Nitro Musk Adducts of Rainbow Trout Hemoglobin: Dose–Response and Toxicokinetics Determination by GC-NICI-MS for a Sentinel Species

Rainbow trout and other fish species can serve as “sentinel” species for the assessment of ecological status and the presence of certain environmental contaminants. As such, they act as bioindicators of exposure. This paper presents seminal data regarding dose–response and toxicokinetics of trout hemoglobin adduct formation from exposure to nitro musks that are frequently used as fragrance ingredients in formulations of personal-care products.

Hemoglobin adducts serve as biomarkers of exposure of the sentinel species, as has been shown in previous studies of hemoglobin adducts formed in trout and environmental carp exposed to musk xylene (MX) and musk ketone (MK). Gas chromatography-electron capture negative ion chemical ionization-mass spectrometry (GC-NICI-MS) employing selected ion monitoring (SIM) was used to measure 4-amino-MX (4-AMX), 2-amino-MX (2-AMX), and 2-amino-MK (2-AMK) released by alkaline hydrolysis from the sulfinamide adducts of hemoglobin.

Dose–response and toxicokinetics were investigated using this sensitive method for the analysis of these metabolites. In the dose–response investigation, the concentrations of 4-AMX and 2-AMX were observed to pass through a maximum at 0.10 mg/g. In the case of 2-AMK, the adduct concentration was almost the same at dosages in the range of 0.030–0.10 mg/g. For toxicokinetics, the concentration of the metabolites in the hemoglobin (Hb) reaches a maximum in the three-day sample after administration of MX or MK. Further elimination of the metabolites exhibited kinetics with a presumed exponential decay and a half-life estimated to be 1–2 days. This suggests that a robust mechanism of elimination of the adducts exists in fish erythrocytes apparently analogous to that observed in mammals. Two sick fish were observed to yield 5–24 times the amount of adducts of similarly exposed fish, suggesting that this elimination mechanism may have been impaired or lacking in susceptible individual fish. It appears that adducts are destroyed in times far shorter than the expected life spans of the erythrocytes. This finding may have implications for the use of Hb biomarkers as integrative measures of exposure in some contexts. Additional conclusions from these preliminary data include the additive burden of exposure to multiple compounds and the increased susceptibility and direct observation of metabolic differences of individual members of the species completely independent of habitat and feeding habit variations.

Musk xylene (1-tert-butyl-3,5-dimethyl-2,4,6-trinitrobezene, MX) and musk ketone (1-tert-butyl-3,5-dimethyl-2,6-dinitro-4-acetylbenzene, MK), the most prevalent synthetic nitro musk compounds, are frequently used as fragrance and additive materials in personal-care products and perfumed household products. They are used as a substitute for expensive natural musk, and their estimated annual production is about 1000 metric tons.1 Due to their persistence in the environment and high potential for bioaccumulation,2 MX and MK have been detected as contaminants in aquatic and terrestrial organisms;3,4 human tissues;5–7 North Sea, river, and freshwater;1,8,9 sewage treatment effluent;10 Norwegian air samples;11 human adipose tissue and breast milk;2,12 developing and adult rats;13 and fish, mussels, and shrimp.14 Metabolites of MX and MK have been identified and quantified in samples of river waters, domestic and industrial sewage sludge,9,15 and homogenized whole fish tissues.16 Some studies have been reported on the ecotoxicity of MX and its metabolites17,18 and toxic effects in mice.19 Several studies suggest that MX is not genotoxic.20–22 MX and MK were identified as inducers of toxifying enzymes, cytochrome P450 1A1, and 1A2 in the rat liver.23

Fish serve as sentinel species for the assessment of ecological health and the presence of certain environmental contaminants.24 As such, they function as bioindicators of exposure and thus provide information on the status of an entire ecosystem or watershed.25 The sentinel species, in turn, may have quantifiable biological responses to the exposure events. Exposure events consist of contact with agents or stressors in the environment that may consist of traditional contaminants such as pesticides, or may be nontraditional substances such as certain endocrine disrupting compounds (EDCs) or pharmaceuticals and personal-care products (PPCPs).26 Because of their aqueous habitat, fish experience continuous exposure to contaminants that perfuse into receiving waters. In addition, there is the factor of cumulative risk as a result of exposure to multiple stressors.27 For example, the protein vitellogenin appears to be a sensitive indicator in some fish species for the presence of estrogenic EDCs.28,29

Hb adducts have served as suitable biomarkers for exposure to carcinogenic aromatic amines and nitroarenes. The metabolites of nitro musks or other related nitroarenes, bound to Hb as biomarkers of exposure, can potentially be used to integrate continuous exposures over a longer time range (possibly over the lifetime of red blood cells), and thus may be better suited for risk assessment than quantitation of urinary metabolites.30,31

Figure 1 - Metabolic pathway of cysteine Hb adduct formation with nitro musk compound using 4-AMX as the example.

Nitroarenes are subject to enzymatic reduction, and their reactive intermediate, nitrosoarenes, react with the sulfhydryl (SH) functional group of cysteine in Hb to form an acid/base labile sulfinamide that hydrolyzes to aromatic amines in the presence of aqueous base.32 This process is shown schematically in Figure 1. The formation of Hb adducts suggests that DNA adducts may be formed as well.

The authors detected 4-AMX metabolite from carp Hb for the purpose of ecological assessment of MX exposures.33 In the course of their earlier studies of trout exposed to MX and MK, a trout Hb adduct of a 4-AMX was found, suggesting that nitro-reduction of MX may occur in fish34,35 as well as humans.36

In the present work, the 4-AMX, 2-AMX, and 2-AMK metabolites bound to Hb, formed by enzymatic reduction of MX and MK, were detected and quantified by GC-NICI-MS using SIM. This was found to be an improvement in sensitivity over earlier methods of analysis.33,37 To the authors’ knowledge, the present investigation is the first report on dose–response and toxicokinetics of nitro musk Hb adducts from the sentinel species of rainbow trout exposed to MX and MK and therefore is the first report on Hb biomarkers of a sentinel species thus exposed.


Standards, chemicals, and solvents

Sodium dodecyl sulfate (SDS), sodium hydroxide pellets, and n-hexane (HPLC-grade) were purchased from Sigma-Aldrich (St. Louis, MO), Fisher Scientific (Pittsburgh, PA), and J.T. Baker (Phillipsburg, NJ), respectively. The internal standard (IS), 2,3,4,5,6-pentafluorobenzophenone (purity 99%), was obtained from Aldrich Chemical Co. (Milwaukee, WI). MX and MK standards were a gift from the Institute of Food Chemistry, University of Hohenheim (Stuttgart, Germany). The metabolites 4-AMX, 2-AMX, and 2-AMK were synthesized from MX and MK by Dr. Lantis I. Osemwengie (U.S. EPA, Las Vegas, NV10). Purities were estimated to be greater than 98%. Tricane methane sulfonate (MS 222) was obtained from Sigma (St. Louis, MO). Deionized water was used for all preparations.

Exposure of trout to nitro musk compounds

Trout exposure experiments were conducted at the Department of Environmental & Molecular Toxicology, Oregon State University (OSU) (Corvallis, OR) for the sampling periods of 24 hr (1 day), 72 hr (3 days), and 168 hr (7 days). A series of standard test solutions containing 10, 30, 100, and 300 mg/mL MX or MK were prepared in salmon oil as the vehicle (pharmaceutical grade, Yukon Nutritional Co., Lake Wales, FL) for trout exposure to MX and MK. At the highest intended concentration, neither the MX nor the MK dissolved completely in the oil, but instead formed an emulsion. Well-shaken standard solutions were injected intraperitoneally into fish that were anesthetized in an aqueous MS 222 solution containing 75 mg/L in a 15-L tank. The anesthetized trout were weighed before injecting the standard solutions into the fish.

For the dose–response study, 24 trout were exposed to MX or MK solutions, three trout for each level. For the toxicokinetic investigations, 12 more trout were exposed to 30 mg/mL MX or MK, six with each standard solution. For control work, nine fish were exposed to the vehicle (no MX or MK) for the same sampling period. After exposure, the fish were returned to labeled tanks with circulating water at 13 °C.

Observation of trout during sampling period

In total, 45 trout were exposed to MX or MK solutions and the salmon oil vehicle, followed by sampling periods of 1 day, 3 days, and 7 days. Following exposure, no food was given to the fish, which were closely monitored in the labeled circulating water tank. Two trout exposed to the 30-mg/mL MX solution were observed to be sick in the tank on day 1 and day 7. One fish exposed to 10 mg/mL MK died on day 1 and was not included in the study.

Collection of trout blood and separation of Hb

Before drawing the blood samples, the trout were anesthetized with MS 222 (250 mg/L). This concentration of the MS 222 solution was also fatal to the trout. Blood samples were drawn from the trout into heparinized individual syringes from the caudal vein and placed into heparinized individual sterile interior Vacutainers (Becton Dickinson and Co., Franklin Lakes, NJ). After a 1-day exposure, 23 blood samples were drawn from 23 trout that were exposed to 10, 30, 100, and 300 mg/mL MX or MK. Three days and seven days after exposure, 12 blood samples, six on each exposure day, were collected from 12 trout that were exposed to 30-mg/mL MX or MK solutions. Nine control blood samples were also drawn from nine trout for 1 day, 3 days, and 7 days after exposure. All blood samples were placed on ice immediately after collection, and the fish were sacrificed. Erythrocytes or red blood cells were separated from plasma by centrifuging at 3500 × g for 10 min at 4 °C and were washed twice with equal volumes of 0.9% saline and recentrifuged. The red blood cells were lysed by adding two volumes of distilled water. The Hb solutions were solidified in a freezer at –24 °C. The solid Hb solutions were shipped overnight from OSU to the National Exposure Research Laboratory, U.S. EPA (Las Vegas, NV), in an insulated box packed with dry ice. Upon receipt, the water was eliminated from the solid Hb solutions by a freeze-drying procedure using a Sentry Microprocessor Controlled Freezemobile and benchtop freeze-dryer (The VirTis Co., Inc., Gardiner, NY). The dried Hb was then placed in a freezer at –24 °C for subsequent analysis of the nitro musk metabolites.

Liberation of bound amine metabolites from Hb

Ref. 33 reports on the alkaline hydrolysis, extraction, and preconcentration procedures for liberation of the bound amino metabolites from the carp Hb. The same procedures were used in this study. The dried extract (about 45 mL) was concentrated under a stream of nitrogen at 45 °C to a volume of about 50–65 μL, to which 10–100 pg/μL of IS was added. The solution was sealed in GC vials and analyzed by GC-NICI-MS using SIM.

Two nonhydrolyzed Hb control experiments were also performed to investigate the possible presence of unbound amino metabolites in the Hb samples. In the experiments, all chemicals and solvents, except for the NaOH, were added to the Hb (about 50 mg), and the same extraction and preconcentration procedures were followed as described in the alkaline hydrolysis work. A laboratory control experiment was carried out by utilizing the same amounts of solvents, chemicals, and reagents used for the hydrolysis, except no Hb was used.