W. W., Begley, K. B., DeBord, D. G., Swearengin, T. F., Savage, R. E. Jr. (1990). 4,4'-Methylene-bis(2-chloroaniline) (MOCA): comparison of macromolecular adduct formation after oral or dermal administration in the rat. Fundam. Appl. Toxicol. 14, 273–283.[ISI][Medline]

Cho, M., Jedrychoswki, R., Hammock, B., and Buckpitt, A. (1994). Reactive naphthalene metabolite binding to hemoglobin and albumin. Fundam. Appl. Toxicol. 22, 26–33.[ISI][Medline]

DeBord, D. G., Swearengin, T. F., Cheever, K. L., Booth-Jones, A. D., and Wissinger, L. A. (1992). Binding characteristics of ortho-toluidine to rat hemoglobin and albumin. Arch. Toxicol. 66, 231–236.[ISI][Medline]

Eckert, K., G., Eyer, P., Sonnenbichler, J., and Zetl, I. (1990). Activation and detoxication of aminophenols. III. Synthesis and structural elucidation of various glutathione addition products to 1,4-benzoquinone. Xenobiotica 20, 351–361.[ISI][Medline]

Fennell, T. R., Sumner, S. C. J., and Walker, V. E. (1992). A model for the formation and removal of hemoglobin adducts. Cancer Epidemiol. Biomarkers Prev. 1, 213–219.[Abstract]

Gorelick, N. J., Hutchins, D. A., Tannenbaum, S. R., and Wogan, G. N. (1989). Formation of DNA and hemoglobin adducts of fluoranthene after single and multiple exposures. Carcinogenesis 10, 1579–1587.[Abstract]

Granath, F., Ehrenberg, L., and Törnqvist, M. (1992). Degree of alkylation of macromolecules in vivo from variable exposure. Mutat. Res. 284, 297–306.[ISI][Medline]

Greenland, S. R. (1998) Analysis of Polytomous Exposures and Outcomes. In Modern Epidemiology (K. J. Rothman and S. R. Greenland, Eds.). 2nd ed., pp. 301–328. Lippincott-Raven, Philadelphia, PA.

Hiraku, Y., and Kawanishi, S. (1996). Oxidative DNA damage and apoptosis induced by benzene metabolites. Cancer Res. 56, 5172–5178.[Abstract]

Kalf, G. F. (1987). Recent advances in the metabolism and toxicity of benzene. CRC Crit. Rev. Toxicol. 18, 141–159.[ISI]

Kleinbaum, D. G., Kupper, L. L., Muller, K. E., and Nizam, A. (1998). Applied Regression Analysis and Multivariate Methods (D. G. Kleinbaum, L. L. Kupper, K. E. Muller, and A. Nizam, Eds.). 3rd ed., pp. 281–316. Duxbury Press, Pacific Grove, CA.

Lau, S. S., Hill, B. A., Highet, R. J., and Monks, T. J. (1988). Sequential oxidation and glutathione addition to 1,4-benzoquinone: correlation of toxicity with increased glutathione substitution. Mol. Pharmacol. 34, 829–836.[Abstract]

Lewis, J. G., Stewart, W., and Adams, D. O. (1988). Role of oxygen radicals in induction of DNA damage by metabolites of benzene. Cancer Res. 48, 4762–4765.[Abstract]

Lindstrom, A. B., Yeowell-O'Connell, K., Waidyanatha, S., Golding, B. T., Tornero-Velez, R., and Rappaport, S. M. (1997). Measurement of benzene oxide in the blood of rats following administration of benzene. Carcinogenesis 18, 1637–1641.[Abstract]

Lindstrom, A. B., Yeowell-O'Connell, K., Waidyanatha, S., McDonald, T. A., Golding, B. T., and Rappaport, S. M. (1998). Formation of hemoglobin and albumin adducts of benzene oxide in mouse, rat, and human blood. Chem. Res. Toxicol. 11, 302–310.[ISI][Medline]

McDonald, T. A., Waidyanatha, S., and Rappaport, S. M. (1993). Production of benzoquinone adducts with hemoglobin and bone-marrow proteins following administration of [13C6]benzene to rats. Carcinogenesis 14, 1921–1925.[Abstract]

McDonald, T. A., Yeowell-O'Connell, K., and Rappaport, S. M. (1994). Comparison of protein adducts of benzene oxide and benzoquinone in the blood and bone marrow of rats and mice exposed to [14C/13C6]benzene. Cancer Res. 54, 4907–4914.[Abstract]

Melikian, A. A., Prahalad, A. K., and Coleman, S. (1992). Isolation and characterization of two benzene-derived hemoglobin adducts in vivo in rats. Cancer Epidemiol. Biomarkers Prev. 1, 307–313.[Abstract]

Mori, K., Inoue, N., Fujishiro, K., Kikuchi, M., and Chiba, S. (1990). Biochemical changes in rat erythrocytes caused by ethylene oxide exposure. Fundam. Appl. Toxicol. 15, 441–447.[ISI][Medline]

Osterman-Golkar, S., Christakopoulos, A., Zorcec, V., and Svensson, K. (1995). Dosimetry of styrene 7,8-oxide in styrene- and styrene oxide-exposed mice and rats by quantification of haemoglobin adducts. Chem. Biol. Interact. 95, 79–87.[ISI][Medline]

Peters, T., Jr. (1970). Serum albumin. Adv. Clin. Chem. 13, 37–111.[Medline]

Rappaport, S. M., Ting, D., Jin, Z., Yeowell-O'Connell, K., Waidyanatha, S., and McDonald, T. (1993) Application of Raney nickel to measure adducts of styrene oxide with hemoglobin and albumin. Chem. Res. Toxicol. 6, 238–244.[ISI][Medline]

Schalm, O. W., Jain, N.C., and Carroll, E. J. (1975). The erythrocytes: their production, function and destruction. In Veterinary Hematology (O. W. Schalm, N. C. Jain, and E. J. Carroll, Eds.). 3rd ed., pp. 356–404. Lea & Febiger, Philadelphia, PA.

Skipper, P. L., and Tannenbaum, S. R. (1990). Protein adducts in the molecular dosimetry of chemical carcinogens. Carcinogenesis 11, 507–518.[Abstract]

Smith, M. T. (1996). The mechanism of benzene-induced leukemia: a hypothesis and speculations on the causes of leukemia. Environ. Health Perspect. 104 (Suppl. 6), 1219–1225.

Smith, C. V., Jones, D. P., Guenthner, T. M., Lash, L. H., and Lauterburg, B. H. (1996). Compartmentation of glutathione: implications for the study of toxicity and disease. Toxicol. Appl. Pharmacol. 140, 1–12.[ISI][Medline]

Snyder, R., Witz, G., and Goldstein, B. D. (1993). The toxicology of benzene. Environ. Health Perspect. 100, 293–306.[ISI][Medline]

Snyder, R., and Kalf, G. (1994). A perspective on benzene leukemogenesis. Crit. Rev. Toxicol. 24, 177–209.[ISI][Medline]

Snyder, R., and Hedli, C. C. (1996). An overview of benzene metabolism. Environ. Health Perspect. 104 (Suppl. 6), 1165–1171.

Srivastava, S. K. (1971). Metabolism of red cell glutathione. Exp. Eye Res. 11, 294–305.[ISI][Medline]

Sun, J. D., Dahl, A. R., Bond, J. A., Birnbaum, L. S., and Henderson, R. F. (1989). Characterization of hemoglobin adduct formation in mice and rats after administration of [14C]butadiene or [14C]isoprene. Toxicol. Appl. Pharmacol. 100, 86–95.[ISI][Medline]

Viau, C., Mercier, M., and Blondin, O. (1993). Measurement of hemoglobin and albumin adducts of benzo(a)pyrendiolepoxide and their rate of elimination in the female Sprague-Dawley rat. Arch. Toxicol. 67, 468–472.[ISI][Medline]

Waidyanatha, S., Yeowell-O'Connell, K., and Rappaport, S. M. (1998). A new assay for albumin and hemoglobin adducts of 1,2- and 1,4-benzoquinones. Chem. Biol. Interact. 115, 117–139.[ISI][Medline]

Wallace, L. (1996). Environmental exposure to benzene: an update. Environ. Health Perspect. 104 (Suppl. 6), 1129–1136.

Yeowell-O'Connell, K., McDonald, T. A., and Rappaport, S. M. (1996). Analysis of hemoglobin adducts of benzene oxide by gas chromatography-mass spectrometry. Anal. Biochem. 237, 49–55.[ISI][Medline]

Yeowell-O'Connell, K., Rothman, N., Smith, M. T., Hayes, R. B., Li, G., Waidyanatha, S., Dosemeci, M., Zhang, L., Yin, S., Titenko-Holland, N., and Rappaport, S. M. (1998). Hemoglobin and albumin adducts of benzene oxide among workers exposed to high levels of benzene. Carcinogenesis 19, 1565–1571.[Abstract]