Although the DNA adducts of vinyl chloride (VC) have been well characterized, previous studies have used single concentrations of VC that are well above contemporary human exposures. This study examined the exposure response to VC in male Sprague Dawley rats with respect to the molecular dose of the promutagenic DNA adduct N2,3-ethenoguanine (N2,3-εG). Adult rats were exposed by inhalation to 0, 10, 100, or 1100 ppm VC for 1 or 4 weeks (6 h/day, 5 days/week). Weanling rats were similarly exposed for 5 days. The amount of N2,3-εG in hepatocyte (HEP) and nonparenchymal cell (NPC) fractions obtained from the liver was measured with a sensitive immunoaffinity/gas chromatography/high-resolution mass spectrometry assay. Endogenous N2,3-εG was present in HEPs and NPCs from all unexposed rats. The exposure response to VC in each group and cell population was supralinear, with a linear increase from 0 to 100 ppm, and a plateau between 100 and 1100 ppm. There was no statistically significant difference in N2,3-εG concentrations between HEPs and NPCs in any adult exposure group, which suggests that factors other than adduct concentrations contribute to the particular susceptibility of NPCs to VC-induced carcinogenesis. The accumulation of N2,3-εG with respect to time was nearly linear in rats exposed to 600 ppm VC for 1, 2, 4, or 8 weeks (4 h/day, 5 days/week), and no repair of N2,3-εG was detected in rats exposed to VC for 4 weeks and allowed to recover for 1 week. N2,3-εG concentrations in HEPs from weanling rats were 2–3-fold greater than those in adult rats exposed for the same time. Higher adduct concentrations in young rats may contribute to their greater susceptibility to VC-induced hepatic angiosarcoma as well as their particular susceptibility to hepatocellular carcinoma. The molecular dosimetry of N2,3-εG in liver appears to be a sensitive and informative biomarker of genotoxic effect after exposure to VC. N2,3-εG was the predominant etheno adduct measured in vivo after exposure to VC, and the saturable nature of VC metabolism was reflected in its molecular dose. The relationships between endogenous N2,3-εG and that formed by low exposures to VC were demonstrated. Conclusions drawn from these exposures may be more relevant for risk assessment purposes than those drawn from high exposures where activation, detoxication, and repair pathways may be saturated or otherwise perturbed. These data are well suited for consideration in future risk assessments of VC that incorporate nontumor mode of action data.

VC6 is a major industrial chemical and widespread environmental contaminant whose carcinogenicity to humans and animals is well established (1). Concern about its carcinogenicity to humans was raised in 1974 after three workers at a VC polymerization plant were diagnosed with hepatic angiosarcoma, a tumor that is otherwise very rare in humans (2, 3, 4). Soon thereafter, occupational exposure limits were reduced to 1–5 ppm in most countries. No cases of hepatic angiosarcoma have been reported in VC workers who began employment after that time.

VC-induced carcinogenicity occurs by genotoxic mechanisms. VC is oxidized by CYP450 2E1 in a concentration-dependent manner (5, 6). The metabolic elimination of VC departs from first-order kinetics at 200–300 ppm in Rhesus monkeys (7) and at approximately 220 ppm in Sprague Dawley rats (8). The primary oxidation product of VC, the reactive electrophile 2-chloroethylene oxide, appears to be responsible for its genotoxicity (9, 10). As shown in Fig. 1, 2-chloroethylene oxide reacts with DNA to form 7-(2-oxoethyl)guanine and a variety of exocyclic base adducts (9, 11). Although 7-(2-oxoethyl)guanine is the major adduct formed in vitro, it was found to be nonmutagenic when tested with a bacterial DNA polymerase (12). Therefore, the promutagenic exocyclic adducts are believed to be responsible for VC-induced genotoxicity. The formation, role, and detection of these adducts have been reviewed (13).

One such adduct, N2,3-εG, efficiently induces G→A transitions in a variety of test systems (14, 15, 16, 17). This mutation was subsequently found to be common in ras proto-oncogenes from hepatic angiosarcomas of workers or rats exposed to VC (18, 19, 20). Although N2,3-εG is formed in relatively small amounts after exposure to VC, its presence is significant because of its proven mutagenic potential and apparent biological persistence (21, 22). Routine quantitation of this adduct in vivo necessitated the development of an ultrasensitive, highly specific GC/ECNCI/HRMS assay (23). The assay was recently improved by the addition of IA chromatography for adduct purification (24).

A low background concentration of N2,3-εG and other exocyclic adducts has consistently been detected in unexposed animals and humans by several laboratories (24, 25, 26, 27, 28). It is believed that oxidative metabolic processes are at least partially responsible for this background presence. The formation of N2,3-εG, 1,N2-ethenoguanine, 1,N6-εA, 3,N4-εC, and related adducts from the reaction of lipid peroxidation products with nucleosides or DNA has been demonstrated (29, 30, 31, 32, 33, 34). It is conceivable that this background mutagenic potential may be involved in the etiology of some spontaneous tumors.

Hepatic angiosarcoma originates in the sinusoidal-lining endothelial cells that represent the majority of the NPC population (35). However, the mechanisms that underlie the targeting of VC to the NPCs are not understood. Although VC is activated primarily in HEPs (36), its metabolites can migrate to endothelial cells (37). The base excision repair enzyme MPG, which was shown to remove etheno adducts from DNA in vitro(38), was induced in HEPs, but not in NPCs, after exposure of adult Sprague Dawley rats to vinyl fluoride (39). Based on this evidence, it was speculated that this relative deficiency in DNA repair may be a critical determinant for the particular susceptibility of NPCs to the carcinogenic effects of vinyl halides.

Age-dependent differences in adduct formation are known to correlate with susceptibility to VC-induced carcinogenesis. N2,3-εG has been measured in the livers of preweanling rats at concentrations approximately 4-fold higher than those in adult rats simultaneously exposed to VC (21). It is possible that higher DNA adduct concentrations may play an important role in determining the greater susceptibility of young rats to VC-induced liver tumors (40, 41).

VC is an excellent model for studying the exposure response to chemical carcinogens at the molecular level. Although the formation of DNA adducts in rats exposed to VC has been well characterized, these studies all used exposures to relatively high concentrations of VC that are known to saturate its metabolism. In contrast, the present study used a variety of exposures to determine the molecular dose from high saturating concentrations down to concentrations that approach current occupational exposures. Preliminary results from analyses of a small number of mixed liver cell fractions obtained in this study have been described elsewhere (42, 43). This report examines the mechanisms behind VC-induced carcinogenesis by determining the molecular dosimetry of N2,3-εG separately in HEPs and NPCs, by characterizing the accumulation and repair of N2,3-εG, and by studying the influence of age on DNA adduct dosimetry.

Chemicals.

VC (99% chemically pure) was purchased from Supelco (Bellefonte, PA). Clostridium histolyticum type 2 collagenase was obtained from Worthington Biochemical (Freehold, NJ). Nucleic acid purification grade lysis buffer, phenol/chloroform/water, and proteinase K were purchased from PE Biosystems (Foster City, CA). RNases A and T1 and Percoll were purchased from Sigma (St. Louis, MO). Other chemicals and analytical standards were obtained and characterized as described by Ham et al.(24).

Animal Exposures.

Exposures for the exposure response study were conducted at Huntingdon Life Sciences (East Millstone, NJ), a laboratory that is fully accredited by the Association for Assessment and Accreditation of Laboratory Care International. All procedures were in compliance with the appropriate parts of the Animal Welfare Act. For the adult studies, 11-week-old male Sprague Dawley rats (450–550 g) were purchased from Charles River Laboratories (Portage, MI). The rats were housed in stainless steel cages with a 12-h light/dark cycle. Certified Rodent Diet No. 5002 (PMI Feeds, St. Louis, MO) was provided ad libitum during nonexposure periods, and water was provided at all times. Eight animals per group were acclimated for 1 week and then exposed to target concentrations of 0, 10, 100, or 1100 ppm VC in a whole-body inhalation apparatus for 1 or 4 weeks (6 h/day, 5 days/week). Animals in the recovery group were similarly exposed for 4 weeks and maintained in a control environment for 5 days before they were killed. For the weanling study, 8 females with 8 male pups each were purchased, and 40 pups weaned at day 25 were selected for the study. The weanlings were exposed to VC for 5 days as described above. The atmosphere in each exposure chamber was sampled hourly, and the VC concentration was measured with a Miran 1A infrared spectrophotometer. Each individual measurement was within 20% of the target VC concentration, and the mean for each 6-h exposure period was within 10% of the target concentration.

Cell Separations.

After exposure, the rats were anesthetized with i.p. pentobarbital, heparinized i.v., and killed by incision of the abdominal aorta. The liver was perfused in situ with collagenase via the portal vein to obtain a mixed liver cell suspension as described previously (44). A portion of the suspension was retained and frozen on dry ice for mixed liver cell analyses. The remaining suspension was centrifuged (50 × g, 3 min), and the supernatant was transferred to another tube. The pellet was resuspended in 30 ml of PBS (140 mm NaCl, 2.7 mm KCl, 8.1 mm Na2HPO4, and 1.5 mm KH2PO4), and this process was repeated.

Percoll density medium was diluted to the appropriate concentrations with PBS. HEPs were purified by adding 30 ml of 60% Percoll to the pellet and centrifuging the resulting suspension (50 × g, 10 min). The supernatant was discarded, and the tube was wiped clean above the pellet. The HEPs were resuspended in 20 ml of PBS, which was then aliquoted into two 15-ml centrifuge tubes. These tubes were centrifuged (350 × g, 5 min), and the supernatant was discarded. The HEP pellets were frozen on dry ice and stored at −80°C until analysis.

NPCs were isolated by combining and centrifuging the supernatant fractions that were retained earlier (350 × g, 5 min). The supernatant was discarded, and the pellet was redissolved in 20 ml of PBS. NPCs were purified essentially as described by Pertoft and Smedsrød (35). Equal amounts of the suspension were carefully transferred to each of two centrifuge tubes prepared by layering 20 ml of 25% Percoll above 15 ml of 50% Percoll. The tubes were centrifuged without braking (800 × g, 20 min), and NPCs were harvested at the interface. The purity of the NPC fraction was estimated to be >95% when aliquots of selected samples were stained with trypan blue and examined with a hemocytometer. Each sample was centrifuged again (350 × g, 5 min), the supernatant was discarded, and the purified NPCs were frozen on dry ice and stored at −80°C.

For the accumulation study, whole liver samples were obtained from male Sprague Dawley rats exposed by inhalation to 500 ppm VC for 1, 2, 4, or 8 weeks (4 h/day, 5 days/week) at IARC (Lyon, France) as described in detail by Guichard et al.(45). The samples were transferred on dry ice and stored at −80°C.

DNA Extraction.

Cellular DNA was isolated using a modified phenol/chloroform extraction procedure with 2% BHT in isopropanol (w/v) added to reagents and samples as an antioxidant. The HEP pellet was suspended in 5 ml of lysis buffer containing 0.04% BHT (w/v), and an additional 25 μl of 2% BHT solution were added. RNases A (5 units) and T1 (270 units) were added, and the samples were incubated for 2 h at 37°C. Next, proteinase K (45 units) was added, and the samples were incubated for an additional 2 h at 37°C. The samples were extracted twice with 5 ml of phenol/chloroform/water containing 0.04% BHT (w/v) and once with 5 ml of alumina-purified chloroform containing 0.04% BHT (w/v). DNA was precipitated by the addition of 0.5 ml of 3 m sodium acetate and 10 ml of cold 95% ethanol. The pellet was washed with cold 70% ethanol and redissolved overnight in water at 4°C. The concentration of each sample was estimated by A260 nm, and the DNA was stored at −80°C until analysis. The liver samples provided by the IARC group were first homogenized in 10 ml of PBS and centrifuged (1000 × g, 15 min). Nuclear DNA was harvested by an automated phenol/chloroform extraction procedure (Applied Biosystems 340A; Applied Biosystems, Foster City, CA). RNA contamination of DNA was measured essentially as described by Boucheron et al.(46).

Quantitation of N2,3-εG/Unmodified Guanine.

N2,3-εG was quantitated with the acid hydrolysis/IA/GC/ECNCI/HRMS method of Ham et al.(24), with minor modifications as described below. Briefly, 70 fmol of [13C4,15N2]N2,3-εG internal standard (30 μl) were added to 50–400 μg of DNA. Water and 100 μl of 1 n HCl were added to yield a total volume of 1000 μl, and the solution was incubated for 30 min at 70°C. The hydrolysate was transferred to Centricon 10 membrane filters (Millipore, Bedford, MA) and purified by centrifugation (5000 × g, 60 min). A 50–100-μl aliquot of the filtrate was retained for guanine quantitation with high-performance liquid chromatography as described previously (24). Initially, the remaining filtrate was neutralized by adding 0.1 volume of 1 n NaOH and 2 ml of PBS to the retentate, followed by centrifugation (5000 × g, 90 min). To minimize pH variation in later analyses, the filtrates were neutralized by the addition of 2 ml of 500 mm sodium phosphate buffer (pH 7.2) to the retentate, followed by centrifugation (5000 × g, 90 min).

N2,3-εG was isolated from the filtrate by IA chromatography as described previously (24). The adduct was eluted with 3 ml of 100% methanol, and the fraction was dried by Speed-vac vacuum centrifugation. To each sample were added ∼25 mg of dry, powdered potassium carbonate, and 500 μl of acetone. Next, 35 μl of 5% pentafluorobenzyl bromide/acetone were added. The tubes were capped and mixed at 50°C for 70 min. The samples were dried by Speed-vac vacuum centrifugation and then extracted and transferred in two 100-μl portions of dichloromethane to disposable silica gel solid-phase extraction columns prepared as described previously (24). The samples were washed with 4 ml of hexane and 6 ml of 5% ethyl acetate/hexane. The derivatives were eluted in 3 ml of ethyl acetate, dried again, redissolved in 15 μl of toluene, and transferred to silanized GC vial inserts for analysis.

The GC/ECNCI/HRMS analyses were made with a HP 5890 GC coupled to a Varian VG70–250SEQII mass spectrometer as described by Ham et al.(24). A 2-μl volume of each sample was injected. Selected ion monitoring at m/z = 354.0413 and m/z = 360.0489 was used to quantitate the [M-181] fragments of 3,5-PFB2-N2,3-εG and 3,5-PFB2-[13C4,15N2]N2,3-εG, respectively. The mass resolving power was 7–10 × 103, and the minimum signal:noise ratio used for quantitation was 5:1. The amount of N2,3-εG in each sample was calculated from the ratio of the analyte and internal standard peak areas. The N2,3-εG concentration in each sample was determined by normalizing the adduct measurement to the amount of guanine measured in the sample by high-performance liquid chromatography. Each result was expressed as the molar ratio between N2,3-εG and unmodified guanine in the sample (mol N2,3-εG:mol guanine).

Statistics.

Analyses were performed with the aid of Instat version 3.00 for Windows 95 software (GraphPad Software, San Diego, CA). Differences in N2,3-εG concentrations between HEP and NPC populations within exposure groups were compared using the unpaired t test with Welch correction. Pairing of HEP and NPC measurements from the same rats was not attempted because fewer NPC than HEP samples were analyzed. Adduct concentrations were compared between exposure regimens with one-way ANOVA followed by multiple comparisons tests performed as follows. N2,3-εG concentrations from exposure groups within each exposure regimen were compared with the matched controls using Dunnett tests. Comparisons between different age groups or exposure regimens were made with the Tukey-Kramer multiple comparisons test. In each case, P ≤ 0.05 (two-sided) was considered to be statistically significant. If a sample was analyzed more than once, the result with the strongest internal standard signal (i.e., highest sensitivity) was used.

Although the DNA adducts induced by VC have been well studied, this is the first comprehensive description of the exposure response to VC across a broad range of exposure concentrations. The formation of N2,3-εG in DNA was used as a biomarker of VC exposure down to a concentration that is within an order of magnitude of current occupational exposure limits. Because VC carcinogenicity preferentially targets the sinusoidal-lining endothelial cells, it was important to consider the molecular dosimetry of N2,3-εG separately in NPCs and HEPs.

The IA/GC/ECNCI/HRMS assay used for these determinations is very sensitive and highly specific (24). The lower limit of quantitation was 1–2 fmol N2,3-εG/sample, which corresponded to approximately 1 mol N2,3-εG/108 mol unmodified guanine in 250 μg of DNA. Due to a low yield of NPC from some livers, fewer NPC than HEP samples were available for analysis. Although the rapid DNA extraction procedure (which was designed to minimize the potential for artifactual adduct formation) allowed RNA to remain in some samples, the assay was found to be unaffected by this contamination (data not shown).

A low background amount of endogenous N2,3-εG was consistently detected in HEP and NPC DNA from control rats (Table 1; Fig. 2). This observation is consistent with previous measurements of N2,3-εG in the liver of unexposed rats (24, 26, 34). This background concentration represents a threshold for the detection of N2,3-εG induced by exposure to VC, but it is not known whether this also reflects a threshold for the carcinogenic response to VC. It will be important to assess the biological role of endogenous N2,3-εG in future studies because its presence could profoundly influence the estimation of the risk associated with exposure to low concentrations of VC (43).

N2,3-εG was clearly induced in HEPs and NPCs after exposure of adult rats to VC for 1 or 4 weeks (Table 1; Fig. 2). The exposure response curves derived from the HEP and NPC populations were supralinear after the 1- and 4-week exposures, with a linear increase from 0 to 100 ppm and a marked plateau between 100 and 1000 ppm (Fig. 3). This response is similar to that previously characterized in mixed liver cells obtained from some of these rats (42, 43). The supralinear curve is thought to be a consequence of the saturable nature of VC metabolism and is consistent with the pattern of tumor incidences observed in a long-term cancer bioassay (40). These correlations suggest that the molecular dosimetry of N2,3-εG in liver is predictive of cancer risk in this model. By demonstrating the importance of metabolism to the genotoxic response, these data support the relevance of target tissue dose metrics calculated with PBPK models to the cancer risk assessment of VC (47).

Due to the presence of endogenous N2,3-εG, it is important to consider the relative increases in adduct concentrations over controls in addition to their absolute values. When adult rats were exposed to 10 ppm VC for 4 weeks, there was a 5-fold increase in N2,3-εG over the endogenous concentration measured in control rats. If this result is linearly interpolated to a 1 ppm occupational exposure limit, it is estimated that this concentration of VC would increase the biological burden of N2,3-εG by only 50%. In contrast, N2,3-εG concentrations in HEPs were increased 37-fold when rats were exposed to 100 or 1100 ppm VC for 4 weeks. The disparity between the relative increases in adduct concentrations associated with each of these exposures is consistent with the epidemiological observation that VC-induced hepatic angiosarcoma was largely associated with autoclave workers who were routinely exposed to VC concentrations estimated at 500 ppm or more (48, 49).

Based upon previous analyses of a small number of HEP and NPC fractions from vinyl fluoride-exposed rats with an earlier version of this assay (42), we hypothesized that N2,3-εG would be present in NPCs at higher concentrations than in HEPs after exposure to VC. However, the response of the HEP and NPC populations to VC was similar in adult rats in this study. This suggests that differences in adduct concentrations between HEPs and NPCs are not indicative of the susceptibility of each population to VC-induced carcinogenicity. Due to the presence of other cell types in the NPC fraction, the possibility of a small difference in N2,3-εG between HEPs and endothelial cells cannot be excluded, however.

Important differences between HEPs and NPCs that were not characterized in this experiment might also contribute to the susceptibility of the latter population to VC-induced carcinogenesis. Because the basal proliferation rate of NPCs is more than 2-fold greater than that of HEPs in rats (50), a similar adduct concentration may cause more mutations to be fixed in the NPC DNA. Another potential factor is the demonstrated difference in DNA repair capacity between HEPs and NPCs (39). However, the rates of accumulation in the two cell populations would be expected to be different if significant differences in repair existed. The parallel nature of the molecular dosimetry curves between HEPs and NPCs at 1 and 4 weeks supports a similar rate of repair.

There was a 2–3-fold accumulation of N2,3-εG between 1 and 4 weeks in both cell populations (Table 1; Fig. 3). N2,3-εG concentrations in HEP and NPC DNA from rats exposed for 4 weeks and allowed to recover for 1 week were similar to those measured in rats exposed for 4 weeks and killed immediately (Table 1; Fig. 4). The lack of a decrease in N2,3-εG concentrations in the recovery group is consistent with a previous half-life estimate of more than 30 days in preweanling rats (21). There appeared to be a small but consistent increase in N2,3-εG concentrations in HEPs after the recovery period, although this difference was only significant in rats exposed to 1100 ppm VC. Although the underlying mechanisms are not understood, there is precedence for an increase in adduct concentrations during a recovery period; a similar phenomenon was reported in a study of the malondialdehyde-induced guanine adduct M1G (51).

To further examine the repair of N2,3-εG, tissues from a collaborative study with the IARC were examined. Whole liver DNA samples from male Sprague Dawley rats exposed to 500 ppm VC for 1, 2, 4, or 8 weeks (4 h/day, 5 days/week) were analyzed using the present method. The high degree of linearity (r2 > 0.99) in the resulting curve (Fig. 5) is consistent with the slow repair of N2,3-εG in this model. It is important to note that these results contradict the previous finding of a plateau in N2,3-εG concentrations between 4 and 8 weeks of exposure when this DNA was analyzed with a less sophisticated version of this assay (42), and the accumulation of N2,3-εG now appears to be similar to that of 1,N6 -εA and 3,N4-εC in these liver samples (45). The concentration of N2,3-εG measured in these rats was much greater than the concentrations of 1,N6-εA and 3,N4-εC previously measured in the livers of the same rats with an IA/32P postlabeling assay (45). After 8 weeks of exposure to 500 ppm VC, the concentration of N2,3-εG was approximately 30-fold greater than that of 1,N6 -εA and 20-fold greater than that of 3,N4-εC. The remaining etheno adduct, 1,N2-ethenoguanine, was not detected in HEPs from rats exposed to VC and thus is not believed to play a significant role in VC-induced carcinogenesis (34).

Although no loss of N2,3-εG from DNA was observed after the 1-week recovery period, the accumulation of this adduct was disproportional to the time of the exposure. For example, there was only a 2–3-fold accumulation of N2,3-εG between 1 and 4 weeks of exposure in HEP DNA from the present study as well as liver DNA from the IARC collaboration. These results strongly suggest that N2,3-εG is slowly repaired in vivo. Furthermore, the persistence of N2,3-εG observed in these studies does not necessarily contradict the published data regarding the repair of this adduct by MPG. Although it was shown that etheno adducts were released from DNA in the presence of MPG, N2,3-εG was released to a much smaller extent than 1,N6-εA (38). It will be necessary to measure N2,3-εG after a longer recovery period to determine whether differences in repair might contribute to the greater susceptibility of NPC to VC-induced carcinogenesis. In any case, the persistence of N2,3-εG effectively amplifies the mutagenic potential of this adduct in vivo. These findings demonstrate the value of molecular dosimetry to the study of DNA repair. As was reported with the M1G adduct (51), a repair mechanism demonstrated in vitro appeared to be inefficient in the whole organism.

N2,3-εG was detected as both an endogenous and VC-induced lesion in HEPs and NPCs harvested from weanling rats (Table 1; Fig. 6). N2,3-εG was measured at significantly higher concentrations in weanling rats than in adult rats exposed for the same time. This is thought to be a consequence of greater CYP450 2E1 protein concentrations in the livers of young rats.7 In this study, the concentration of N2,3-εG measured in HEPs and NPCs of weanling rats was approximately 2–3-fold greater than that in similarly exposed adults. This finding is consistent with the 4-fold difference reported between preweanling rats and their dams after exposure to 600 ppm VC for 5 days (21). (These adduct measurements are expected to be slightly underestimated due to dilution as a consequence of the relatively high cell proliferation rate in the livers of young rats.) A greater burden of adducts, coupled with the high cell proliferation rate in the livers of young rats, may play a role in determining the higher susceptibility of young rats to VC-induced carcinogenesis compared with adults (41). These findings provide mechanistic support for the decision to include an additional protection factor for young populations in the recent United States Environmental Protection Agency VC risk assessment.8

In contrast to adults, a difference in adduct concentrations was detected between the HEP and NPC populations from the weanlings. The concentration of N2,3-εG in HEPs was significantly greater than that measured in NPCs after the 10 and 100 ppm VC exposures (P < 0.01), although the adduct concentration in both populations was similar at 1100 ppm. This differential response of the HEP and NPC populations may contribute to the particular susceptibility of young rats to VC-induced hepatocellular carcinoma (40, 41).

After exposure to genotoxic carcinogens, DNA adducts are sensitive biomarkers that provide information about promutagenic effects on the target molecule. When the concentration of a DNA adduct is measured, aspects of absorption, transport, activation, and detoxication are incorporated into the result. This can greatly simplify extrapolations and help define differences between species, strains, and routes of exposure. Furthermore, this end point makes it possible to characterize not only the formation but also the repair of a DNA adduct. This allows the study of DNA repair in the presence of all of the repair pathways that are present under relevant conditions in the whole organism. Finally, advances in technology now allow the study of the effects of doses at or near concentrations commonly experienced by humans. Conclusions drawn from these exposures may be more relevant for risk assessment purposes than those drawn from high exposures, at which activation, detoxication, and repair pathways may be saturated or otherwise perturbed (26, 52).

The United States Environmental Protection Agency made provisions in the 1996 Proposed Guidelines for Carcinogen Risk Assessment (53) for the consideration of nontumor data in chemical risk assessments. These results provide mode of action evidence that serves to increase confidence in the agency’s recent VC risk assessment, which relied largely on the PBPK model of Clewell et al.(47). If appropriate human dosimetry data become available in the future, these results could be incorporated into the PBPK model to extrapolate the genotoxic effects of VC metabolites across species. Such a model might also consider the influence of endogenous DNA adducts on the risk posed by exposure to VC. It is hoped that the inclusion of such experimental data may improve our ability to evaluate carcinogenic risk in the future, particularly at the low concentrations that are relevant to most contemporary occupational and environmental exposures.

The findings detailed in this report strongly support a causal role for N2,3-εG in VC-induced carcinogenesis. The exposure-response curve correlates well with the kinetics of VC metabolism as well as tumor incidences observed in cancer bioassays. N2,3-εG appears to be the predominant etheno adduct present in the livers of rats exposed to VC and is readily induced in the NPC population that is targeted by VC. The long persistence and potent mutagenicity of N2,3-εG suggest that it would effectively miscode in vivo. Taken together, these findings implicate N2,3-εG as an important precursor lesion to carcinogenesis induced by VC.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported in part by NIH grants (ES07126, CA72319, CA 83369, CA16086, ES05948, and ES10126), and the Chemical Manufacturers’ Association.

6

The abbreviations used are: VC, vinyl chloride; N2,3-εG, N2,3-ethenoguanine; GC, gas chromatography; ECNCI, electron capture negative chemical ionization; HRMS, high-resolution mass spectrometry; IA, immunoaffinity; 1,N6-εA, 1,N6-ethenoadenine; 3,N4-εC, 3,N4-ethenocytosine; NPC, nonparenchymal cell; HEP, hepatocyte; MPG, N-methylpurine DNA glycosylase; BHT, butylated hydroxytoluene; PBPK, physiologically based pharmacokinetic; M1G, pyrimidol [1,2-α]purin-10(3H)-one.

7

F. P. Guengerich, unpublished data.

8

United States Environmental Protection Agency. Toxicological review of vinyl chloride in support of summary information on the Integrated Risk Information System (2000). Available online from National Center for Environmental Assessment (http://www.epa.gov/iris).

We thank Raymond Schroeder (Huntingdon Life Sciences, East Millstone, NJ) for directing the rat inhalation exposures. We greatly appreciate the assistance of Brandon Aron in the cell separation procedures. The liver samples used for the accumulation experiment were kindly provided as part of a collaborative effort with Drs. Alan Barbin and Helmut Bartsch (IARC, Lyon, France).

1
IARC. Overall evaluations of carcinogenicity: an updating of IARC Monographs Volumes 1 to 42.
IARC Monogr. Eval. Carcinog. Risks Hum.
,
S7
:
373
-376,  
1987
.
2
Creech J. L., Jr., Johnson M. N. Angiosarcoma of liver in the manufacture of polyvinyl chloride.
J. Occup. Med.
,
16
:
150
-151,  
1974
.
3
Baxter P. J., Anthony P. P., Macsween R. N., Scheuer P. J. Angiosarcoma of the liver: annual occurrence and aetiology in Great Britain.
Br. J. Ind. Med.
,
37
:
213
-221,  
1980
.
4
Falk H., Herbert J., Crowley S., Ishak K. G., Thomas L. B., Popper H., Caldwell G. G. Epidemiology of hepatic angiosarcoma in the United States: 1964–1974.
Environ. Health Perspect.
,
41
:
107
-113,  
1981
.
5
Guengerich F. P., Kim D. H., Iwasaki M. Role of human cytochrome P-450 IIE1 in the oxidation of many low molecular weight cancer suspects.
Chem. Res. Toxicol.
,
4
:
168
-179,  
1991
.
6
Watanabe P. G., Gehring P. J. Dose-dependent fate of vinyl chloride and its possible relationship to oncogenicity in rats.
Environ. Health Perspect.
,
17
:
145
-152,  
1976
.
7
Buchter A., Filser J. G., Peter H., Bolt H. M. Pharmacokinetics of vinyl chloride in the Rhesus monkey.
Toxicol. Lett.
,
6
:
33
-36,  
1980
.
8
Hefner R. E., Watanabe P. G., Gehring P. J. Preliminary studies of the fate of inhaled vinyl chloride monomer in rats.
Ann. N. Y. Acad. Sci.
,
246
:
135
-148,  
1975
.
9
Guengerich F. P. Roles of the vinyl chloride oxidation products 2-chlorooxirane and 2-chloroacetaldehyde in the in vitro formation of etheno adducts of nucleic acid bases.
Chem. Res. Toxicol.
,
5
:
2
-5,  
1992
.
10
Guengerich F. P., Crawford W. M., Watanabe P. G. Activation of vinyl chloride to covalently bound metabolites: roles of 2-chloroethylene oxide and 2-chloroacetaldehyde.
Biochemistry
,
18
:
5177
-5182,  
1979
.
11
Guengerich F. P., Persmark M. Mechanism of formation of ethenoguanine adducts from 2-haloacetaldehydes: 13C-labeling patterns with 2-bromoacetaldehyde.
Chem. Res. Toxicol.
,
7
:
205
-208,  
1994
.
12
Barbin A., Laib R. J., Bartsch H. Lack of miscoding properties of 7-(2-oxoethyl)guanine, the major vinyl chloride-DNA adduct.
Cancer Res.
,
45
:
2440
-2444,  
1985
.
13
Bartsch H., Barbin A., Marion M. J., Nair J., Guichard Y. Formation, detection, and role in carcinogenesis of ethenobases in DNA.
Drug Metab. Rev.
,
26
:
349
-371,  
1994
.
14
Singer B., Spengler S. J., Chavez F., Kusmierek J. T. The vinyl chloride-derived nucleoside, N2,3-ethenoguanosine, is a highly efficient mutagen in transcription.
Carcinogenesis (Lond.)
,
8
:
745
-747,  
1987
.
15
Mroczkowska M. M., Kusmierek J. T. Miscoding potential of N2,3-ethenoguanine studied in an Escherichia coli DNA-dependent RNA polymerase in vitro system and possible role of this adduct in vinyl chloride-induced mutagenesis.
Mutagenesis
,
6
:
385
-390,  
1991
.
16
Cheng K. C., Preston B. D., Cahill D. S., Dosanjh M. K., Singer B., Loeb L. A. The vinyl chloride DNA derivative N2,3-ethenoguanine produces G→A transitions in Escherichia coli.
Proc. Natl. Acad. Sci. USA
,
88
:
9974
-9978,  
1991
.
17
Singer B., Kusmierek J. T., Folkman W., Chavez F., Dosanjh M. K. Evidence for the mutagenic potential of the vinyl chloride induced adduct, N2,3-etheno-deoxyguanosine, using a site-directed kinetic assay.
Carcinogenesis (Lond.)
,
12
:
745
-747,  
1991
.
18
Marion M. J., Froment O., Trepo C. Activation of ki-ras gene by point mutation in human liver angiosarcoma associated with vinyl chloride exposure.
Mol. Carcinogen.
,
4
:
450
-454,  
1991
.
19
Froment O., Boivin S., Barbin A., Bancel B., Trepo C., Marion M. J. Mutagenesis of ras proto-oncogenes in rat liver tumors induced by vinyl chloride.
Cancer Res.
,
54
:
5340
-5345,  
1994
.
20
DeVivo I., Marion M. J., Smith S. J., Carney W. P., Brandt-Rauf P. W. Mutant c-Ki-ras p21 protein in chemical carcinogenesis in humans exposed to vinyl chloride.
Cancer Causes Control
,
5
:
273
-278,  
1994
.
21
Fedtke N., Boucheron J. A., Walker V. E., Swenberg J. A. Vinyl chloride-induced DNA adducts. II. Formation and persistence of 7-(2′-oxoethyl)guanine and N2,3-ethenoguanine in rat tissue DNA.
Carcinogenesis (Lond.)
,
11
:
1287
-1292,  
1990
.
22
Swenberg J. A., Fedtke N., Ciroussel F., Barbin A., Bartsch H. Etheno adducts formed in DNA of vinyl chloride-exposed rats are highly persistent in liver.
Carcinogenesis (Lond.)
,
13
:
727
-729,  
1992
.
23
Fedtke N., Boucheron J. A., Turner M. J., Swenberg J. A. Vinyl chloride-induced DNA adducts. I. Quantitative determination of N2,3-ethenoguanine based on electrophore labeling.
Carcinogenesis (Lond.)
,
11
:
1279
-1285,  
1990
.
24
Ham A. J. L., Ranasinghe A., Morinello E. J., Nakamura J., Upton P. B., Johnson F., Swenberg J. A. Immunoaffinity/gas chromatography/high resolution mass spectrometry method for the detection of N2,3-ethenoguanine.
Chem. Res. Toxicol.
,
12
:
1240
-1246,  
1999
.
25
Chaudhary A. K., Nokubo M., Reddy G. R., Yeola S. N., Morrow J. D., Blair I. A., Marnett L. J. Detection of endogenous malondialdehyde-deoxyguanosine adducts in human liver.
Science (Wash. DC)
,
265
:
1580
-1582,  
1994
.
26
Swenberg J. A., La D. K., Scheller N. A., Wu K. Y. Dose-response relationships for carcinogens.
Toxicol. Lett.
,
82–83
:
751
-756,  
1995
.
27
Nair J., Barbin A., Guichard Y., Bartsch H. 1. N6-Ethenodeoxyadenosine and 3,N4-ethenodeoxycytidine in liver DNA from humans and untreated rodents detected by immunoaffinity/32P-postlabelling.
Carcinogenesis (Lond.)
,
16
:
613
-617,  
1995
.
28
Nath R. G., Ocando J. E., Chung F. L. Detection of 1,N2-propanodeoxyguanosine adducts as potential endogenous DNA lesions in rodent and human tissues.
Cancer Res.
,
56
:
452
-456,  
1996
.
29
Sodum R. S., Chung F. L. 1,N2-Ethenodeoxyguanosine as a potential marker for DNA adduct formation by trans-4-hydroxy-2-nonenal.
Cancer Res.
,
48
:
320
-323,  
1988
.
30
Sodum R. S., Chung F. L. Structural characterization of adducts formed in the reaction of 2,3-epoxy-4-hydroxynonanal with deoxyguanosine.
Chem. Res. Toxicol.
,
2
:
23
-28,  
1989
.
31
El-Ghissassi F., Barbin A., Nair J., Bartsch H. Formation of 1,N6-ethenoadenine and 3,N4-ethenocytosine by lipid peroxidation products and nucleic acid bases.
Chem. Res. Toxicol.
,
8
:
278
-283,  
1995
.
32
Chen H. J., Zhang L., Cox J., Cunningham J. A., Chung F. L. DNA adducts of 2,3-epoxy-4-hydroxynonanal: detection of 7-(1′,2′-dihydroxyheptyl)-3H-imidazo[2,1-i]purine and 1,N6-ethenoadenine by gas chromatography/negative ion chemical ionization/mass spectrometry.
Chem. Res. Toxicol.
,
11
:
1474
-1480,  
1998
.
33
Ham A. J. L., Ranasinghe A., Koc H., Swenberg J. A. 4-Hydroxy-2-nonenal and ethyl linoleate form N2,3-ethenoguanine under peroxidizing conditions.
Chem. Res. Toxicol.
,
13
:
1243
-1250,  
2000
.
34
Morinello E. J., Ham A. J. L., Ranasinghe A., Sangaiah R., Swenberg J. A. Simultaneous quantitation of N2,3-ethenoguanine and 1,N2-ethenoguanine with an immunoaffinity/gas chromatography/high resolution mass spectrometry assay.
Chem. Res. Toxicol.
,
14
:
327
-334,  
2001
.
35
Pertoft H., Smedsrød B. Separation and characterization of liver cells Pretlow T. G. P. Pretlow T. P. eds. .
Cell Separation: Methods and Selected Applications
,
Vol. 4
:
1
-24, Academic Press Orlando, FL  
1987
.
36
Ottenwalder H., Bolt H. M. Metabolic activation of vinyl chloride and vinyl bromide by isolated hepatocytes and hepatic sinusoidal cells.
J. Environ. Pathol. Toxicol.
,
4
:
411
-417,  
1980
.
37
Kuchenmeister F., Wang M., Klein R. G., Schmezer P. Transport of reactive metabolites of procarcinogens between different liver cell types, as demonstrated by the single cell microgel electrophoresis assay.
Toxicol. Lett.
,
88
:
29
-34,  
1996
.
38
Dosanjh M. K., Chenna A., Kim E., Fraenkel-Conrat H., Samson L., Singer B. All four known cyclic adducts formed in DNA by the vinyl chloride metabolite chloroacetaldehyde are released by a human DNA glycosylase.
Proc. Natl. Acad. Sci. USA
,
91
:
1024
-1028,  
1994
.
39
Holt S., Roy G., Mitra S., Upton P. B., Bogdanffy M. S., Swenberg J. A. Deficiency of N-methylpurine-DNA-glycosylase expression in nonparenchymal cells, the target cell for vinyl chloride and vinyl fluoride.
Mutat. Res.
,
460
:
105
-115,  
2000
.
40
Maltoni C., Lefemine G., Ciliberti A., Cotti G., Carretti D. Carcinogenicity bioassays of vinyl chloride monomer: a model of risk assessment on an experimental basis.
Environ. Health Perspect.
,
41
:
3
-29,  
1981
.
41
Maltoni C., Cotti G. Carcinogenicity of vinyl chloride in Sprague-Dawley rats after prenatal and postnatal exposure.
Ann. N. Y. Acad. Sci.
,
534
:
145
-159,  
1988
.
42
Swenberg J. A., Bogdanffy M. S., Ham A. J., Holt S., Kim A., Morinello E. J., Ranasinghe A., Scheller N., Upton P. B. Formation and repair of DNA adducts in vinyl chloride- and vinyl fluoride-induced carcinogenesis Singer B. Bartsch H. eds. .
Exocyclic DNA Adducts in Mutagenesis and Carcinogenesis
,
29
-43, IARC Scientific Publications Lyon, France  
1999
.
43
Swenberg J. A., Ham A. J., Koc H., Morinello E., Ranasinghe A., Tretyakova N., Upton P. B., Wu K. Y. DNA adducts: effects of low exposure to ethylene oxide, vinyl chloride and butadiene.
Mutat. Res.
,
464
:
77
-86,  
2000
.
44
Lindamood C., Bedell M. A., Billings K. C., Swenberg J. A. Alkylation and de novo synthesis of liver cell DNA from C3H mice during continuous dimethylnitrosamine exposure.
Cancer Res.
,
42
:
4153
-4157,  
1982
.
45
Guichard Y., El-Ghissassi F., Nair J., Bartsch H., Barbin A. Formation and accumulation of DNA ethenobases in adult Sprague-Dawley rats exposed to vinyl chloride.
Carcinogenesis (Lond.)
,
17
:
1553
-1559,  
1996
.
46
Boucheron J. A., Richardson F. C., Morgan P. H., Swenberg J. A. Molecular dosimetry of O4-ethyldeoxythymidine in rats continuously exposed to diethylnitrosamine.
Cancer Res.
,
47
:
1577
-1581,  
1987
.
47
Clewell H. J., Gentry P. R., Gearhart J. M., Allen B. C., Andersen M. E. Considering pharmacokinetic and mechanistic information in cancer risk assessments for environmental contaminants: examples with vinyl chloride and trichloroethylene.
Chemosphere
,
31
:
2561
-2578,  
1995
.
48
Jones R. D., Smith D. M., Thomas P. G. A mortality study of vinyl chloride monomer workers employed in the United Kingdom in 1940–1974.
Scand. J. Work Environ. Health
,
14
:
153
-160,  
1988
.
49
Simonato L., L’Abbe K. A., Andersen A., Belli S., Comba P., Engholm G., Ferro G., Hagmar L., Langard S., Lundberg I. A collaborative study of cancer incidence and mortality among vinyl chloride workers.
Scand. J. Work Environ. Health
,
17
:
159
-169,  
1991
.
50
Lewis J. G., Swenberg J. A. Effect of 1,2-dimethylhydrazine and diethylnitrosamine on cell replication and unscheduled DNA synthesis in target and nontarget cell populations in rat liver following chronic administration.
Cancer Res.
,
42
:
89
-92,  
1982
.
51
Kautiainen A., Vaca C. E., Granath F. Studies on the relationship between hemoglobin and DNA adducts of malonaldehyde and their stability in vivo.
Carcinogenesis (Lond.)
,
14
:
705
-708,  
1993
.
52
La D. K., Swenberg J. A. DNA adducts: biological markers of exposure and potential applications to risk assessment.
Mutat. Res.
,
365
:
129
-146,  
1996
.
53
United States Environmental Protection Agency Proposed Guidelines for Carcinogen Risk Assessment.
Fed. Reg.
,
61
:
17960
-18011,  
1996
.