Metabolism and Hemoglobin Adduct Formation of Acrylamide in Humans
http://www.100md.com
《毒物学科学杂志》
Research Triangle Institute, Research Triangle Park, North Carolina 27709
Covance Clinical Research Unit, Inc., Madison, Wisconsin 53703
University of Medicine and Dentistry of New Jersey, Newark, New Jersey 07103
ABSTRACT
Acrylamide (AM), used in the manufacture of polyacrylamide and grouting agents, is produced during the cooking of foods. Workplace exposure to AM can occur through the dermal and inhalation routes. The objectives of this study were to evaluate the metabolism of AM in humans following oral administration, to compare hemoglobin adduct formation on oral and dermal administration, and to measure hormone levels. The health of the people exposed under controlled conditions was continually monitored. Prior to conducting exposures in humans, a low-dose study was conducted in rats administered 3 mg/kg (1,2,3-13C3) AM by gavage. The study protocol was reviewed and approved by Institute Review Boards both at RTI, which performed the sample analysis, and the clinical research center conducting the study. (1,2,3-13C3) AM was administered in an aqueous solution orally (single dose of 0.5, 1.0, or 3.0 mg/kg) or dermally (three daily doses of 3.0 mg/kg) to sterile male volunteers. Urine samples (3 mg/kg oral dose) were analyzed for AM metabolites using 13C NMR spectroscopy. Approximately 86% of the urinary metabolites were derived from GSH conjugation and excreted as N-acetyl-S-(3-amino-3-oxopropyl)cysteine and its S-oxide. Glycidamide, glyceramide, and low levels of N-acetyl-S-(3-amino-2-hydroxy-3-oxopropyl)cysteine were detected in urine. On oral administration, a linear dose response was observed for N-(2-carbamoylethyl)valine (AAVal) and N-(2-carbamoyl-2-hydroxyethyl)valine (GAVal) in hemoglobin. Dermal administration resulted in lower levels of AAVal and GAVal. This study indicated that humans metabolize AM via glycidamide to a lesser extent than rodents, and dermal uptake was approximately 6.6% of that observed with oral uptake.
Key Words: acrylamide; glycidamide; hemoglobin; adducts.
INTRODUCTION
Acrylamide (AM) is used in the manufacture of water-soluble polymers (European Union, 2002). These polymers are then used for wastewater and sludge treatment, paper manufacture, soil stabilization, mining, and many other uses (European Union, 2002). AM is also a chemical intermediate in the manufacture of other monomeric chemicals and is used for grouting and preparation of laboratory gels for electrophoresis. Human exposure through these applications is very small, with the largest mean estimated exposures of 0.07 mg/kg/day thought to arise during manufacture of AM (European Union, 2002). Previously, it has been postulated that dermal absorption was the major route of human exposure to AM (European Union, 2002). The magnitude of this dermal absorption is highly relevant, as one of the uses of AM-based polymers is in the formulation of skin creams (European Union, 2002). Estimates of dermal absorption based on in vitro and rodent studies have ranged from 3 to 100% (European Union, 2002). Recently, exposure to AM in a variety of cooked foods has been described (Rosen and Hellenas, 2002; Tareke et al., 2002). Human exposure via this route is substantial, with estimated exposures as high as 70 μg per day proposed (Tareke et al., 2002).
AM is metabolized by two main pathways: glutathione conjugation (Dixit et al., 1982; Edwards, 1975; Hashimoto and Aldridge, 1970; Miller et al., 1982; Sumner et al., 1992), and epoxidation to glycidamide (GA) (Calleman et al., 1990; Sumner et al., 1992). The metabolism of AM in vivo results in the formation of a number of metabolites. The metabolism of AM in vivo has been investigated by administration of (1,2,3-13C3) AM to rodents, with the detection and quantitation of metabolites by 13C NMR spectroscopy (Sumner et al., 1992, 1999, 2003). The epoxidation reaction to GA is catalyzed by cytochrome P450 2E1 in rodents (Sumner et al., 1999). Both AM and GA react with hemoglobin, producing a stable adduct which can be measured as an indicator of exposure. Correlations have been made with hemoglobin adducts and neurotoxicity, but there has been no systematic standardization of hemoglobin adducts with dose. GA is mutagenic in the Salmonella test (Hashimoto and Tanii, 1985) and in human bronchial epithelial cells and Big Blue mouse embryonic fibroblasts (Besaratinia and Pfeifer, 2004). It can react with DNA in vitro to produce a guanine derivative N7-(2-carbamoyl-2-hydroxyethyl)guanine, and two adenine derivatives, N3-(2-carbamoyl-2-hydroxyethyl)adenine and N1-(2-carboxy-2-hydroxyethyl)-2'-deoxyadenosine (Gamboa da Costa et al., 2003; Segerback et al., 1995). In vivo, administration of AM to rats and mice produces low levels of N7-(2-carbamoyl-2-hydroxyethyl)guanine and N3-(2-carbamoyl-2-hydroxyethyl)adenine (Gamboa da Costa et al., 2003; Segerback et al., 1995).
AM induces a characteristic peripheral neurotoxicity in animals and man (Spencer and Schaumburg, 1974a,b, 1975). This toxicity manifests itself as a distal to proximal loss of nerve function and dying back of cells. AM also affects rodent reproduction, namely smaller litter size. At elevated AM doses other reproductive effects are seen, likely as a consequence of dominant lethal mutations at low doses and neurotoxicity at higher doses (Tyl and Friedman, 2003).
AM is carcinogenic in drinking water studies in laboratory rats (Friedman et al., 1995; Johnson et al., 1986). In male rats, it induces tumors of the tunica vaginalis testes and the thyroid, while in females, it induces mammary fibroadenomas and thyroid tumors (Friedman et al., 1995). The mechanism for this tumorigenicity is unclear, although interaction with the dopamine receptor has been postulated, as well as genotoxicity (Tyl and Friedman, 2003). Increases in DNA synthesis have been reported after subacute exposure in target tissues for tumor development (thyroid, testicular mesothelium, and adrenal medulla) and have been suggested to play a role in the carcinogenicity of AM (Lafferty et al., 2004). Understanding the mechanism of tumorigenicity is important, since conventional risk assessment techniques place the order of magnitude of the risk at approximately 10–3 with exposures of 70 μg/ day (Dybing and Sanner, 2003).
The relative contributions of AM and GA in the mode of action of AM are the subject of debate and current research. The conversion of AM to GA and differences that may occur between species, exposure route, and dose are important considerations in assessing the risk of the possible effects of AM exposures in the diet, in consumer products, and in the workplace.
The primary objectives of this study were to evaluate the conversion of AM to GA in people exposed to AM and to evaluate the extent of uptake following dermal administration. This was conducted by administering a low dose of 13C-labeled (substituted) AM to volunteers orally or dermally, and by measuring urinary metabolites or hemoglobin adducts derived from the GA pathway and comparing them to metabolites and hemoglobin adducts derived from AM directly. More specifically, we intended to evaluate urinary metabolites and hemoglobin adducts and to measure hormone levels after exposure to a known dose of AM. As a secondary and no less important objective, we intended to carefully monitor the health of people exposed to AM under these controlled conditions.
MATERIALS AND METHODS
Institutional Review Board approval.
This study was conducted in accordance with the Code of Federal Regulations (CFRs) governing Protection of Human Subjects (21 CFR 50), IRB (21 CFR 56), retention of data (21 CFR 312) as applicable and consistent with the Declaration of Helsinki. The administration of 13C AM to the study subjects was conducted at Covance Clinical Laboratories. Institutional Review Board approval of the protocol and the consent form was obtained at Covance Clinical Research Unit (CRU). Written informed consent was obtained from all study participants prior to study participation. Institutional Review Board approval was also obtained at RTI International, where the analysis of the samples was conducted. The study participants were compensated for time spent in the Covance Clinical Research Unit. The stipends were reviewed by the IRB as part of the study protocol, and those approved for this study were within the general stipend range approved by the IRB for other studies conducted at Covance CRU.
Chemicals.
(1,2,3-13C3) AM (CLM-813, lot number 11085, 99% labeled) was obtained from Cambridge Isotopes Limited. Identity and purity were confirmed by 1H and 13C NMR spectroscopy. (Note that the molecular weight of the (1,2,3-13C3) AM is 74, versus 71.08 for natural abundance AM). GA was synthesized by H2O2 oxidation of acrylonitrile (Payne and Williams, 1961), and stored at –20°C. N-(2-Carbamoylethyl)valine (AAVal), N-(2-carbamoylethyl)valine-13C5 (AAVal-13C5), N-(2-carbamoyl-2-hydroxyethyl) valine (GAVal), and N-(2-carbamoyl-2-hydroxyethyl)valine-13C5 (GAVal-13C5) were synthesized and purified as described previously by Fennell et al. (2003). The AAVal phenylthiohydantoin derivative (AAVal PTH), the corresponding 13C-labeled standard AAVal-13C5, GAVal PTH, and 13C-GAVal PTH standards were prepared as described by Fennell et al. (2003). AAVal-leu-anilide was obtained from Bachem Bioscience Inc. (King of Prussia, PA).
Human Study
AM exposure.
Twenty-four volunteers participated in this study. They were all male Caucasians (with the exception of one Native American) weighing between 71 and 101 kg and between 26 and 68 years of age. All volunteers were aspermic and had not used tobacco products for the past 6 months. They passed a drug screen and had not taken prescription drugs or caffeinated products over the previous 3 days. They had not consumed alcohol-containing beverages or medications within 7 days of study entry, and for the duration of the study. Each experimental group consisted of six individuals of whom one was administered a placebo. There were two phases to this study: an oral phase and a dermal phase.
A comprehensive physical exam was conducted on each individual upon check-in to the clinic, at 24 h after compound administration, and 7 days after check out. This exam included medical history, demographic data, neurological examination, 12-lead ECG, vital signs (including oral temperature, respiratory rate, and automated seated pulse and blood pressure), clinical laboratory evaluation (including clinical chemistry, hematology, and complete urinalysis). Each individual also had screens for HIV, hepatitis, and selected drugs of abuse and provided a semen sample to confirm aspermia. Additional ECG, neurological evaluation, an abbreviated physical examination, and subjective evaluations were conducted at 4 h after each AM administration.
In the oral phase, three groups of six people were administered 0.5, 1.0, or 3.0 mg/kg 13C3 AM. Individuals were presented with test substance at approximately 9:00 in the morning to initiate the study. Urine was collected at 0–2, 2–4, 4–8, 8–16, and 16–24 h. Blood was collected immediately prior to compound administration and 24 h later. Hormone blood samples (testosterone, LH, and prolactin levels) were drawn immediately prior to compound administration, 24 h later, and on the follow up visit on day 8.
In the dermal phase, a 50% solution of 13C3 AM was applied directly on the skin to a clean, dry, marked off, 24-cm2 (3 cm x 8 cm) area on the volar forearm. After applying the appropriate amount of material, the liquid was evaporated to dryness using a commercial hair dryer and covered with a sterile gauze pad. After drying the AM solution, the tape which had been used to demark the area of application was removed and placed in a vial containing 20 ml of water. The water (dermal dam solution) was analyzed for AM by HPLC. The site of application was covered with gauze for 24 h, at which time the gauze was removed and the area was washed with 1000 ml of water. The recovered wash water and gauze were analyzed by HPLC for AM. Dermal applications alternated between left and right arms, starting with the subject's dominant arm. Blood was collected immediately prior to compound administration and 24, 48, 72, and 96 h later (immediately prior to administration of the second and third doses, after gauze removal, and prior to leaving the clinic). Hormone blood samples were drawn immediately prior to compound administration, after 24 h, and on day 5 when the volunteers left the clinic.
Each exposure group contained six volunteers. Of the six volunteers in each group, five
Blood samples were processed for storage and shipment at the Covance Clinical Research Unit as follows: the blood samples were centrifuged, and plasma was removed; an equal volume of isotonic saline was added to the red cell pellets; the remaining washed red blood cells were washed by centrifugation with isotonic saline. The washing procedure was repeated a total of three times. The samples were stored frozen until shipped to RTI.
Urine samples were collected at intervals of 0–2, 2–4, 4–8, 8–16, and 16–24 h following administration of AM. The volume of urine in each sample was recorded, and sample aliquots were transferred to sample vials for storage.
Samples of urine and washed red blood cells were shipped to RTI from Covance Clinical Laboratories on dry ice, and were stored at –20°C until processed for analysis.
AM analysis.
The dose solutions, dermal dam solutions, and wash solutions provided by the Covance Clinical Research Unit were analyzed for the concentration of AM using a reversed-phase HPLC method. A calibration curve was prepared with AM over a concentration range of 5–200 μg/ml. Analysis was conducted with a Waters HPLC system consisting of two 515 pumps, a 717 autosampler, and an Applied Biosystems 759A UV detector. Data were recorded with a Waters Millennium data system. Chromatography was conducted on a Beckman Ultrasphere ODS column (4.5 mm x 25 cm) eluted at a flow rate of 1 ml/min with 100% water. Elution was monitored by measuring UV absorbance at 195 nm. The measurements of AM concentration in dose solutions were used to confirm the amount of AM administered in oral dose solutions, and in the dose solution used for dermal applications. The amounts of AM recovered from dermal application in both the dermal dam, which was used to outline the skin area for application, and the skin washings following removal of the gauze covering the application site were also measured. The total amount of AM recovered following dermal application was used to calculate the maximum amount of AM that was available for absorption (total dose applied – amount recovered in the dermal dam and washing solutions).
The AM recovered in the dermal dam and wash solutions would not be available systemically. The amount of dose that could have been taken up was calculated as:
(1)
Urinary metabolite analysis.
Metabolites of (1,2,3-13C3) AM in urine were analyzed from the group of volunteers exposed to 3 mg/kg AM orally by 13C NMR spectroscopy, essentially as described by Sumner et al.,(1992). No analyses were conducted on the samples from the volunteers administered 0.5 or 1.0 mg/kg orally, or those administered AM dermally, due to the insensitivity of the methodology. Individual samples were prepared for all urine samples from subject 015 (3 mg/kg), and these were analyzed individually. To reduce the amount of NMR instrument time required for analysis, composite urine samples were prepared for analysis of each subject administered 3.0 mg/kg orally. For each individual, sample aliquots of each time point were combined in the appropriate proportion (for each time point, added 10 ml x [volume of each time point/ total urine volume]) to make a total volume of 10 ml. Samples were concentrated by mixing 5.0 ml of pooled urine with 10 ml of methanol, and centrifuging at 5000 x g for 10 min. The supernatant was reduced in volume under a stream of nitrogen in a preweighed tube to approximately 300–600 μl. The weight of the concentrated urine was recorded. Water was added to make a total volume of 600 μl, and a solution of dioxane in D2O was added (200 μl).
NMR analysis of urinary metabolites.
Initial analysis of urine samples was conducted on a Bruker 500 MHz NMR spectrometer operating a 125 MHz for 13C. Quantitative analysis of metabolites was conducted on a Varian 500 MHz NMR spectrometer operating at 125 MHz. Samples were prepared by adding D2O, or D2O containing dioxane at a known concentration (200 μl) to an aliquot of a urine sample, a composite urine sample, or a concentrated composite urine sample (800 μl). Carbon-carbon connectivity was established using two-dimensional incredible natural abundance double quantum transfer spectra (INADEQUATE) using the Varian pulse sequence.
LC-MS/MS analysis of hemoglobin adducts.
Quantitative analysis of all of the human globin samples was conducted using an API-4000 LC/MS/MS system with a heated nebulizer source coupled to an Agilent 1100 HPLC system. Data was processed using Analyst version 1.3 software. The HPLC system was composed of a binary HPLC pump, a refrigerated vial/96 well plate autosampler, a photodiodearray detector, and a column heater.
Hemoglobin adduct analysis.
N-(2-Carbamoylethyl)valine (AAVal) and N-(2-carbamoyl-2-hydroxyethyl)valine (GAVal), formed by reaction of AM and GA, respectively, with the N-terminal valine residue in hemoglobin, were measured by an LC-MS/MS method. Globin was isolated from washed red cells (Mowrer et al., 1986). Samples were derivatized with phenylisothiocyanate in formamide to form adduct phenylthiohydantoin derivatives in a manner analogous to the modified Edman degradation (Bergmark, 1997; Perez et al., 1999; Trnqvist et al., 1986). Internal standards, AAValPTH-13C5 and GAVal PTH-13C5, were added, and the samples were extracted using a Waters Oasis HLB 3 cc (60 mg) extraction cartridge (Milford, MA). The samples were eluted with methanol, dried, and reconstituted in 100 μl of 50:50 MeOH:H2O (containing 0.1% formic acid). Analysis was conducted using an HP 1100 HPLC system interfaced to a PE Sciex API 4000 LC-MS with a Turboionspray interface. Chromatography was conducted on a Phenomenex Luna Phenyl-Hexyl Column (50 mm x 2 mm, 3 mm) eluted with 0.1% acetic acid in water and methanol at a flow rate of 350 μl/min, with a gradient of 45–55% methanol in 2.1 min. The elution of adducts was monitored by Multiple Reaction Monitoring (MRM) in the negative ion mode for the following ions:
Natural Abundance Analytes
AAVal-PTH: m/z 304 233 (M-H– M-H– – CH2-CH2-CONH2)
GAVal-PTH: m/z 320 233 (M-H– M-H– – CH2-CHOH-CONH2)
Adducts Derived from 1,2,3-13C AM
13C3-AAVal-PTH: m/z 307 233 (M-H– M-H– – 13CH2-13CH2-13CONH2)
13C3-GAVal-PTH: m/z 323 233 (M-H– M-H– – 13CH2-13CHOH-13 CONH2)
Internal Standards
AAVal-PTH-13C5: m/z 309 238 (M-H– M-H– – CH2-CH2-CONH2)
GAVal-PTH-13C5: m/z 325 238 (M-H– M-H– – CH2-CHOH-CONH2)
Quantitation of AAVal was conducted using the ratio of analyte to internal standard, with a calibration curve generated using AAVal-leu-anilide. Quantitation of GAVal was conducted using the ratio of analyte to internal standard.
Rat Study.
Prior to conducting the study in humans, four male Fischer 344 rats (body weights from 202 to 212 g) were administered (1,2,3-13C) AM at a dose of 3 mg/kg by gavage in distilled water. This component of the study was conducted in parallel with administration of 50 mg (1,2,3-13C) AM/kg to rats by gavage described in detail previously (Sumner et al., 2003). The rats were placed in metabolism cages for 24 h following dosing for collection of urine. After 24 h, the rats were euthanized, and blood samples were collected by cardiac puncture. Washed red blood cells were prepared by centrifugation, and globin was isolated by the method of Mowrer et al. (1986). Globin samples were analyzed by LC-MS/MS for AAVal, 13C3-AAVal, GAVal, and 13C3-GAVal as described by Sumner et al. (2003). Urine samples were analyzed by 13C NMR spectroscopy for metabolites of AM (Sumner et al., 2003). For quantitation of metabolites, samples were concentrated by addition of methanol to a 3-ml sample of urine, centrifugation, and evaporation under a stream of nitrogen. D2O and dioxane were added.
In vitro reaction rate constants.
Washed red cells from male F-344 rats, or from human blood, were lysed with an equal volume of distilled water and incubated with AM or GA at a concentration of 100 mM 37°C. Samples were removed at 0, 2, 5, 10, 15, 30, and 60 min, and the reactions were terminated by placing the tubes on ice and immediately adding 3 ml of 50 mM HCl in isopropanol. Globin was then isolated as described by Mowrer et al. (1986).
RESULTS
Rat Study
Analysis of urinary metabolites.
The analysis of urinary metabolites was conducted as described previously (Sumner et al., 2003), and the metabolites detected in rats and humans are shown in Figure 1. Prior to conducting the administration of AM in humans, the ability of the methods used to detect urinary metabolites and hemoglobin adducts was evaluated with a single dose of 3 mg/kg 13C3 AM administered by gavage in rats. The metabolites could be detected in urine samples. However, for quantitation of urinary metabolites, concentration of urine samples was required. The results of the quantitative analysis are shown in Table 1. Conjugation with GSH to form metabolite 1 accounted for 59% of the total metabolites. Metabolites 2,2' and 3,3', produced by GSH conjugation of GA, accounted for 25 and 16% of the urinary metabolites, respectively. GA was detected in urine samples from two of the rats prior to concentration, but was below the limit of quantitation in the concentrated samples. Metabolism via epoxidation to GA (2,2'-5, Table 1) accounted for approximately 41% of the urinary metabolites.
Analysis of the hemoglobin adducts from rats administered 3 mg/kg 13C3 AM is shown in Table 2. 13C3-AAVal and 13C3-GAVal were both increased by administration of AM. The ratio of 13C3-GAVal: 13C3-AAVal was 0.84 ± 0.07.
Human Study
Clinical findings.
No adverse events were reported in the oral phase of the study. With the dermal administration, one individual appeared to have a mild contact dermatitis, which is a known response to AM and was part of the informed consent. This individual was seen by a dermatologist who performed a skin biopsy which was consistent with a delayed hypersensitivity reaction. The skin reaction resolved 39 days after the first application of AM and 23 days after the reaction was manifested. Thus, the AM caused a delayed hypersensitivity reaction when placed on the skin, a reaction that took more than 3 weeks to resolve completely. An increase in the liver enzyme alanine aminotransferase (ALT) was observed above the upper limit of the reference range (normal) in four of the five individuals who
Dose administered.
For oral administration, AM was administered in a constant volume of 200 ml water to give the appropriate dose of 0.5, 1.0, or 3.0 mg/kg. The actual amount of AM administered was verified by analysis of aliquots of the dose solution with calculation of the amount of AM in 200 ml for each subject and the amount of AM administered per kg body weight. One individual from each dose group did not receive AM, and this was verified by analysis of the administered dose. The mean doses calculated were 0.44 ± 0.01, 0.92 ± 0.01, and 2.87 ± 0.03 mg/kg, and were 88, 92, and 96% of the nominal dose at 0.5, 1.0, and 3.0 mg/kg, respectively.
With dermal administration, the appropriate volume of a 50% solution of AM was applied to the skin. The dose applied each day based on HPLC analysis was calculated as 2.48 mg/kg, and was 83% of the nominal dose. Recovery of the dose after application was evaluated by HPLC. The tape used to demark the area of application (dermal dam solution) was found to contain between 9 and 54 mg of AM (Table 3). The amount of AM recovered by washing the application site with water after removal of the gauze covering ranged between 62 and 154 mg. The total AM per day recovered in dermal dam and wash solutions ranged between 85 and 190 mg and accounted for 36–86% of the applied AM. While considerable variability was observed in the recovery of AM, the mean recovery in dam and washes ranged between 65 and 71% of the total AM applied. The mean absorbed dose ranged from 0.73 to 0.86 mg/kg/day (Table 3). The cumulative absorbed dose (over the 3 days of administration) was 2.35 ± 0.50 mg/kg (Table 3). This value includes material that could be retained at the site of application and probably represents the maximum that could be absorbed rather than the actual amount absorbed.
Analysis of urinary metabolites.
Urine samples from a single individual administered 3 mg/kg orally were evaluated qualitatively by 13C NMR spectroscopy prior to quantitative analysis. A sample of each time point collection was analyzed from subject 15. The majority of the metabolite signals were found in the 2–4, 4–8, and 8–16 h samples. No signals indicative of the presence of 13C3 AM or its metabolites were found in the predose urine.
To achieve the necessary signal to noise ratio for quantitative analysis, and to collect data that were similar to those obtained from studies in rodents, aliquots of the urine samples from individuals were combined in the appropriate proportions to make a 24-h composite sample, which was then concentrated. Dioxane was added as an internal standard for quantitation of the metabolites.
The 13C NMR spectrum of a composite urine sample obtained from a volunteer who did not receive AM is shown in Supplementary Data. A number of signals from endogenous metabolites, including urea at 162.5 ppm were seen. In the samples of urine from volunteers administered 3.0 mg/kg (1,2,3-13C3) AM, there are additional signals arising from AM and its metabolites (Supplementary Data). These signals showed characteristic multiplets arising from coupling with adjacent labeled carbon atoms. Many but not all of the signals present in urine samples from rats and mice administered (1,2,3-13C3) AM were present in the human urine samples. Some additional signals that had not been previously observed were also present. These signals are presented in Table 4. The nomenclature used previously (Sumner et al., 1992, 1997, 2003) is also used to describe the metabolites in this paper. The major metabolites present in all of the composite urine samples were derived from direct glutathione conjugation of AM (Sumner et al., 1992, 1997, 2003).
The major metabolites, 1 and 1', corresponding to N-acetyl-S-(3-amino-3-oxopropyl)cysteine and S-(3-amino-3-oxopropyl)cysteine, showed signals at 27 ppm (doublet), 34.7 ppm (2 doublets of doublets), and at 177 ppm (2 doublets). Signals associated with GA, metabolite 4, were observed at 46.9 ppm (4a), and at 48.5 ppm (4b). Signals from the hydrolysis product of GA (2,3-dihydroxypropionamide) were observed at 62.99 ppm (5a, doublet), at 72 ppm (5,5'b, doublet of doublets), and at 175.5 ppm (5,5'c, doublet). Low-intensity signals that may be due to metabolite 2 (N-acetyl-S-(3-amino-2-hydroxy-3-oxopropyl)cysteine) were observed at 35.8 ppm and at 70.2 ppm. Signals that are associated with metabolite 3 (N-acetyl-S-(1-carbamoyl-2-hydroxyethyl)cysteine) were not observed. A metabolite that had not been previously observed in rats and mice gave a signal at 46.47 ppm (doublet, J = 37 Hz). An INADEQUATE spectrum was used to establish the carbon–carbon connectivity of the main signals of metabolite 1, GA, and glyceramide. In addition, the connectivity of the doublet at 46.47 ppm (6b) was established with a complex doublet of doublets at 27.60 ppm (6a), and a doublet at 175.4 ppm (6c). These signals have been assigned to the labeled carbon atoms of N-acetyl-S-(3-amino-3-oxopropyl)cysteine-S-oxide, based on the comparison with a synthesized standard (to be reported in an additional publication).
Quantitation of the metabolites present was conducted using the integral of the metabolite signals and dioxane added as internal standard. Spectra for quantitation were acquired with decoupling only during acquisition to ensure that nuclear Overhauser enhancement was minimized so that accurate quantitation could be conducted. This resulted in the decrease in intensity of some of the low-intensity signals to the point were quantitation was not readily possible (e.g., with metabolite 2). Although signals for AM were readily apparent, these were not quantitated because of the long relaxation time for the signals of AM. A summary of the quantitative information for urinary metabolites is presented in Table 5. Approximately 34% of the administered dose of AM was recovered in the total urinary metabolites within 24 h of administration. Metabolite 1 accounted for approximately 72% of the metabolites excreted. The sulfoxide derived from metabolite 1 accounted for approximately 14% of the metabolites measured. Metabolites that are known to be derived from GA (4 and 5) represented approximately 14% of the metabolites.
Analysis of hemoglobin adducts.
Analysis of AAVal and GAVal was set up and validated as described previously (Fennell et al., 2003). A standard curve was developed with AAVal-leu-anilide standard and 13C5-AAVal PTH. GAVal was measured based on the ratio of analyte to added 13C5 GAVal PTH. In each of the samples obtained prior to administration of (1,2,3-13C3) AM, measurable adduct backgrounds for AAVal and GAVal were detected.
Typical chromatograms for AAVal and GAVal from a volunteer administered 13C3 AM orally (0.5 mg/kg) are shown in Supplementary Data (both before and following administration). The administration of (1,2,3-13C3) AM resulted in an increase in the peak height and area for 13C3-AAVal and 13C3-GAVal. The mean values for hemoglobin adduct levels from the various groups administered AM orally are presented in Table 6. AAVal, GAVal, 13C3-AAVal, and 13C3-GAVal were measured for each individual prior to exposure to 13C3 AM and at 24 h following administration. The majority of the individual values (not shown) for AAVal prior to exposure were in the range of 40–200 fmol/mg globin. Most of the values for AAVal measured before and after exposure were similar. One exception to this was noted. One volunteer had high levels of AAVal in the first sample (986 fmol/mg), which dropped in the second sample (43 fmol/mg). The reason for this discrepancy cannot be explained, and the value for the pre-exposure sample has been excluded from calculations of statistical parameters. Prior to administration of AM, the levels of GAVal were in the range of 16–67 fmol/mg globin, and peaks associated with 13C3-AAVal and 13C3-GAVal were not detected. Following administration of AM, 13C3-AAVal and 13C3-GAVal adduct levels increased in five out of the six members of each group. The subjects who
On oral administration of (1,2,3-13C3) AM, levels of 13C3-AAVal increased in a dose-dependent manner (Table 6). A plot of 13C3-AAVal versus the nominal dose administered was linear (Fig. 2). Similarly, on oral administration of (1,2,3-13C3) AM, levels of 13C3-GAVal increased in a dose-dependent manner (Table 6). A plot of 13C3-GAVal versus the nominal dose administered was also linear (Fig. 2). The levels of 13C3-GAVal formed were considerably lower than those of 13C3-AAVal, with a ratio of 13C3-GAVal: 13C3-AAVal of ranging from 0.36 to 0.44 (Table 6).
Following dermal administration of AM (days 1, 2, and 3), both AAVal and GAVal increased in a linear manner after each dose, on days 2, 3, and 4 (Table 6). Little change was noted between days 4 and 5 when no administration took place. The relationship between 13C3-AAVal, 13C3-GAVal, and the cumulative daily exposure was linear (not shown). The ratio of 13C3-GAVal:13C3-AAVal ranged from 0.48 to 0.68.
To enable the comparison of dose groups and routes of exposure, the data for hemoglobin adducts have been normalized by the actual dose estimated from analysis of AM administered and recovered (Table 7). The amount of 13C3-AAVal and 13C3-GAVal formed following dermal exposure was considerably lower than that observed on oral administration, based on the dermal Administered Dose. Calculation of the dermal dose, taking into consideration the AM recovered in the dermal washes (Absorbed Dose) resulted in an increase in the amount of AAVal and GAVal formed per unit dose, but the yield from dermal administration was still considerably lower than that with oral administration (12.7 vs 74.7 nmol AAVal/g globin/mmol AM/kg). Hemoglobin adducts formed in humans on oral administration have been compared with those formed in rats on gavage administration (Table 7). Adduct levels normalized for dose are approximately 3-fold higher in humans for AAVal and 1.7-fold higher for GAVal in humans. The ratio of GAVal:AAVal in humans on oral administration was similar to that previously reported (Fennell et al., 2003) for rats administered AM by gavage at 50 mg/kg (Table 8). On dermal administration in humans, the ratio of GAVal:AAVal was slightly increased compared with oral administration (0.57 vs. 0.39); the magnitude of the increase was much less than that observed in rats (1.7 vs. 0.38).
Hemoglobin adducts can be used to calculate the internal dose or area under the curve in blood, using the reaction rate constant measured in vitro (Ehrenberg and Osterman-Golkar, 1980), using the relationship:
(2)
where k, the second order reaction rate constant, is expressed in units of l/g globin/h, and [RHb] is the adduct concentration, and [Hb] is the concentration of hemoglobin. In Table 8, the rate constants measured in this laboratory are presented. These are similar in magnitude to those reported previously by Bergmark et al. (1993). For the human AAVal rate constant, our value of 4.27 x 10–6 l/g globin/h agrees well with the previously reported value of 4.4 x 10–6 l/g globin/h. However, our values for GA reaction with rat and human hemoglobin are approximately half of those reported previously. It is possible that the differences in the method of analysis could account for this difference. The calibration of AAVal measurements conducted here used commercially available AAVal-leu-anilide together with AAVal-13C5PTH. The agreement between values calculated with a standard curve and those calculated from the ratio of analyte:internal standard were within 1%. However, GAVal-leu-anilide was not commercially available at the time that these samples were analyzed, and standardization was conducted by the ratio of analyte:internal standard alone. However, using the rate constant to calculate the AUC for AM and GA from the adduct data, with the rate constants measured under the same conditions should enable comparison of data with other laboratories in the future. Calculated values for AAVal and GAVal were converted to "Dose" or AUC normalized for administered or absorbed dose (Table 9). Compared with the area under the curve (AUC) calculated for AM in rats administered 3 mg/kg, the AUC in humans ranged from 2.75- to 3.7-fold higher. In contrast, the AUC for GA in humans was similar to that in the rat, ranging from approximately 1.2 to 1.4 times that of rat.
DISCUSSION
This study was undertaken to understand the metabolism of AM in humans, and the relationship between hemoglobin adducts of AM and GA and exposure to AM via the oral and dermal routes. During the design of the study, the finding of AM in the foods had not been disclosed, and the main impetus behind the study was to understand the fate of AM in scenarios that could occur with occupational exposure. The controlled exposure of human subjects was conducted in a dose-escalating manner with careful monitoring of the subjects.
The administration of a low dose (3 mg/kg) of AM by gavage to rats resulted in a greater amount of metabolism via GA (41% of the urinary metabolites) compared with a higher dose of 59 mg/kg (28% of the urinary metabolites, (Sumner et al., 2003)). The fate of GA was primarily conjugation with GSH, resulting in the excretion of two mercapturic acids (metabolites 2 and 3). The total amount of AM metabolites recovered by 24 h after dosing (50%) is similar to that reported by Miller et al. (1982), and by Kadry et al. (1999).
Hemoglobin adducts represent useful dosimeters for reactive chemicals and metabolites, with the extent of adduct formation related to the concentration of reactant integrated over time. The ratio of GAVal:AAVal has been observed to differ with route of exposure (Sumner et al., 2003). This should reflect differences in the relative AUCs for AM and GA in blood (Calleman, 1996; Calleman et al., 1992). At a dose of 3 mg/kg in rats, GAVal:AAVal was higher (0.84) than that observed at 50 mg/kg (0.38, (Fennell et al., 2003)). Both this observation and the larger percentage of metabolites in urine derived from GA at the lower dose are consistent with saturation of the epoxidation of AM at the higher dose administered (Calleman et al., 1992).
The urinary metabolites of AM in humans showed similarities and differences with data obtained previously in the rat and mouse. The main pathway of metabolism in humans was via direct glutathione conjugation, forming N-acetyl-S-(3-amino-3-oxopropyl)cysteine, as observed in the rat and mouse, and its S-oxide, which has not been reported previously. Epoxidation to GA was the other important pathway, with glyceramide formed as a major metabolite in humans. GA was detected in low amounts. The glutathione conjugation of GA, which is a major pathway in rodents, appeared to occur at very low levels in humans, with metabolite 2 detected, but not quantitated, and metabolite 3 not detected. Metabolism via GA (derived from GA and glyceramide) in humans was approximately 12% of the total urinary metabolites. This is considerably lower than the amount of GA derived metabolites reported for oral administration of AM in rats (28% at 50 mg/kg, (Sumner et al., 2003) and in mice (59% at 50 mg/kg, (Sumner et al., 1992)).
This study has provided data on the amount of hemoglobin adducts derived from AM and GA following administration of a defined dose of AM to people. Both AAVal and GAVal increased linearly with increasing dose of AM administered orally, suggesting that, over the range of 0.5–3.0 mg/kg, there is no saturation of metabolism of AM to GA. The ratio of GAVal:AAVal produced by administration of AM was similar to the ratio of the background adducts prior to exposure. Compared with the equivalent oral administration in rats (3 mg/kg), the ratio of 13C3-GAVal: 13C3-AAVal in humans was lower (0.44 ± 0.06) than in rats (0.84 ± 0.07), and the absolute amount of 13C3-AAVal formed in humans was approximately 2.7-fold higher than in the rat. The absolute amount of 13C3-GAVal was approximately 1.4-fold higher than that formed in the rat.
In risk assessment, the extrapolation of dose between species is generally conducted using a scaling factor of body weight3/4 (U.S. General Accounting Office, 2001). Thus, for extrapolation from rats to humans, a scaling factor of approximately 4 would be used. For effects that are mediated via the action of AM, based on the AAVal comparison between rats and humans, a factor of 2.7 would appear appropriate, whereas for effects mediated by GA, a factor of 1.4 would appear appropriate.
The average AAVal and GAVal formed on dermal administration were calculated from the values in Table 7. Dermal administration of AM resulted in much lower levels of AAVal and GAVal formed compared with an equivalent dose by the oral route. Comparing AAVal after dermal (Table 7, 4.9 nmol/g globin/mmol AM/kg) and oral administration (Table 7, 74.7 nmol/g globin/mmol AM/kg) indicated approximately 6.6% of the dermally administered dose (Table 3) was taken up, assuming 100% oral absorption. Similarly, dermal administration also resulted in much lower formation of GAVal (9.7% of that formed on oral administration). Approximately 66% of the administered dose of AM was recovered in dam solutions and the wash solutions (data not shown), and thus was not systemically absorbed on dermal administration. This suggests that only approximately 33% of the dermally applied dose could have been absorbed (Absorbed Dose, Table 3). The AAVal formed on dermal administration normalized with the absorbed dose (Table 3) was only 17.0% of that expected (Table 7, 12.7 nmol/g globin/mmol AM/kg for dermal vs. 74.7 nmol/g globin/mmol AM/kg for oral). This may indicate that 83% of the AM that penetrated the skin was not available systemically. GAVal on dermal administration normalized for the absorbed dose in a similar manner was 25.3% of that formed on oral administration (Table 7, 7.3 pmol/g globin/mmol AM/kg for dermal versus 28.9 pmol/g globin/mmol AM/kg for oral). An alternative explanation is that the AM and GA may be more rapidly metabolized on dermal exposure, resulting in a lower AUC and lower adduct formation per mg/kg. This may be clarified by analysis of the urinary metabolites in the dermally exposed individuals.
An important result from this study is the capability to define the amount of AAVal adduct expected from exposure to AM via different routes. The expected amount of adduct that would accumulate from continuous exposure can be calculated from a knowledge of the amount of adduct formed per day of exposure, and from the kinetics of the erythrocyte (Fennell et al., 1992; Osterman-Golkar et al., 1976). Exposure via oral intake to 1 μg AM/kg (1.05 fmol AAVal/mg globin/day) for the lifespan of the erythrocyte (120 days) would result in the accumulation of adducts to 63 fmol/mg globin. Daily dermal exposure to 1 μg AM/kg (0.18 fmol AAVal/mg globin/day) for the lifespan of the erythrocyte (120 days) would result in the accumulation of adducts to 10.8 fmol AAVal/mg globin. With workplace exposure of 5 days per week, this would decrease to approximately 7.8 fmol AAVal/mg globin.
Exposure to AM can occur through occupational exposure, via ingestion in food, and via cigarette smoking (Bergmark, 1997; Bergmark et al., 1993; Tareke et al., 2002). Occupational exposure can occur via inhalation of AM vapor and dust and via the dermal route. The adduct yield for a given exposure differs between oral and dermal administration in rodents (Sumner et al., 2003) and in humans (this study). The contribution of AM exposure via diet and cigarette smoking will complicate the assessment of low level workplace exposure via adduct measurement. However, different studies have described AAVal measured in occupational settings ranging from 300 to 34,000 fmol/mg (Calleman et al., 1994), up to 17,700 fmol/mg (Hagmar et al., 2001), and 71–1854 fmol/mg in a plant that was in production for a period of 2 weeks (Perez et al., 1999). Lifestyle and dietary exposures appear to cause much lower levels of adducts.
Using the data obtained for hemoglobin adducts from the oral administration of AM, it is possible to estimate the amount of AM taken in from the diet over the lifespan of the erythrocyte, and compare this with the estimates of AM in the diet. A detailed consideration of the dietary exposure to AM and comparison with the dietary exposure estimates reported previously (Svensson et al., 2003; Tareke et al., 2002; Trnqvist et al., 1998; WHO, 2002) is included in the Appendix.
In summary, this study has examined the metabolism of AM in people and compared the internal dose of AM and GA in people with that observed in rats. The data reported are consistent with slower elimination of AM in humans and slower metabolism of AM to GA in humans. The hemoglobin adduct measurements obtained will provide a calibration for estimates of exposure from diet, lifestyle, and the workplace.
Appendix
Estimates of Dietary Intake from AAVal Measurements
The normalized formation of AAVal per unit dose of unlabeled AM, averaged for all of the oral dose groups, was calculated as 1050 fmol/mg globin/mg AM/kg. Using this value, the amount of AM exposure that would be expected in a human from the diet can be calculated. In this study, the average pre-exposure AAVal level was 79 ± 49 fmol/mg globin (excluding subject 13). Hagmar et al. (2001) reported a range of 0.02–0.07 nmol AAVal/g globin (equivalent to 20–70 fmol AAVal/mg globin) in unexposed individuals. The steady state adduct level from continuous exposure is calculated to be a.ter/2, where a is the daily adduct increment, and ter is the erythrocyte lifespan, which in humans is approximately 120 days. The amount of adduct formed per day, assuming similar exposure per day over the erythrocyte lifespan, would be 1/60th of the daily adduct increment. For the average level of 80 fmol/mg:
(3)
The amount of AM taken in can be estimated by:
(4)
Similarly for the lower level of 20 fmol/mg, the daily adduct increment is 0.33 fmol/mg globin/day, and AM intake is 0.31 μg/kg/day.
These estimates suggest a daily intake in the range of 22–88 μg for an average 70-kg person and can be compared with a variety of estimates that have been produced for the average daily intake of AM. The World Health Organization consultation (2002) estimated a daily intake of 0.8 μg AM/kg/day for the average consumer, based on AM levels in foods. Similarly, (Svensson et al., 2003) estimated a mean daily intake of 31 μg/day based on food consumption data, which corresponds to 0.44 μg/kg/day for a 70-kg person. Trnqvist et al. (1998) have estimated a daily intake of approximately 100 μg per person based on hemoglobin adduct measurements of 30 fmol/mg globin, reported in Tareke et al. (2002). These calculations are based on a complex equation that relates the estimated dose to adduct level accumulated over the erythrocyte lifespan, rate of elimination (k), the reaction rate constant for adduct formation, the volume of distribution, and the erythrocyte lifespan (ter).
(5)
Of the parameters in this equation, two are estimated: the elimination rate constant in humans (Calleman, 1996), and the volume of distribution (assumed to be 1). Using the data obtained in this study for a single exposure in a modified equation enables the estimation of possible values for the elimination rate constant and volume of distribution.
(6)
Rearranging and solving for VD yields a value of 0.38 l/kg, or distribution into 27 liters in a 70-kg person. This is approximately 64% of the estimated 42 liters of total body water. Using the value for VD of 0.38 l/kg and an adduct level of 30 pmol adduct/g globin, an uptake of AM of 0.48 μg/kg/day is calculated, or 33 μg /day for a 70-kg person. This estimated uptake is more in line with the estimates of AM exposure based on food consumption.
SUPPLEMENTARY DATA
Supplementary data are available at Toxicological Sciences online.
NOTES
The authors certify that all research involving human subjects was done under full compliance with all government policies and the Helsinki Declaration.
ACKNOWLEDGMENTS
The authors would like to acknowledge CIIT Centers for Health Research, Research Triangle Park, NC, where the administration of AM to rats and analyses of rat urine and globin were conducted.
This study was funded by SNF SA, which is a manufacturer of acrylamide and polyacrylamide.
REFERENCES
Bergmark, E. (1997). Hemoglobin adducts of acrylamide and acrylonitrile in laboratory workers, smokers and nonsmokers. Chem. Res. Toxicol. 10, 78–84.
Bergmark, E., Calleman, C. J., He, F., and Costa, L. G. (1993). Determination of hemoglobin adducts in humans occupationally exposed to acrylamide. Toxicol. Appl. Pharmacol. 120, 45–54.
Besaratinia, A., and Pfeifer, G. P. (2004). Genotoxicity of acrylamide and glycidamide. J. Natl. Cancer Inst. 96, 1023–1029.
Calleman, C. J. (1996). The metabolism and pharmacokinetics of acrylamide: Implications for mechanisms of toxicity and human risk estimation. Drug Metab. Rev. 28, 527–590.
Calleman, C. J., Bergmark, E., and Costa, L. G. (1990). Acrylamide is metabolized to glycidamide in the rat: Evidence from hemoglobin adduct formation. Chem. Res. Toxicol. 3, 406–412.
Calleman, C. J., Stern, L. G., Bergmark, E., and Costa, L. G. (1992). Linear versus nonlinear models for hemoglobin adduct formation by acrylamide and its metabolite glycidamide: Implications for risk estimation. Cancer Epidemiol. Biomarkers Prev. 1, 361–368.
Calleman, C. J., Wu, Y., He, F., Tian, G., Bergmark, E., Zhang, S., Deng, H., Wang, Y., Crofton, K. M., Fennell, T., et al.(1994). Relationships between biomarkers of exposure and neurological effects in a group of workers exposed to acrylamide. Toxicol. Appl. Pharmacol. 126, 361–371.
Dixit, R., Seth, P. K., and Mukjtar, H. (1982). Metabolism of acrylamide into urinary mercapturic acid and cysteine conjugates in rats. Drug Metab. Dispos. 10, 196–197.
Dybing, E., and Sanner, T. (2003). Risk assessment of acrylamide in foods. Toxicol. Sci. 75, 7–15.
Edwards, P. M. (1975). The distribution and metabolism of acrylamide and its neurotoxic analogues in rats. Biochem. Pharmacol. 24, 1277–1282.
Ehrenberg, L., and Osterman-Golkar, S. (1980). Alkylation of macromolecules for detecting mutagenic agents. Teratog. Carcinog. Mutagen 1, 105–127.
European Union (2002). European Union Risk Assessment Report: Acrylamide. Luxembourg. 210 p.
Fennell, T. R., Snyder, R. W., Krol, W. L., and Sumner, S. C. (2003). Comparison of the hemoglobin adducts formed by administration of N-methylolacrylamide and acrylamide to rats. Toxicol Sci 71, 164–175.
Fennell, T. R., Sumner, S. C., and Walker, V. E. (1992). A model for the formation and removal of hemoglobin adducts. Cancer Epidemiol. Biomarkers Prev. 1, 213–219.
Friedman, M. A., Dulak, L. H., and Stedham, M. A. (1995). A lifetime oncogenicity study in rats with acrylamide. Fundam. Appl. Toxicol. 27, 95–105.
Gamboa da Costa, G., Churchwell, M. I., Hamilton, L. P., Von Tungeln, L. S., Beland, F. A., Marques, M. M., and Doerge, D. R.(2003). DNA adduct formation from acrylamide via conversion to glycidamide in adult and neonatal mice. Chem. Res. Toxicol. 16, 1328–1337.
Hagmar, L., Trnqvist, M., Nordander, C., Rosen, I., Bruze, M., Kautiainen, A., Magnusson, A. L., Malmberg, B., Aprea, P., Granath, F., et al.(2001). Health effects of occupational exposure to acrylamide using hemoglobin adducts as biomarkers of internal dose. Scand. J. Work Environ. Health 27, 219–226.
Hashimoto, K., and Aldridge, W. N. (1970). Biochemical studies on acrylamide, a neurotoxic agent. Biochem. Pharmacol. 19, 2591–2604.
Hashimoto, K., and Tanii, H. (1985). Mutagenicity of acrylamide and its analogues in Salmonella typhimurium. Mutat. Res. 158, 129–133.
Johnson, K. A., Gorzinski, S. J., Bodner, K. M., Campbell, R. A., Wolf, C. H., Friedman, M. A., and Mast, R. W. (1986). Chronic toxicity and oncogenicity study on acrylamide incorporated in the drinking water of Fischer 344 rats. Toxicol. Appl. Pharmacol. 85, 154–168.
Kadry, A. M., Friedman, M. A., and Abdel Rahman, M. S. (1999). Pharmacokinetics of acrylamide after oral administration in male rats. Environ. Toxicol. Pharmacol. 7, 127–133.
Lafferty, J. S., Kamendulis, L. M., Kaster, J., Jiang, J., and Klaunig, J. E. (2004). Subchronic acrylamide treatment induces a tissue-specific increase in DNA synthesis in the rat. Toxicol. Lett. 154, 95–103.
Miller, M. J., Carter, D. E., and Sipes, I. G. (1982). Pharmacokinetics of acrylamide in Fisher-344 rats. Toxicol. Appl. Pharmacol. 63, 36–44.
Mowrer, J., Trnqvist, M., Jensen, S., and Ehrenberg, L. (1986). Modified Edman Degradation applied to hemoglobin for monitoring occupational exposure to alkylating agents. Toxicol. Env. Chem. 11, 215–231.
Osterman-Golkar, S., Ehrenberg, L., Segerback, D., and Hallstrom, I. (1976). Evaluation of genetic risks of alkylating agents. II. Haemoglobin as a dose monitor. Mutat. Res. 34, 1–10.
Payne, G. B., and Williams, P. H. (1961). Reactions of Hydrogen Peroxide. VI. Alkaline epoxidation of acrylonitrile. J. Org. Chem. 26, 651–659.
Perez, H. L., Cheong, H. K., Yang, J. S., and Osterman-Golkar, S. (1999). Simultaneous analysis of hemoglobin adducts of acrylamide and glycidamide by gas chromatography-mass spectrometry. Anal. Biochem. 274, 59–68.
Rosen, J., and Hellenas, K. E. (2002). Analysis of acrylamide in cooked foods by liquid chromatography tandem mass spectrometry. Analyst 127, 880–882.
Segerback, D., Calleman, C. J., Schroeder, J. L., Costa, L. G., and Faustman, E. M. (1995). Formation of N-7-(2-carbamoyl-2-hydroxyethyl)guanine in DNA of the mouse and the rat following intraperitoneal administration of [14C]acrylamide. Carcinogenesis 16, 1161–1165.
Spencer, P. S., and Schaumburg, H. H. (1974a). A review of acrylamide neurotoxicity. Part I. Properties, uses and human exposure. Can. J. Neurol. Sci. 1, 143–150.
Spencer, P. S., and Schaumburg, H. H. (1974b). A review of acrylamide neurotoxicity. Part II. Experimental animal neurotoxicity and pathologic mechanisms. Can. J. Neurol. Sci. 1, 152–169.
Spencer, P. S., and Schaumburg, H. H. (1975). Nervous system degeneration produced by acrylamide monomer. Environ. Health Perspect. 11, 129–133.
Sumner, S. C., Fennell, T. R., Moore, T. A., Chanas, B., Gonzalez, F., and Ghanayem, B. I. (1999). Role of cytochrome P450 2E1 in the metabolism of acrylamide and acrylonitrile in mice. Chem. Res. Toxicol. 12, 1110–1116.
Sumner, S. C., MacNeela, J. P., and Fennell, T. R. (1992). Characterization and quantitation of urinary metabolites of [1,2,3-13C]acrylamide in rats and mice using 13C nuclear magnetic resonance spectroscopy. Chem. Res. Toxicol. 5, 81–89.
Sumner, S. C., Selvaraj, L., Nauhaus, S. K., and Fennell, T. R. (1997). Urinary metabolites from F344 rats and B6C3F1 mice coadministered acrylamide and acrylonitrile for 1 or 5 days. Chem. Res. Toxicol. 10, 1152–1160.
Sumner, S. C., Williams, C. C., Snyder, R. W., Krol, W. L., Asgharian, B., and Fennell, T. R. (2003). Acrylamide: A comparison of metabolism and hemoglobin adducts in rodents following dermal, intraperitoneal, oral, or inhalation exposure. Toxicol. Sci. 75, 260–270.
Svensson, K., Abramsson, L., Becker, W., Glynn, A., Hellenas, K. E., Lind, Y., and Rosen, J. (2003). Dietary intake of acrylamide in Sweden. Food Chem. Toxicol. 41, 1581–1586.
Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S., and Trnqvist, M. (2002). Analysis of acrylamide, a carcinogen formed in heated foodstuffs. J. Agric. Food Chem. 50, 4998–5006.
Trnqvist, M., Bergmark, E., Ehrenberg, L., and Granath, F. (1998). Risk Assessment of Acrylamide. Swedish Chemicals Inspectorate, Solna, Sweden.
Trnqvist, M., Mowrer, J., Jensen, S., and Ehrenberg, L. (1986). Monitoring of environmental cancer initiators through hemoglobin adducts by a modified Edman degradation method. Anal. Biochem. 154, 255–266.
Tyl, R. W., and Friedman, M. A. (2003). Effects of acrylamide on rodent reproductive performance. Reprod. Toxicol. 17, 1–13.
U.S. General Accounting Office (2001). Chemical Risk Assessment, Selected Federal Agencies' Procedures, Assumptions, and Policies. United States General Accounting Office, Washington, DC.
WHO (2002). Joint FAO/WHO Consultation on the Health Implication of Acrylamide in Food. World Health Organization, Geneva, Switzerland. 33 p.(Timothy R. Fennell, Susan)
Covance Clinical Research Unit, Inc., Madison, Wisconsin 53703
University of Medicine and Dentistry of New Jersey, Newark, New Jersey 07103
ABSTRACT
Acrylamide (AM), used in the manufacture of polyacrylamide and grouting agents, is produced during the cooking of foods. Workplace exposure to AM can occur through the dermal and inhalation routes. The objectives of this study were to evaluate the metabolism of AM in humans following oral administration, to compare hemoglobin adduct formation on oral and dermal administration, and to measure hormone levels. The health of the people exposed under controlled conditions was continually monitored. Prior to conducting exposures in humans, a low-dose study was conducted in rats administered 3 mg/kg (1,2,3-13C3) AM by gavage. The study protocol was reviewed and approved by Institute Review Boards both at RTI, which performed the sample analysis, and the clinical research center conducting the study. (1,2,3-13C3) AM was administered in an aqueous solution orally (single dose of 0.5, 1.0, or 3.0 mg/kg) or dermally (three daily doses of 3.0 mg/kg) to sterile male volunteers. Urine samples (3 mg/kg oral dose) were analyzed for AM metabolites using 13C NMR spectroscopy. Approximately 86% of the urinary metabolites were derived from GSH conjugation and excreted as N-acetyl-S-(3-amino-3-oxopropyl)cysteine and its S-oxide. Glycidamide, glyceramide, and low levels of N-acetyl-S-(3-amino-2-hydroxy-3-oxopropyl)cysteine were detected in urine. On oral administration, a linear dose response was observed for N-(2-carbamoylethyl)valine (AAVal) and N-(2-carbamoyl-2-hydroxyethyl)valine (GAVal) in hemoglobin. Dermal administration resulted in lower levels of AAVal and GAVal. This study indicated that humans metabolize AM via glycidamide to a lesser extent than rodents, and dermal uptake was approximately 6.6% of that observed with oral uptake.
Key Words: acrylamide; glycidamide; hemoglobin; adducts.
INTRODUCTION
Acrylamide (AM) is used in the manufacture of water-soluble polymers (European Union, 2002). These polymers are then used for wastewater and sludge treatment, paper manufacture, soil stabilization, mining, and many other uses (European Union, 2002). AM is also a chemical intermediate in the manufacture of other monomeric chemicals and is used for grouting and preparation of laboratory gels for electrophoresis. Human exposure through these applications is very small, with the largest mean estimated exposures of 0.07 mg/kg/day thought to arise during manufacture of AM (European Union, 2002). Previously, it has been postulated that dermal absorption was the major route of human exposure to AM (European Union, 2002). The magnitude of this dermal absorption is highly relevant, as one of the uses of AM-based polymers is in the formulation of skin creams (European Union, 2002). Estimates of dermal absorption based on in vitro and rodent studies have ranged from 3 to 100% (European Union, 2002). Recently, exposure to AM in a variety of cooked foods has been described (Rosen and Hellenas, 2002; Tareke et al., 2002). Human exposure via this route is substantial, with estimated exposures as high as 70 μg per day proposed (Tareke et al., 2002).
AM is metabolized by two main pathways: glutathione conjugation (Dixit et al., 1982; Edwards, 1975; Hashimoto and Aldridge, 1970; Miller et al., 1982; Sumner et al., 1992), and epoxidation to glycidamide (GA) (Calleman et al., 1990; Sumner et al., 1992). The metabolism of AM in vivo results in the formation of a number of metabolites. The metabolism of AM in vivo has been investigated by administration of (1,2,3-13C3) AM to rodents, with the detection and quantitation of metabolites by 13C NMR spectroscopy (Sumner et al., 1992, 1999, 2003). The epoxidation reaction to GA is catalyzed by cytochrome P450 2E1 in rodents (Sumner et al., 1999). Both AM and GA react with hemoglobin, producing a stable adduct which can be measured as an indicator of exposure. Correlations have been made with hemoglobin adducts and neurotoxicity, but there has been no systematic standardization of hemoglobin adducts with dose. GA is mutagenic in the Salmonella test (Hashimoto and Tanii, 1985) and in human bronchial epithelial cells and Big Blue mouse embryonic fibroblasts (Besaratinia and Pfeifer, 2004). It can react with DNA in vitro to produce a guanine derivative N7-(2-carbamoyl-2-hydroxyethyl)guanine, and two adenine derivatives, N3-(2-carbamoyl-2-hydroxyethyl)adenine and N1-(2-carboxy-2-hydroxyethyl)-2'-deoxyadenosine (Gamboa da Costa et al., 2003; Segerback et al., 1995). In vivo, administration of AM to rats and mice produces low levels of N7-(2-carbamoyl-2-hydroxyethyl)guanine and N3-(2-carbamoyl-2-hydroxyethyl)adenine (Gamboa da Costa et al., 2003; Segerback et al., 1995).
AM induces a characteristic peripheral neurotoxicity in animals and man (Spencer and Schaumburg, 1974a,b, 1975). This toxicity manifests itself as a distal to proximal loss of nerve function and dying back of cells. AM also affects rodent reproduction, namely smaller litter size. At elevated AM doses other reproductive effects are seen, likely as a consequence of dominant lethal mutations at low doses and neurotoxicity at higher doses (Tyl and Friedman, 2003).
AM is carcinogenic in drinking water studies in laboratory rats (Friedman et al., 1995; Johnson et al., 1986). In male rats, it induces tumors of the tunica vaginalis testes and the thyroid, while in females, it induces mammary fibroadenomas and thyroid tumors (Friedman et al., 1995). The mechanism for this tumorigenicity is unclear, although interaction with the dopamine receptor has been postulated, as well as genotoxicity (Tyl and Friedman, 2003). Increases in DNA synthesis have been reported after subacute exposure in target tissues for tumor development (thyroid, testicular mesothelium, and adrenal medulla) and have been suggested to play a role in the carcinogenicity of AM (Lafferty et al., 2004). Understanding the mechanism of tumorigenicity is important, since conventional risk assessment techniques place the order of magnitude of the risk at approximately 10–3 with exposures of 70 μg/ day (Dybing and Sanner, 2003).
The relative contributions of AM and GA in the mode of action of AM are the subject of debate and current research. The conversion of AM to GA and differences that may occur between species, exposure route, and dose are important considerations in assessing the risk of the possible effects of AM exposures in the diet, in consumer products, and in the workplace.
The primary objectives of this study were to evaluate the conversion of AM to GA in people exposed to AM and to evaluate the extent of uptake following dermal administration. This was conducted by administering a low dose of 13C-labeled (substituted) AM to volunteers orally or dermally, and by measuring urinary metabolites or hemoglobin adducts derived from the GA pathway and comparing them to metabolites and hemoglobin adducts derived from AM directly. More specifically, we intended to evaluate urinary metabolites and hemoglobin adducts and to measure hormone levels after exposure to a known dose of AM. As a secondary and no less important objective, we intended to carefully monitor the health of people exposed to AM under these controlled conditions.
MATERIALS AND METHODS
Institutional Review Board approval.
This study was conducted in accordance with the Code of Federal Regulations (CFRs) governing Protection of Human Subjects (21 CFR 50), IRB (21 CFR 56), retention of data (21 CFR 312) as applicable and consistent with the Declaration of Helsinki. The administration of 13C AM to the study subjects was conducted at Covance Clinical Laboratories. Institutional Review Board approval of the protocol and the consent form was obtained at Covance Clinical Research Unit (CRU). Written informed consent was obtained from all study participants prior to study participation. Institutional Review Board approval was also obtained at RTI International, where the analysis of the samples was conducted. The study participants were compensated for time spent in the Covance Clinical Research Unit. The stipends were reviewed by the IRB as part of the study protocol, and those approved for this study were within the general stipend range approved by the IRB for other studies conducted at Covance CRU.
Chemicals.
(1,2,3-13C3) AM (CLM-813, lot number 11085, 99% labeled) was obtained from Cambridge Isotopes Limited. Identity and purity were confirmed by 1H and 13C NMR spectroscopy. (Note that the molecular weight of the (1,2,3-13C3) AM is 74, versus 71.08 for natural abundance AM). GA was synthesized by H2O2 oxidation of acrylonitrile (Payne and Williams, 1961), and stored at –20°C. N-(2-Carbamoylethyl)valine (AAVal), N-(2-carbamoylethyl)valine-13C5 (AAVal-13C5), N-(2-carbamoyl-2-hydroxyethyl) valine (GAVal), and N-(2-carbamoyl-2-hydroxyethyl)valine-13C5 (GAVal-13C5) were synthesized and purified as described previously by Fennell et al. (2003). The AAVal phenylthiohydantoin derivative (AAVal PTH), the corresponding 13C-labeled standard AAVal-13C5, GAVal PTH, and 13C-GAVal PTH standards were prepared as described by Fennell et al. (2003). AAVal-leu-anilide was obtained from Bachem Bioscience Inc. (King of Prussia, PA).
Human Study
AM exposure.
Twenty-four volunteers participated in this study. They were all male Caucasians (with the exception of one Native American) weighing between 71 and 101 kg and between 26 and 68 years of age. All volunteers were aspermic and had not used tobacco products for the past 6 months. They passed a drug screen and had not taken prescription drugs or caffeinated products over the previous 3 days. They had not consumed alcohol-containing beverages or medications within 7 days of study entry, and for the duration of the study. Each experimental group consisted of six individuals of whom one was administered a placebo. There were two phases to this study: an oral phase and a dermal phase.
A comprehensive physical exam was conducted on each individual upon check-in to the clinic, at 24 h after compound administration, and 7 days after check out. This exam included medical history, demographic data, neurological examination, 12-lead ECG, vital signs (including oral temperature, respiratory rate, and automated seated pulse and blood pressure), clinical laboratory evaluation (including clinical chemistry, hematology, and complete urinalysis). Each individual also had screens for HIV, hepatitis, and selected drugs of abuse and provided a semen sample to confirm aspermia. Additional ECG, neurological evaluation, an abbreviated physical examination, and subjective evaluations were conducted at 4 h after each AM administration.
In the oral phase, three groups of six people were administered 0.5, 1.0, or 3.0 mg/kg 13C3 AM. Individuals were presented with test substance at approximately 9:00 in the morning to initiate the study. Urine was collected at 0–2, 2–4, 4–8, 8–16, and 16–24 h. Blood was collected immediately prior to compound administration and 24 h later. Hormone blood samples (testosterone, LH, and prolactin levels) were drawn immediately prior to compound administration, 24 h later, and on the follow up visit on day 8.
In the dermal phase, a 50% solution of 13C3 AM was applied directly on the skin to a clean, dry, marked off, 24-cm2 (3 cm x 8 cm) area on the volar forearm. After applying the appropriate amount of material, the liquid was evaporated to dryness using a commercial hair dryer and covered with a sterile gauze pad. After drying the AM solution, the tape which had been used to demark the area of application was removed and placed in a vial containing 20 ml of water. The water (dermal dam solution) was analyzed for AM by HPLC. The site of application was covered with gauze for 24 h, at which time the gauze was removed and the area was washed with 1000 ml of water. The recovered wash water and gauze were analyzed by HPLC for AM. Dermal applications alternated between left and right arms, starting with the subject's dominant arm. Blood was collected immediately prior to compound administration and 24, 48, 72, and 96 h later (immediately prior to administration of the second and third doses, after gauze removal, and prior to leaving the clinic). Hormone blood samples were drawn immediately prior to compound administration, after 24 h, and on day 5 when the volunteers left the clinic.
Each exposure group contained six volunteers. Of the six volunteers in each group, five
Blood samples were processed for storage and shipment at the Covance Clinical Research Unit as follows: the blood samples were centrifuged, and plasma was removed; an equal volume of isotonic saline was added to the red cell pellets; the remaining washed red blood cells were washed by centrifugation with isotonic saline. The washing procedure was repeated a total of three times. The samples were stored frozen until shipped to RTI.
Urine samples were collected at intervals of 0–2, 2–4, 4–8, 8–16, and 16–24 h following administration of AM. The volume of urine in each sample was recorded, and sample aliquots were transferred to sample vials for storage.
Samples of urine and washed red blood cells were shipped to RTI from Covance Clinical Laboratories on dry ice, and were stored at –20°C until processed for analysis.
AM analysis.
The dose solutions, dermal dam solutions, and wash solutions provided by the Covance Clinical Research Unit were analyzed for the concentration of AM using a reversed-phase HPLC method. A calibration curve was prepared with AM over a concentration range of 5–200 μg/ml. Analysis was conducted with a Waters HPLC system consisting of two 515 pumps, a 717 autosampler, and an Applied Biosystems 759A UV detector. Data were recorded with a Waters Millennium data system. Chromatography was conducted on a Beckman Ultrasphere ODS column (4.5 mm x 25 cm) eluted at a flow rate of 1 ml/min with 100% water. Elution was monitored by measuring UV absorbance at 195 nm. The measurements of AM concentration in dose solutions were used to confirm the amount of AM administered in oral dose solutions, and in the dose solution used for dermal applications. The amounts of AM recovered from dermal application in both the dermal dam, which was used to outline the skin area for application, and the skin washings following removal of the gauze covering the application site were also measured. The total amount of AM recovered following dermal application was used to calculate the maximum amount of AM that was available for absorption (total dose applied – amount recovered in the dermal dam and washing solutions).
The AM recovered in the dermal dam and wash solutions would not be available systemically. The amount of dose that could have been taken up was calculated as:
(1)
Urinary metabolite analysis.
Metabolites of (1,2,3-13C3) AM in urine were analyzed from the group of volunteers exposed to 3 mg/kg AM orally by 13C NMR spectroscopy, essentially as described by Sumner et al.,(1992). No analyses were conducted on the samples from the volunteers administered 0.5 or 1.0 mg/kg orally, or those administered AM dermally, due to the insensitivity of the methodology. Individual samples were prepared for all urine samples from subject 015 (3 mg/kg), and these were analyzed individually. To reduce the amount of NMR instrument time required for analysis, composite urine samples were prepared for analysis of each subject administered 3.0 mg/kg orally. For each individual, sample aliquots of each time point were combined in the appropriate proportion (for each time point, added 10 ml x [volume of each time point/ total urine volume]) to make a total volume of 10 ml. Samples were concentrated by mixing 5.0 ml of pooled urine with 10 ml of methanol, and centrifuging at 5000 x g for 10 min. The supernatant was reduced in volume under a stream of nitrogen in a preweighed tube to approximately 300–600 μl. The weight of the concentrated urine was recorded. Water was added to make a total volume of 600 μl, and a solution of dioxane in D2O was added (200 μl).
NMR analysis of urinary metabolites.
Initial analysis of urine samples was conducted on a Bruker 500 MHz NMR spectrometer operating a 125 MHz for 13C. Quantitative analysis of metabolites was conducted on a Varian 500 MHz NMR spectrometer operating at 125 MHz. Samples were prepared by adding D2O, or D2O containing dioxane at a known concentration (200 μl) to an aliquot of a urine sample, a composite urine sample, or a concentrated composite urine sample (800 μl). Carbon-carbon connectivity was established using two-dimensional incredible natural abundance double quantum transfer spectra (INADEQUATE) using the Varian pulse sequence.
LC-MS/MS analysis of hemoglobin adducts.
Quantitative analysis of all of the human globin samples was conducted using an API-4000 LC/MS/MS system with a heated nebulizer source coupled to an Agilent 1100 HPLC system. Data was processed using Analyst version 1.3 software. The HPLC system was composed of a binary HPLC pump, a refrigerated vial/96 well plate autosampler, a photodiodearray detector, and a column heater.
Hemoglobin adduct analysis.
N-(2-Carbamoylethyl)valine (AAVal) and N-(2-carbamoyl-2-hydroxyethyl)valine (GAVal), formed by reaction of AM and GA, respectively, with the N-terminal valine residue in hemoglobin, were measured by an LC-MS/MS method. Globin was isolated from washed red cells (Mowrer et al., 1986). Samples were derivatized with phenylisothiocyanate in formamide to form adduct phenylthiohydantoin derivatives in a manner analogous to the modified Edman degradation (Bergmark, 1997; Perez et al., 1999; Trnqvist et al., 1986). Internal standards, AAValPTH-13C5 and GAVal PTH-13C5, were added, and the samples were extracted using a Waters Oasis HLB 3 cc (60 mg) extraction cartridge (Milford, MA). The samples were eluted with methanol, dried, and reconstituted in 100 μl of 50:50 MeOH:H2O (containing 0.1% formic acid). Analysis was conducted using an HP 1100 HPLC system interfaced to a PE Sciex API 4000 LC-MS with a Turboionspray interface. Chromatography was conducted on a Phenomenex Luna Phenyl-Hexyl Column (50 mm x 2 mm, 3 mm) eluted with 0.1% acetic acid in water and methanol at a flow rate of 350 μl/min, with a gradient of 45–55% methanol in 2.1 min. The elution of adducts was monitored by Multiple Reaction Monitoring (MRM) in the negative ion mode for the following ions:
Natural Abundance Analytes
AAVal-PTH: m/z 304 233 (M-H– M-H– – CH2-CH2-CONH2)
GAVal-PTH: m/z 320 233 (M-H– M-H– – CH2-CHOH-CONH2)
Adducts Derived from 1,2,3-13C AM
13C3-AAVal-PTH: m/z 307 233 (M-H– M-H– – 13CH2-13CH2-13CONH2)
13C3-GAVal-PTH: m/z 323 233 (M-H– M-H– – 13CH2-13CHOH-13 CONH2)
Internal Standards
AAVal-PTH-13C5: m/z 309 238 (M-H– M-H– – CH2-CH2-CONH2)
GAVal-PTH-13C5: m/z 325 238 (M-H– M-H– – CH2-CHOH-CONH2)
Quantitation of AAVal was conducted using the ratio of analyte to internal standard, with a calibration curve generated using AAVal-leu-anilide. Quantitation of GAVal was conducted using the ratio of analyte to internal standard.
Rat Study.
Prior to conducting the study in humans, four male Fischer 344 rats (body weights from 202 to 212 g) were administered (1,2,3-13C) AM at a dose of 3 mg/kg by gavage in distilled water. This component of the study was conducted in parallel with administration of 50 mg (1,2,3-13C) AM/kg to rats by gavage described in detail previously (Sumner et al., 2003). The rats were placed in metabolism cages for 24 h following dosing for collection of urine. After 24 h, the rats were euthanized, and blood samples were collected by cardiac puncture. Washed red blood cells were prepared by centrifugation, and globin was isolated by the method of Mowrer et al. (1986). Globin samples were analyzed by LC-MS/MS for AAVal, 13C3-AAVal, GAVal, and 13C3-GAVal as described by Sumner et al. (2003). Urine samples were analyzed by 13C NMR spectroscopy for metabolites of AM (Sumner et al., 2003). For quantitation of metabolites, samples were concentrated by addition of methanol to a 3-ml sample of urine, centrifugation, and evaporation under a stream of nitrogen. D2O and dioxane were added.
In vitro reaction rate constants.
Washed red cells from male F-344 rats, or from human blood, were lysed with an equal volume of distilled water and incubated with AM or GA at a concentration of 100 mM 37°C. Samples were removed at 0, 2, 5, 10, 15, 30, and 60 min, and the reactions were terminated by placing the tubes on ice and immediately adding 3 ml of 50 mM HCl in isopropanol. Globin was then isolated as described by Mowrer et al. (1986).
RESULTS
Rat Study
Analysis of urinary metabolites.
The analysis of urinary metabolites was conducted as described previously (Sumner et al., 2003), and the metabolites detected in rats and humans are shown in Figure 1. Prior to conducting the administration of AM in humans, the ability of the methods used to detect urinary metabolites and hemoglobin adducts was evaluated with a single dose of 3 mg/kg 13C3 AM administered by gavage in rats. The metabolites could be detected in urine samples. However, for quantitation of urinary metabolites, concentration of urine samples was required. The results of the quantitative analysis are shown in Table 1. Conjugation with GSH to form metabolite 1 accounted for 59% of the total metabolites. Metabolites 2,2' and 3,3', produced by GSH conjugation of GA, accounted for 25 and 16% of the urinary metabolites, respectively. GA was detected in urine samples from two of the rats prior to concentration, but was below the limit of quantitation in the concentrated samples. Metabolism via epoxidation to GA (2,2'-5, Table 1) accounted for approximately 41% of the urinary metabolites.
Analysis of the hemoglobin adducts from rats administered 3 mg/kg 13C3 AM is shown in Table 2. 13C3-AAVal and 13C3-GAVal were both increased by administration of AM. The ratio of 13C3-GAVal: 13C3-AAVal was 0.84 ± 0.07.
Human Study
Clinical findings.
No adverse events were reported in the oral phase of the study. With the dermal administration, one individual appeared to have a mild contact dermatitis, which is a known response to AM and was part of the informed consent. This individual was seen by a dermatologist who performed a skin biopsy which was consistent with a delayed hypersensitivity reaction. The skin reaction resolved 39 days after the first application of AM and 23 days after the reaction was manifested. Thus, the AM caused a delayed hypersensitivity reaction when placed on the skin, a reaction that took more than 3 weeks to resolve completely. An increase in the liver enzyme alanine aminotransferase (ALT) was observed above the upper limit of the reference range (normal) in four of the five individuals who
Dose administered.
For oral administration, AM was administered in a constant volume of 200 ml water to give the appropriate dose of 0.5, 1.0, or 3.0 mg/kg. The actual amount of AM administered was verified by analysis of aliquots of the dose solution with calculation of the amount of AM in 200 ml for each subject and the amount of AM administered per kg body weight. One individual from each dose group did not receive AM, and this was verified by analysis of the administered dose. The mean doses calculated were 0.44 ± 0.01, 0.92 ± 0.01, and 2.87 ± 0.03 mg/kg, and were 88, 92, and 96% of the nominal dose at 0.5, 1.0, and 3.0 mg/kg, respectively.
With dermal administration, the appropriate volume of a 50% solution of AM was applied to the skin. The dose applied each day based on HPLC analysis was calculated as 2.48 mg/kg, and was 83% of the nominal dose. Recovery of the dose after application was evaluated by HPLC. The tape used to demark the area of application (dermal dam solution) was found to contain between 9 and 54 mg of AM (Table 3). The amount of AM recovered by washing the application site with water after removal of the gauze covering ranged between 62 and 154 mg. The total AM per day recovered in dermal dam and wash solutions ranged between 85 and 190 mg and accounted for 36–86% of the applied AM. While considerable variability was observed in the recovery of AM, the mean recovery in dam and washes ranged between 65 and 71% of the total AM applied. The mean absorbed dose ranged from 0.73 to 0.86 mg/kg/day (Table 3). The cumulative absorbed dose (over the 3 days of administration) was 2.35 ± 0.50 mg/kg (Table 3). This value includes material that could be retained at the site of application and probably represents the maximum that could be absorbed rather than the actual amount absorbed.
Analysis of urinary metabolites.
Urine samples from a single individual administered 3 mg/kg orally were evaluated qualitatively by 13C NMR spectroscopy prior to quantitative analysis. A sample of each time point collection was analyzed from subject 15. The majority of the metabolite signals were found in the 2–4, 4–8, and 8–16 h samples. No signals indicative of the presence of 13C3 AM or its metabolites were found in the predose urine.
To achieve the necessary signal to noise ratio for quantitative analysis, and to collect data that were similar to those obtained from studies in rodents, aliquots of the urine samples from individuals were combined in the appropriate proportions to make a 24-h composite sample, which was then concentrated. Dioxane was added as an internal standard for quantitation of the metabolites.
The 13C NMR spectrum of a composite urine sample obtained from a volunteer who did not receive AM is shown in Supplementary Data. A number of signals from endogenous metabolites, including urea at 162.5 ppm were seen. In the samples of urine from volunteers administered 3.0 mg/kg (1,2,3-13C3) AM, there are additional signals arising from AM and its metabolites (Supplementary Data). These signals showed characteristic multiplets arising from coupling with adjacent labeled carbon atoms. Many but not all of the signals present in urine samples from rats and mice administered (1,2,3-13C3) AM were present in the human urine samples. Some additional signals that had not been previously observed were also present. These signals are presented in Table 4. The nomenclature used previously (Sumner et al., 1992, 1997, 2003) is also used to describe the metabolites in this paper. The major metabolites present in all of the composite urine samples were derived from direct glutathione conjugation of AM (Sumner et al., 1992, 1997, 2003).
The major metabolites, 1 and 1', corresponding to N-acetyl-S-(3-amino-3-oxopropyl)cysteine and S-(3-amino-3-oxopropyl)cysteine, showed signals at 27 ppm (doublet), 34.7 ppm (2 doublets of doublets), and at 177 ppm (2 doublets). Signals associated with GA, metabolite 4, were observed at 46.9 ppm (4a), and at 48.5 ppm (4b). Signals from the hydrolysis product of GA (2,3-dihydroxypropionamide) were observed at 62.99 ppm (5a, doublet), at 72 ppm (5,5'b, doublet of doublets), and at 175.5 ppm (5,5'c, doublet). Low-intensity signals that may be due to metabolite 2 (N-acetyl-S-(3-amino-2-hydroxy-3-oxopropyl)cysteine) were observed at 35.8 ppm and at 70.2 ppm. Signals that are associated with metabolite 3 (N-acetyl-S-(1-carbamoyl-2-hydroxyethyl)cysteine) were not observed. A metabolite that had not been previously observed in rats and mice gave a signal at 46.47 ppm (doublet, J = 37 Hz). An INADEQUATE spectrum was used to establish the carbon–carbon connectivity of the main signals of metabolite 1, GA, and glyceramide. In addition, the connectivity of the doublet at 46.47 ppm (6b) was established with a complex doublet of doublets at 27.60 ppm (6a), and a doublet at 175.4 ppm (6c). These signals have been assigned to the labeled carbon atoms of N-acetyl-S-(3-amino-3-oxopropyl)cysteine-S-oxide, based on the comparison with a synthesized standard (to be reported in an additional publication).
Quantitation of the metabolites present was conducted using the integral of the metabolite signals and dioxane added as internal standard. Spectra for quantitation were acquired with decoupling only during acquisition to ensure that nuclear Overhauser enhancement was minimized so that accurate quantitation could be conducted. This resulted in the decrease in intensity of some of the low-intensity signals to the point were quantitation was not readily possible (e.g., with metabolite 2). Although signals for AM were readily apparent, these were not quantitated because of the long relaxation time for the signals of AM. A summary of the quantitative information for urinary metabolites is presented in Table 5. Approximately 34% of the administered dose of AM was recovered in the total urinary metabolites within 24 h of administration. Metabolite 1 accounted for approximately 72% of the metabolites excreted. The sulfoxide derived from metabolite 1 accounted for approximately 14% of the metabolites measured. Metabolites that are known to be derived from GA (4 and 5) represented approximately 14% of the metabolites.
Analysis of hemoglobin adducts.
Analysis of AAVal and GAVal was set up and validated as described previously (Fennell et al., 2003). A standard curve was developed with AAVal-leu-anilide standard and 13C5-AAVal PTH. GAVal was measured based on the ratio of analyte to added 13C5 GAVal PTH. In each of the samples obtained prior to administration of (1,2,3-13C3) AM, measurable adduct backgrounds for AAVal and GAVal were detected.
Typical chromatograms for AAVal and GAVal from a volunteer administered 13C3 AM orally (0.5 mg/kg) are shown in Supplementary Data (both before and following administration). The administration of (1,2,3-13C3) AM resulted in an increase in the peak height and area for 13C3-AAVal and 13C3-GAVal. The mean values for hemoglobin adduct levels from the various groups administered AM orally are presented in Table 6. AAVal, GAVal, 13C3-AAVal, and 13C3-GAVal were measured for each individual prior to exposure to 13C3 AM and at 24 h following administration. The majority of the individual values (not shown) for AAVal prior to exposure were in the range of 40–200 fmol/mg globin. Most of the values for AAVal measured before and after exposure were similar. One exception to this was noted. One volunteer had high levels of AAVal in the first sample (986 fmol/mg), which dropped in the second sample (43 fmol/mg). The reason for this discrepancy cannot be explained, and the value for the pre-exposure sample has been excluded from calculations of statistical parameters. Prior to administration of AM, the levels of GAVal were in the range of 16–67 fmol/mg globin, and peaks associated with 13C3-AAVal and 13C3-GAVal were not detected. Following administration of AM, 13C3-AAVal and 13C3-GAVal adduct levels increased in five out of the six members of each group. The subjects who
On oral administration of (1,2,3-13C3) AM, levels of 13C3-AAVal increased in a dose-dependent manner (Table 6). A plot of 13C3-AAVal versus the nominal dose administered was linear (Fig. 2). Similarly, on oral administration of (1,2,3-13C3) AM, levels of 13C3-GAVal increased in a dose-dependent manner (Table 6). A plot of 13C3-GAVal versus the nominal dose administered was also linear (Fig. 2). The levels of 13C3-GAVal formed were considerably lower than those of 13C3-AAVal, with a ratio of 13C3-GAVal: 13C3-AAVal of ranging from 0.36 to 0.44 (Table 6).
Following dermal administration of AM (days 1, 2, and 3), both AAVal and GAVal increased in a linear manner after each dose, on days 2, 3, and 4 (Table 6). Little change was noted between days 4 and 5 when no administration took place. The relationship between 13C3-AAVal, 13C3-GAVal, and the cumulative daily exposure was linear (not shown). The ratio of 13C3-GAVal:13C3-AAVal ranged from 0.48 to 0.68.
To enable the comparison of dose groups and routes of exposure, the data for hemoglobin adducts have been normalized by the actual dose estimated from analysis of AM administered and recovered (Table 7). The amount of 13C3-AAVal and 13C3-GAVal formed following dermal exposure was considerably lower than that observed on oral administration, based on the dermal Administered Dose. Calculation of the dermal dose, taking into consideration the AM recovered in the dermal washes (Absorbed Dose) resulted in an increase in the amount of AAVal and GAVal formed per unit dose, but the yield from dermal administration was still considerably lower than that with oral administration (12.7 vs 74.7 nmol AAVal/g globin/mmol AM/kg). Hemoglobin adducts formed in humans on oral administration have been compared with those formed in rats on gavage administration (Table 7). Adduct levels normalized for dose are approximately 3-fold higher in humans for AAVal and 1.7-fold higher for GAVal in humans. The ratio of GAVal:AAVal in humans on oral administration was similar to that previously reported (Fennell et al., 2003) for rats administered AM by gavage at 50 mg/kg (Table 8). On dermal administration in humans, the ratio of GAVal:AAVal was slightly increased compared with oral administration (0.57 vs. 0.39); the magnitude of the increase was much less than that observed in rats (1.7 vs. 0.38).
Hemoglobin adducts can be used to calculate the internal dose or area under the curve in blood, using the reaction rate constant measured in vitro (Ehrenberg and Osterman-Golkar, 1980), using the relationship:
(2)
where k, the second order reaction rate constant, is expressed in units of l/g globin/h, and [RHb] is the adduct concentration, and [Hb] is the concentration of hemoglobin. In Table 8, the rate constants measured in this laboratory are presented. These are similar in magnitude to those reported previously by Bergmark et al. (1993). For the human AAVal rate constant, our value of 4.27 x 10–6 l/g globin/h agrees well with the previously reported value of 4.4 x 10–6 l/g globin/h. However, our values for GA reaction with rat and human hemoglobin are approximately half of those reported previously. It is possible that the differences in the method of analysis could account for this difference. The calibration of AAVal measurements conducted here used commercially available AAVal-leu-anilide together with AAVal-13C5PTH. The agreement between values calculated with a standard curve and those calculated from the ratio of analyte:internal standard were within 1%. However, GAVal-leu-anilide was not commercially available at the time that these samples were analyzed, and standardization was conducted by the ratio of analyte:internal standard alone. However, using the rate constant to calculate the AUC for AM and GA from the adduct data, with the rate constants measured under the same conditions should enable comparison of data with other laboratories in the future. Calculated values for AAVal and GAVal were converted to "Dose" or AUC normalized for administered or absorbed dose (Table 9). Compared with the area under the curve (AUC) calculated for AM in rats administered 3 mg/kg, the AUC in humans ranged from 2.75- to 3.7-fold higher. In contrast, the AUC for GA in humans was similar to that in the rat, ranging from approximately 1.2 to 1.4 times that of rat.
DISCUSSION
This study was undertaken to understand the metabolism of AM in humans, and the relationship between hemoglobin adducts of AM and GA and exposure to AM via the oral and dermal routes. During the design of the study, the finding of AM in the foods had not been disclosed, and the main impetus behind the study was to understand the fate of AM in scenarios that could occur with occupational exposure. The controlled exposure of human subjects was conducted in a dose-escalating manner with careful monitoring of the subjects.
The administration of a low dose (3 mg/kg) of AM by gavage to rats resulted in a greater amount of metabolism via GA (41% of the urinary metabolites) compared with a higher dose of 59 mg/kg (28% of the urinary metabolites, (Sumner et al., 2003)). The fate of GA was primarily conjugation with GSH, resulting in the excretion of two mercapturic acids (metabolites 2 and 3). The total amount of AM metabolites recovered by 24 h after dosing (50%) is similar to that reported by Miller et al. (1982), and by Kadry et al. (1999).
Hemoglobin adducts represent useful dosimeters for reactive chemicals and metabolites, with the extent of adduct formation related to the concentration of reactant integrated over time. The ratio of GAVal:AAVal has been observed to differ with route of exposure (Sumner et al., 2003). This should reflect differences in the relative AUCs for AM and GA in blood (Calleman, 1996; Calleman et al., 1992). At a dose of 3 mg/kg in rats, GAVal:AAVal was higher (0.84) than that observed at 50 mg/kg (0.38, (Fennell et al., 2003)). Both this observation and the larger percentage of metabolites in urine derived from GA at the lower dose are consistent with saturation of the epoxidation of AM at the higher dose administered (Calleman et al., 1992).
The urinary metabolites of AM in humans showed similarities and differences with data obtained previously in the rat and mouse. The main pathway of metabolism in humans was via direct glutathione conjugation, forming N-acetyl-S-(3-amino-3-oxopropyl)cysteine, as observed in the rat and mouse, and its S-oxide, which has not been reported previously. Epoxidation to GA was the other important pathway, with glyceramide formed as a major metabolite in humans. GA was detected in low amounts. The glutathione conjugation of GA, which is a major pathway in rodents, appeared to occur at very low levels in humans, with metabolite 2 detected, but not quantitated, and metabolite 3 not detected. Metabolism via GA (derived from GA and glyceramide) in humans was approximately 12% of the total urinary metabolites. This is considerably lower than the amount of GA derived metabolites reported for oral administration of AM in rats (28% at 50 mg/kg, (Sumner et al., 2003) and in mice (59% at 50 mg/kg, (Sumner et al., 1992)).
This study has provided data on the amount of hemoglobin adducts derived from AM and GA following administration of a defined dose of AM to people. Both AAVal and GAVal increased linearly with increasing dose of AM administered orally, suggesting that, over the range of 0.5–3.0 mg/kg, there is no saturation of metabolism of AM to GA. The ratio of GAVal:AAVal produced by administration of AM was similar to the ratio of the background adducts prior to exposure. Compared with the equivalent oral administration in rats (3 mg/kg), the ratio of 13C3-GAVal: 13C3-AAVal in humans was lower (0.44 ± 0.06) than in rats (0.84 ± 0.07), and the absolute amount of 13C3-AAVal formed in humans was approximately 2.7-fold higher than in the rat. The absolute amount of 13C3-GAVal was approximately 1.4-fold higher than that formed in the rat.
In risk assessment, the extrapolation of dose between species is generally conducted using a scaling factor of body weight3/4 (U.S. General Accounting Office, 2001). Thus, for extrapolation from rats to humans, a scaling factor of approximately 4 would be used. For effects that are mediated via the action of AM, based on the AAVal comparison between rats and humans, a factor of 2.7 would appear appropriate, whereas for effects mediated by GA, a factor of 1.4 would appear appropriate.
The average AAVal and GAVal formed on dermal administration were calculated from the values in Table 7. Dermal administration of AM resulted in much lower levels of AAVal and GAVal formed compared with an equivalent dose by the oral route. Comparing AAVal after dermal (Table 7, 4.9 nmol/g globin/mmol AM/kg) and oral administration (Table 7, 74.7 nmol/g globin/mmol AM/kg) indicated approximately 6.6% of the dermally administered dose (Table 3) was taken up, assuming 100% oral absorption. Similarly, dermal administration also resulted in much lower formation of GAVal (9.7% of that formed on oral administration). Approximately 66% of the administered dose of AM was recovered in dam solutions and the wash solutions (data not shown), and thus was not systemically absorbed on dermal administration. This suggests that only approximately 33% of the dermally applied dose could have been absorbed (Absorbed Dose, Table 3). The AAVal formed on dermal administration normalized with the absorbed dose (Table 3) was only 17.0% of that expected (Table 7, 12.7 nmol/g globin/mmol AM/kg for dermal vs. 74.7 nmol/g globin/mmol AM/kg for oral). This may indicate that 83% of the AM that penetrated the skin was not available systemically. GAVal on dermal administration normalized for the absorbed dose in a similar manner was 25.3% of that formed on oral administration (Table 7, 7.3 pmol/g globin/mmol AM/kg for dermal versus 28.9 pmol/g globin/mmol AM/kg for oral). An alternative explanation is that the AM and GA may be more rapidly metabolized on dermal exposure, resulting in a lower AUC and lower adduct formation per mg/kg. This may be clarified by analysis of the urinary metabolites in the dermally exposed individuals.
An important result from this study is the capability to define the amount of AAVal adduct expected from exposure to AM via different routes. The expected amount of adduct that would accumulate from continuous exposure can be calculated from a knowledge of the amount of adduct formed per day of exposure, and from the kinetics of the erythrocyte (Fennell et al., 1992; Osterman-Golkar et al., 1976). Exposure via oral intake to 1 μg AM/kg (1.05 fmol AAVal/mg globin/day) for the lifespan of the erythrocyte (120 days) would result in the accumulation of adducts to 63 fmol/mg globin. Daily dermal exposure to 1 μg AM/kg (0.18 fmol AAVal/mg globin/day) for the lifespan of the erythrocyte (120 days) would result in the accumulation of adducts to 10.8 fmol AAVal/mg globin. With workplace exposure of 5 days per week, this would decrease to approximately 7.8 fmol AAVal/mg globin.
Exposure to AM can occur through occupational exposure, via ingestion in food, and via cigarette smoking (Bergmark, 1997; Bergmark et al., 1993; Tareke et al., 2002). Occupational exposure can occur via inhalation of AM vapor and dust and via the dermal route. The adduct yield for a given exposure differs between oral and dermal administration in rodents (Sumner et al., 2003) and in humans (this study). The contribution of AM exposure via diet and cigarette smoking will complicate the assessment of low level workplace exposure via adduct measurement. However, different studies have described AAVal measured in occupational settings ranging from 300 to 34,000 fmol/mg (Calleman et al., 1994), up to 17,700 fmol/mg (Hagmar et al., 2001), and 71–1854 fmol/mg in a plant that was in production for a period of 2 weeks (Perez et al., 1999). Lifestyle and dietary exposures appear to cause much lower levels of adducts.
Using the data obtained for hemoglobin adducts from the oral administration of AM, it is possible to estimate the amount of AM taken in from the diet over the lifespan of the erythrocyte, and compare this with the estimates of AM in the diet. A detailed consideration of the dietary exposure to AM and comparison with the dietary exposure estimates reported previously (Svensson et al., 2003; Tareke et al., 2002; Trnqvist et al., 1998; WHO, 2002) is included in the Appendix.
In summary, this study has examined the metabolism of AM in people and compared the internal dose of AM and GA in people with that observed in rats. The data reported are consistent with slower elimination of AM in humans and slower metabolism of AM to GA in humans. The hemoglobin adduct measurements obtained will provide a calibration for estimates of exposure from diet, lifestyle, and the workplace.
Appendix
Estimates of Dietary Intake from AAVal Measurements
The normalized formation of AAVal per unit dose of unlabeled AM, averaged for all of the oral dose groups, was calculated as 1050 fmol/mg globin/mg AM/kg. Using this value, the amount of AM exposure that would be expected in a human from the diet can be calculated. In this study, the average pre-exposure AAVal level was 79 ± 49 fmol/mg globin (excluding subject 13). Hagmar et al. (2001) reported a range of 0.02–0.07 nmol AAVal/g globin (equivalent to 20–70 fmol AAVal/mg globin) in unexposed individuals. The steady state adduct level from continuous exposure is calculated to be a.ter/2, where a is the daily adduct increment, and ter is the erythrocyte lifespan, which in humans is approximately 120 days. The amount of adduct formed per day, assuming similar exposure per day over the erythrocyte lifespan, would be 1/60th of the daily adduct increment. For the average level of 80 fmol/mg:
(3)
The amount of AM taken in can be estimated by:
(4)
Similarly for the lower level of 20 fmol/mg, the daily adduct increment is 0.33 fmol/mg globin/day, and AM intake is 0.31 μg/kg/day.
These estimates suggest a daily intake in the range of 22–88 μg for an average 70-kg person and can be compared with a variety of estimates that have been produced for the average daily intake of AM. The World Health Organization consultation (2002) estimated a daily intake of 0.8 μg AM/kg/day for the average consumer, based on AM levels in foods. Similarly, (Svensson et al., 2003) estimated a mean daily intake of 31 μg/day based on food consumption data, which corresponds to 0.44 μg/kg/day for a 70-kg person. Trnqvist et al. (1998) have estimated a daily intake of approximately 100 μg per person based on hemoglobin adduct measurements of 30 fmol/mg globin, reported in Tareke et al. (2002). These calculations are based on a complex equation that relates the estimated dose to adduct level accumulated over the erythrocyte lifespan, rate of elimination (k), the reaction rate constant for adduct formation, the volume of distribution, and the erythrocyte lifespan (ter).
(5)
Of the parameters in this equation, two are estimated: the elimination rate constant in humans (Calleman, 1996), and the volume of distribution (assumed to be 1). Using the data obtained in this study for a single exposure in a modified equation enables the estimation of possible values for the elimination rate constant and volume of distribution.
(6)
Rearranging and solving for VD yields a value of 0.38 l/kg, or distribution into 27 liters in a 70-kg person. This is approximately 64% of the estimated 42 liters of total body water. Using the value for VD of 0.38 l/kg and an adduct level of 30 pmol adduct/g globin, an uptake of AM of 0.48 μg/kg/day is calculated, or 33 μg /day for a 70-kg person. This estimated uptake is more in line with the estimates of AM exposure based on food consumption.
SUPPLEMENTARY DATA
Supplementary data are available at Toxicological Sciences online.
NOTES
The authors certify that all research involving human subjects was done under full compliance with all government policies and the Helsinki Declaration.
ACKNOWLEDGMENTS
The authors would like to acknowledge CIIT Centers for Health Research, Research Triangle Park, NC, where the administration of AM to rats and analyses of rat urine and globin were conducted.
This study was funded by SNF SA, which is a manufacturer of acrylamide and polyacrylamide.
REFERENCES
Bergmark, E. (1997). Hemoglobin adducts of acrylamide and acrylonitrile in laboratory workers, smokers and nonsmokers. Chem. Res. Toxicol. 10, 78–84.
Bergmark, E., Calleman, C. J., He, F., and Costa, L. G. (1993). Determination of hemoglobin adducts in humans occupationally exposed to acrylamide. Toxicol. Appl. Pharmacol. 120, 45–54.
Besaratinia, A., and Pfeifer, G. P. (2004). Genotoxicity of acrylamide and glycidamide. J. Natl. Cancer Inst. 96, 1023–1029.
Calleman, C. J. (1996). The metabolism and pharmacokinetics of acrylamide: Implications for mechanisms of toxicity and human risk estimation. Drug Metab. Rev. 28, 527–590.
Calleman, C. J., Bergmark, E., and Costa, L. G. (1990). Acrylamide is metabolized to glycidamide in the rat: Evidence from hemoglobin adduct formation. Chem. Res. Toxicol. 3, 406–412.
Calleman, C. J., Stern, L. G., Bergmark, E., and Costa, L. G. (1992). Linear versus nonlinear models for hemoglobin adduct formation by acrylamide and its metabolite glycidamide: Implications for risk estimation. Cancer Epidemiol. Biomarkers Prev. 1, 361–368.
Calleman, C. J., Wu, Y., He, F., Tian, G., Bergmark, E., Zhang, S., Deng, H., Wang, Y., Crofton, K. M., Fennell, T., et al.(1994). Relationships between biomarkers of exposure and neurological effects in a group of workers exposed to acrylamide. Toxicol. Appl. Pharmacol. 126, 361–371.
Dixit, R., Seth, P. K., and Mukjtar, H. (1982). Metabolism of acrylamide into urinary mercapturic acid and cysteine conjugates in rats. Drug Metab. Dispos. 10, 196–197.
Dybing, E., and Sanner, T. (2003). Risk assessment of acrylamide in foods. Toxicol. Sci. 75, 7–15.
Edwards, P. M. (1975). The distribution and metabolism of acrylamide and its neurotoxic analogues in rats. Biochem. Pharmacol. 24, 1277–1282.
Ehrenberg, L., and Osterman-Golkar, S. (1980). Alkylation of macromolecules for detecting mutagenic agents. Teratog. Carcinog. Mutagen 1, 105–127.
European Union (2002). European Union Risk Assessment Report: Acrylamide. Luxembourg. 210 p.
Fennell, T. R., Snyder, R. W., Krol, W. L., and Sumner, S. C. (2003). Comparison of the hemoglobin adducts formed by administration of N-methylolacrylamide and acrylamide to rats. Toxicol Sci 71, 164–175.
Fennell, T. R., Sumner, S. C., and Walker, V. E. (1992). A model for the formation and removal of hemoglobin adducts. Cancer Epidemiol. Biomarkers Prev. 1, 213–219.
Friedman, M. A., Dulak, L. H., and Stedham, M. A. (1995). A lifetime oncogenicity study in rats with acrylamide. Fundam. Appl. Toxicol. 27, 95–105.
Gamboa da Costa, G., Churchwell, M. I., Hamilton, L. P., Von Tungeln, L. S., Beland, F. A., Marques, M. M., and Doerge, D. R.(2003). DNA adduct formation from acrylamide via conversion to glycidamide in adult and neonatal mice. Chem. Res. Toxicol. 16, 1328–1337.
Hagmar, L., Trnqvist, M., Nordander, C., Rosen, I., Bruze, M., Kautiainen, A., Magnusson, A. L., Malmberg, B., Aprea, P., Granath, F., et al.(2001). Health effects of occupational exposure to acrylamide using hemoglobin adducts as biomarkers of internal dose. Scand. J. Work Environ. Health 27, 219–226.
Hashimoto, K., and Aldridge, W. N. (1970). Biochemical studies on acrylamide, a neurotoxic agent. Biochem. Pharmacol. 19, 2591–2604.
Hashimoto, K., and Tanii, H. (1985). Mutagenicity of acrylamide and its analogues in Salmonella typhimurium. Mutat. Res. 158, 129–133.
Johnson, K. A., Gorzinski, S. J., Bodner, K. M., Campbell, R. A., Wolf, C. H., Friedman, M. A., and Mast, R. W. (1986). Chronic toxicity and oncogenicity study on acrylamide incorporated in the drinking water of Fischer 344 rats. Toxicol. Appl. Pharmacol. 85, 154–168.
Kadry, A. M., Friedman, M. A., and Abdel Rahman, M. S. (1999). Pharmacokinetics of acrylamide after oral administration in male rats. Environ. Toxicol. Pharmacol. 7, 127–133.
Lafferty, J. S., Kamendulis, L. M., Kaster, J., Jiang, J., and Klaunig, J. E. (2004). Subchronic acrylamide treatment induces a tissue-specific increase in DNA synthesis in the rat. Toxicol. Lett. 154, 95–103.
Miller, M. J., Carter, D. E., and Sipes, I. G. (1982). Pharmacokinetics of acrylamide in Fisher-344 rats. Toxicol. Appl. Pharmacol. 63, 36–44.
Mowrer, J., Trnqvist, M., Jensen, S., and Ehrenberg, L. (1986). Modified Edman Degradation applied to hemoglobin for monitoring occupational exposure to alkylating agents. Toxicol. Env. Chem. 11, 215–231.
Osterman-Golkar, S., Ehrenberg, L., Segerback, D., and Hallstrom, I. (1976). Evaluation of genetic risks of alkylating agents. II. Haemoglobin as a dose monitor. Mutat. Res. 34, 1–10.
Payne, G. B., and Williams, P. H. (1961). Reactions of Hydrogen Peroxide. VI. Alkaline epoxidation of acrylonitrile. J. Org. Chem. 26, 651–659.
Perez, H. L., Cheong, H. K., Yang, J. S., and Osterman-Golkar, S. (1999). Simultaneous analysis of hemoglobin adducts of acrylamide and glycidamide by gas chromatography-mass spectrometry. Anal. Biochem. 274, 59–68.
Rosen, J., and Hellenas, K. E. (2002). Analysis of acrylamide in cooked foods by liquid chromatography tandem mass spectrometry. Analyst 127, 880–882.
Segerback, D., Calleman, C. J., Schroeder, J. L., Costa, L. G., and Faustman, E. M. (1995). Formation of N-7-(2-carbamoyl-2-hydroxyethyl)guanine in DNA of the mouse and the rat following intraperitoneal administration of [14C]acrylamide. Carcinogenesis 16, 1161–1165.
Spencer, P. S., and Schaumburg, H. H. (1974a). A review of acrylamide neurotoxicity. Part I. Properties, uses and human exposure. Can. J. Neurol. Sci. 1, 143–150.
Spencer, P. S., and Schaumburg, H. H. (1974b). A review of acrylamide neurotoxicity. Part II. Experimental animal neurotoxicity and pathologic mechanisms. Can. J. Neurol. Sci. 1, 152–169.
Spencer, P. S., and Schaumburg, H. H. (1975). Nervous system degeneration produced by acrylamide monomer. Environ. Health Perspect. 11, 129–133.
Sumner, S. C., Fennell, T. R., Moore, T. A., Chanas, B., Gonzalez, F., and Ghanayem, B. I. (1999). Role of cytochrome P450 2E1 in the metabolism of acrylamide and acrylonitrile in mice. Chem. Res. Toxicol. 12, 1110–1116.
Sumner, S. C., MacNeela, J. P., and Fennell, T. R. (1992). Characterization and quantitation of urinary metabolites of [1,2,3-13C]acrylamide in rats and mice using 13C nuclear magnetic resonance spectroscopy. Chem. Res. Toxicol. 5, 81–89.
Sumner, S. C., Selvaraj, L., Nauhaus, S. K., and Fennell, T. R. (1997). Urinary metabolites from F344 rats and B6C3F1 mice coadministered acrylamide and acrylonitrile for 1 or 5 days. Chem. Res. Toxicol. 10, 1152–1160.
Sumner, S. C., Williams, C. C., Snyder, R. W., Krol, W. L., Asgharian, B., and Fennell, T. R. (2003). Acrylamide: A comparison of metabolism and hemoglobin adducts in rodents following dermal, intraperitoneal, oral, or inhalation exposure. Toxicol. Sci. 75, 260–270.
Svensson, K., Abramsson, L., Becker, W., Glynn, A., Hellenas, K. E., Lind, Y., and Rosen, J. (2003). Dietary intake of acrylamide in Sweden. Food Chem. Toxicol. 41, 1581–1586.
Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S., and Trnqvist, M. (2002). Analysis of acrylamide, a carcinogen formed in heated foodstuffs. J. Agric. Food Chem. 50, 4998–5006.
Trnqvist, M., Bergmark, E., Ehrenberg, L., and Granath, F. (1998). Risk Assessment of Acrylamide. Swedish Chemicals Inspectorate, Solna, Sweden.
Trnqvist, M., Mowrer, J., Jensen, S., and Ehrenberg, L. (1986). Monitoring of environmental cancer initiators through hemoglobin adducts by a modified Edman degradation method. Anal. Biochem. 154, 255–266.
Tyl, R. W., and Friedman, M. A. (2003). Effects of acrylamide on rodent reproductive performance. Reprod. Toxicol. 17, 1–13.
U.S. General Accounting Office (2001). Chemical Risk Assessment, Selected Federal Agencies' Procedures, Assumptions, and Policies. United States General Accounting Office, Washington, DC.
WHO (2002). Joint FAO/WHO Consultation on the Health Implication of Acrylamide in Food. World Health Organization, Geneva, Switzerland. 33 p.(Timothy R. Fennell, Susan)