Environmental Toxicology and Chemistry, Vol. 34, No. 4, pp. 880–886, 2015 # 2014 SETAC Printed in the USA
BIOACCUMULATION KINETICS OF THE CONVENTIONAL ENERGETICS TNT AND RDX RELATIVE TO INSENSITIVE MUNITIONS CONSTITUENTS DNAN AND NTO IN RANA PIPIENS TADPOLES GUILHERME R. LOTUFO,*y JAMES M. BIEDENBACH,y JERRE G. SIMS,y PORNSAWAN CHAPPELL,z JACOB K. STANLEY,y and KURT A. GUSTy yUS Army Engineer Research and Development Center, Vicksburg, Mississippi, USA zBadger Technical Services, San Antonio, Texas, USA
(Submitted 19 September 2014; Returned for Revision 10 December 2014; Accepted 15 December 2014) Abstract: The manufacturing of explosives and their loading, assembling, and packing into munitions for use in testing on training sites or battlefields has resulted in contamination of terrestrial and aquatic sites that may pose risk to populations of sensitive species. The bioaccumulative potential of the conventional explosives 2,4,6-trinitrotoluene (TNT) and hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) and of the insensitive munitions (i.e., less shock sensitive) compound 2,4-dinitroanisole (DNAN) were assessed using the Northern leopard frog, Rana pipiens. Trinitrotoluene entering the organism was readily biotransformed to aminodinitrotoluenes, whereas no transformation products were measured for RDX or DNAN. Uptake clearance rates were relatively slow and similar among compounds (1.32–2.19 L kg1 h1). Upon transfer to uncontaminated water, elimination rate was very fast, resulting in the prediction of fast time to approach steady state (5 h or less) and short elimination half-lives (1.2 h or less). A preliminary bioconcentration factor of 0.25 L kg1 was determined for the insensitive munitions compound 3-nitro-1,2,4-trizole-5-one (NTO) indicating negligible bioaccumulative potential. Because of the rapid elimination rate for explosives, tadpoles inhabiting contaminated areas are expected to experience harmful effects only if under constant exposure conditions given that body burdens can rapidly depurate preventing tissue concentrations from persisting at levels that may cause detrimental biological effects. Environ Toxicol Chem 2015;34:880–886. # 2014 SETAC Keywords: Amphibians; Bioconcentration; Energetics; Toxicokinetics
and bioaccumulation potential associated with the presence of DNAN and NTO in the environment. A primary objective of the present research effort was to comparatively assess the bioaccumulation potential of the insensitive munitions compounds (DNAN and NTO) relative to the conventional munitions compounds (TNT and RDX) to determine potential differences in bioconcentration using a novel amphibian test species. Bioconcentration is the process by which a chemical is absorbed by an organism from the water through its respiratory and dermal surfaces. It is the net result of competing rates of uptake and elimination, including biotransformation [12]. In addition, the octanol–water partition coefficient (KOW) is widely used as an indicator of hydrophobicity and thus the partitioning of a chemical from water into lipids and other organic phases [13]. Relatively low log KOW values have been reported for TNT (1.86 [14]), RDX (0.90 [15]), and DNAN (1.61 [16]). According to predictive models (e.g., Arnot and Goas [17] and Meylan et al. [18]), those explosives are weakly hydrophobic and therefore have low potential to bioconcentrate. Reports on the bioconcentration potential of explosives in fish and aquatic invertebrates, which investigated only a relatively small number of species, confirmed the expected low potential to bioconcentrate [8,9]. Moreover, studies examining uptake from water, as well as elimination kinetics and biotransformation potential, were conducted for only a few compounds and even fewer aquatic species [8,9]. Report of studies investigating the bioaccumulation, bioconcentration, or toxicokinetics of explosives in larval amphibians were not found in the available literature. Aquatic toxicology studies using larval amphibians are limited to evaluations of the lethal effects of TNT to Xenopus laevis [19] and to Rana catesbeiana [20]. Given the broad public
INTRODUCTION
Manufacturing of explosives and their loading, assembling, and packing into munitions for use in testing on training sites and use in the battlefield has resulted in contamination of terrestrial and aquatic sites [1,2]. Some sites have been reported to contain explosives and associated compounds in soil, sediment, groundwater or surface water at concentrations that span several orders of magnitude [3,4]. Explosives compounds released from discarded shells as well as fragments of explosives formulations remaining following incomplete detonations may be present in surface soils and in aquatic habitats. Underwater unexploded and discarded munitions may also contribute to contamination of the water column from the slow release of explosives from corroded and breached shells [5–7]. Research on the ecotoxicity of explosive compounds has focused on munitions compounds, which have been heavily used by the military for many decades and continue to be used around the world [8–10]. Several insensitive munitions compounds that are chemically stable enough to withstand mechanical shocks without unintentional detonation have been developed and are being evaluated for future weapon systems to replace more sensitive explosives. Among these are 2,4dinitroanisole (DNAN) as a substitute of 2,4,6-trinitrotoluene (TNT) and 3-nitro-1,2,4-trizole-5-one (NTO) as a replacement for hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), a polynitramine compound [11]. Little is known about the aquatic toxicity * Address correspondence to
[email protected]. Published online 18 December 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2863 880
Energetic compound bioaccumulation kinetics in tadpoles
concern over reported declines in global amphibian populations and the suspected contribution of chemical contamination to the complex interaction of stressors contributing to those declines [21], investigations of the potential for ecototoxicological effects of explosives on the sensitive larval life stages of amphibians were pursued. The present study investigated the Northern leopard frog, Rana pipiens, an important ecological indicator within North America and a closely-related phylogenetic relative of Rana capito [22], a candidate for endangered species listing and inhabitant of military installations across the southeastern United States [23]. The objective of the present study was to develop a basic understanding of the bioconcentration potential and the bioaccumulation kinetics (uptake, elimination and biotransformation) of TNT, RDX, and DNAN in R. pipiens tadpoles using multi-time-point exposure to sublethal concentrations of each compound. In addition, the potential of NTO to bioconcentrate in R. pipiens was investigated in a single time point exposure. The present study was conducted in coordination with an investigation of the acute and chronic toxicity of TNT, RDX, DNAN, and NTO in R. pipiens tadpoles [24] to comprehensively assess potential for biological impacts. MATERIALS AND METHODS
Chemicals and experimental organisms
The energetic compounds TNT, RDX, DNAN, and NTO were obtained from the Holston Army Ammunition Plant (Kingsport, TN, USA). Egg masses of the leopard frog (R. pipiens) were purchased from a commercial supplier (Nasco) and hatched in the laboratory. Tadpoles were fed ad libitum once a day a commercially available frog brittle for tadpoles (Nasco), and their diet was supplemented with washed organic lettuce. Animals were 7 d to 10 d old and within a 24-h age range at the initiation of each experiment. All animal testing was conducted using methods and protocols approved by the US Army Engineer Research and Development Center’s Institutional Animal Care and Use Committee (Protocol #EL-60092011-3). TNT, RDX, and DNAN accumulation kinetics experiments
To obviate the need to add organic solvents directly to exposure water, solve-carrier-free solutions of TNT, RDX, and DNAN were created in water using a “shell coating” technique [25]. Briefly, the appropriate amount of chemical dissolved in acetone was added to 16-L glass carboys forming a thin layer around the sides of the container. Following complete evaporation of the solvent, 15 L of dechlorinated water (Vicksburg, MS, USA, municipal tap water dechlorinated using activated carbon filtration) was added and stirred using a magnetic stir plate for 24 h. After obtaining concentration measurement (analytical methods described below), stock solutions were mixed with uncontaminated dechlorinated water to create target exposure water concentrations of 1 mg L1 for TNT, 10 mg L1 for RDX, and 6 mg L1 for DNAN selected based on no-observed-effects levels determined by Stanley et al. [24]. Uptake and elimination kinetics experiments were conducted under static conditions at 23 8C in 40-L glass aquaria containing 30 L of exposure solution. For each chemical investigated, 2 aquaria containing exposure solutions were set up for each chemical: 1 for multi-time-point sampling for uptake kinetics determination and 1 contaminant bioaccumulation prior to transferring to uncontaminated water for elimination kinetics
Environ Toxicol Chem 34, 2015
881
determination. Tadpoles (individual weight, 100–140 mg) were added to the exposure tanks, 88 in the uptake kinetics tank and 100 in the tank used for contaminant bioaccumulation prior to elimination kinetics determination. Four sets of 3 arbitrarily selected tadpoles were taken from uptake kinetics tanks after 1 h, 3 h, 6 h, 12 h, and 24 h representing 4 replicates for body residue determination. Four additional individuals were collected and archived at each time point for gene expression analysis (K. Gust, unpublished manuscript). This exposure design was based on those used previously for assessing the toxicokinetics of munitions compounds in fish [26] and mussel [27]. Tadpole samples for body residue analysis were blotted dry on paper towel, placed in preweighed bead beater vials, weighed, and stored frozen at –20 8C. Following a 24-h contaminant loading phase conducted in parallel with the uptake kinetics exposure, tadpoles were transferred to 40-L glass aquaria containing 30 L of uncontaminated dechlorinated water, 1 per chemical investigated. Four sets of 3 arbitrarily selected tadpoles were taken from elimination kinetics tanks after 0 h, 1 h, 3 h, 6 h, 12 h, and 24 h (for TNT and RDX) or after 0 h, 0.5 h, 1 h, 2 h, 4 h, and 8 h (for DNAN), representing 4 replicates for body residue determination. The animals were fed Artemia nauplii approximately midway through the exposure period. Aquaria were held at 23 1 8C temperature. Confirmatory water samples (1 mL) were collected at initiation and termination of the uptake and elimination kinetics experiments to confirm concentrations during uptake and loading, and confirm lack of measurable explosives in the water at initiation of the elimination kinetics exposure. Water samples were filtered to 0.45 mm and immediately stored at 4 8C in the dark for future solvent extraction and chemical analysis. Preliminary assessment of NTO bioconcentration potential
Twenty tadpoles weighing approximately 140 mg were exposed to a single concentration (100 mg L1) in duplicate using 1-L beakers for 24 h. The exposure solution was one of the toxicity test treatments described in [24], where acute exposure (96 h) with NTO indicated no impacts to survival up to a treatment level of 500 mg L1. The NTO was dissolved directly into water with no carrier solvent using magnetic stirring for 24 h before use. Because of the acidic nature of NTO, NTO solutions were pH adjusted to a pH of 7.5 using NaOH. All surviving tadpoles from each beaker were pooled, blotted dry on paper towel, placed in preweighed bead beater vials, weighed, and stored frozen at –20 8C prior to body burden analysis. Analytical chemistry
Water samples. Aqueous samples were analyzed for TNT, the TNT major degradation products aminodinitrotoluenes (ADNTs) and diaminonitrotoluenes (DANTs), RDX, HMX (a potential impurity present in the RDX stock), and DNAN using a modified version of United States Environmental Protection Agency method 8330 A [28]. Chemical analysis of water samples was conducted using an Agilent 1100 highperformance liquid chromatograph (HPLC) equipped with a diode-array detector. A Supelco RPAmide C-16 column was used to separate analytes. A sample injection volume of 100 mL was used with a flow rate of 1 mL/min. Solvent ratios were 45% water and 55% methanol (isocratic elution), and ultraviolet absorbance was measured at 230 nm. Laboratory reporting
882
Environ Toxicol Chem 34, 2015
G.R. Lotufo et al.
limits for the analysis of water samples using this method were 0.1 mg L1 for all analytes. Tadpoles from the TNT, RDX, and DNAN experiments. After removal from the freezer, each bead beater vial containing tadpole samples (120–510 mg) received 1.0 mm glass beads and 0.5 mL HPLC-grade acetonitrile. Samples were homogenized on a mini-bead beater (Biospec) for 180 s at 4200 oscillations/min and sonicated (Branson 3200; Branson Ultrasonics) for 1 h at 18 8C in a water bath (Neslab RTE-111; Neslab). Samples were centrifuged for 10 min at 7500 g at 4 8C. Aliquots of the supernatants (0.25 mL) added to plastic syringes, received 0.25 mL of a 1% CaCl2 solution and were filtered through 0.45 mm polytetrafluoroethylene filters (Nalge Nunc) into amber sample vials, and refrigerated. Analytes were isolated and identified as described for the water samples. Laboratory reporting limits for the analysis of tissue samples using this method were approximately 0.4 mg g1 for all analytes. Tadpoles from the NTO experiment. Tadpoles exposed to NTO were homogenized in methanol as described in Tadpoles from the TNT, RDX, and DNAN experiments. The supernatants were analyzed using an Agilent 1200 HPLC equipped with a with a Thermo Hypercarb column (5 mm, 100 mm 3 mm). A sample injection volume of 50 mL was used with a flow rate of 0.8 mL/min. The isocratic mobile phase consisted of a 1:1 ratio of acetonitrile containing 1% trifluoroacetic acid and distilled water containing 1% trifluoroacetic acid. Ultraviolet absorbance was measured at 315 nm. Analyte confirmation by mass spectrometric analysis was carried out using a Bruker Daltonics Esquire 6000 ion trap mass spectrometer equipped with an electrospray ion source. The operating conditions for the mass spectrometer are described in Russell et al. [29]. Data analysis
Bioaccumulation factors (BCFs) for each compound were calculated as Css Cw
Steady-state BCF ¼
ð3Þ
where Css is the concentration of explosive compounds in the tissue at steady state, defined in the present study as the mean concentration of all tadpoles from the uptake exposure. The experimentally measured elimination rate (ke(m)) was determined by fitting the data from the elimination exposure to the first-order decay in Equation 4 [31] Cðt¼tÞ ¼ Cðt¼0Þ eke ðmÞt a a
ð4Þ
where Ca(t ¼ 0) is the initial chemical concentration in the tadpoles. The biological half-lives of the chemicals (t1/2) were determined using Equation 5 [31] t1=2 ¼
0:693 ke ðmÞ
ð5Þ
The time for the body residue to achieve 95% of steady-state concentration (TSS95%) was calculated by Equation 6 [31] TSS95% ¼
2:99 ke ðmÞ
ð6Þ
The data were fit by nonlinear regression analyses using SigmaPlot (Release 10.0; SPSS). RESULTS
Exposure water
Changes in body burdens were tested using one-way analysis of variance across time points (1 h, 3 h, 6 h, 12 h, and 24 h) from the uptake exposure. In addition, toxicokinentic parameters were established using Equation 1. During the uptake phase, the compounds enter the tadpoles at a rate characterized by the uptake clearance rate constant (ku). Similarly, the elimination rate constant (ke) describes the loss of the chemicals from the tadpoles. Rate constants were estimated by numeric integration using the differential form of the first-order, 2 compartment model containing water and organism compartments [30] and using water concentration and temporal bioaccumulation data from the uptake phase of the toxicokinetic experiment dCa ¼ ðku Cw Þ ðke Ca Þ dt
ð1Þ
In Equation 1, Ca is the concentration of chemical in the animal (mg kg1), Cw is the concentration of the chemical in water (mg L1), ku is the uptake clearance coefficient (L kg1 h1), ke is the chemical elimination rate constant (h1), and t is time (h). Because Cw was essentially kept constant (confirmed experimentally) and Ca was not detectable at t ¼ 0, Equation 1 can then be expressed as Ca ¼
k u Cw ð1 eke t Þ ke
ð2Þ
Water concentrations measured during the uptake phase of the toxicokinetics experiments (Table 1) varied minimally during the 24-h uptake exposure period, with compound loss remaining below 4% in all exposures, meeting the 2compartment model criterion for constant exposure [30]. Detectable concentrations of TNT transformation products in the exposure water were not observed at any time point during the uptake phase. The concentration of HMX in the exposure water during the uptake phase was measured at 0.6% of the total (RDX þ HMX) concentration. In the elimination phase, the
Table 1. Mean and percent decrease in concentration of explosives in the water sampled at the initiation and termination of the uptake exposure of the toxicokinetics experiments using larval Rana pipensa
Compound TNT RDX DNAN
Concentration (mg L1)
Percent decrease
Solubility at 25 8C (mg L1)
4-d LC50 or LOEC (mg L1)b
0.9 9.7 6.2
3.4 0 1.0
130c 56.3c 216d
4.4 (LC50) 25.3 (LOEC) 24.3 (LC50)
a The solubility limits and effects concentrations, expressed as median lethal concentration (LC50) or lowest-observed-effect concentration (LOEC) for survival in 4-d exposures, are presented for comparison purposes. b See Stanley et al. [24]. c See US Environmental Protection Agency [43]. d See Boddu et al. [16]. TNT ¼ 2,4,6-trinitrotoluene; RDX ¼ hexahydro-1,3,5-trinitro-1,3,5-triazine; DNAN ¼ 2,4-dinitroanisole.
Energetic compound bioaccumulation kinetics in tadpoles
Environ Toxicol Chem 34, 2015
Figure 1. Mean relative percentage of trinitrotoluene (TNT) and its major transformation products, 2-aminodinitrotoluene (2-ADNT) and 4-aminodinitrotoluene (4-ADNT), measured in larval Rana pipens tissues exposed to water amended with TNT at different time points during the uptake exposure. Error bars represent 1 standard deviation.
water concentrations of the explosives were below the laboratory reporting limit (0.1 mg L1), and therefore, the amounts of contaminant potentially available for re-uptake were negligible. The measured concentration of NTO in the exposure water was 100.1 mg L1. Water quality was found to be within targeted ranges [32] for all exposure tanks (temperature range, 23.1–24.0 8C; dissolved oxygen range, 7.1–8.4 mg L1; unionized ammonia, 0.05). For the sake of providing rough estimates for comparative purposes, the toxicokinetics parameters ku and ke derived using all data points (TNT) or using only mean concentrations (DNAN) are reported in Table 2. Elimination rates measured using rate of loss of compound following exposure and transfer to clean water were very fast for all the explosives investigated, resulting in the prediction of very fast times to approach steady state. The elimination experiments showed undetectable body burden of STNT, RDX, and DNAN within 6 h or less (30 min for DNAN), on transferring to clean water (Figure 2). The fast elimination of TNT and RDX was associated with half-lives of approximately 1 h (Table 2), indicating that release from exposure results in very fast loss of those compounds from the tadpole tissues. The experimentally measured rate elimination could not be determined for DNAN because that compound was below the level of detection in tadpoles sampled after 30 min, the earliest sampling period. The fast elimination of TNT and RDX indicate that the estimated time to approach steady state is very short, 5 h or less (Table 2). The BCF values were relatively low and overall similar for all 3 chemicals investigated (Table 2). The concentration of STNT and DNAN in tadpole tissue (as mg kg1) remained below the concentration of TNT and DNAN in the exposure water (as mg L1). Bioconcentration of NTO
Following a 24-h exposure to 100 mg L1 to NTO, no mortality was observed and the body residue in tadpoles was 25 mg kg1. This value was associated with a BCF value of 0.25 L kg1. Because of the observed dilution of the concentration of NTO in the tissues relative to the source concentration in the water (1:4), this compound was not further evaluated for uptake and elimination kinetics. DISCUSSION
The explosive compounds TNT, RDX, and DNAN are very weakly hydrophobic (low KOW) and are predicted to have a low tendency to bioconcentrate [8]. The present study showed low rates of uptake and very high rates of elimination in R. pipiens tadpoles, which resulted in very low BCFs (
