Environmentally relevant concentrations of ... - Wiley Online Library

Environmental Toxicology and Chemistry, Vol. 21, No. 2, pp. 424–430, 2002 q 2002 SETAC Printed in the USA 0730-7268/02 $9.00 1 .00

ENVIRONMENTALLY RELEVANT CONCENTRATIONS OF AMMONIUM PERCHLORATE INHIBIT DEVELOPMENT AND METAMORPHOSIS IN XENOPUS LAEVIS WANDA L. GOLEMAN,† LINA J. URQUIDI,† TODD A. ANDERSON,‡ ERNEST E. SMITH,‡ RONALD J. KENDALL,‡ and JAMES A. CARR*†‡ †Department of Biological Sciences, Texas Tech University, Lubbock, Texas 79409, USA ‡The Institute of Environmental and Human Health, Texas Tech University, Lubbock, Texas 79416, USA ( Received 11 April 2001; Accepted 7 August 2001) Abstract—We determined whether environmentally relevant concentrations of ammonium perchlorate alter development and metamorphosis in Xenopus laevis. Eggs and larvae were exposed to varying concentrations of ammonium perchlorate or control medium for 70 d. Most treatment-related mortality was observed within 5 d after exposure and was due in large part to reduced hatching success. The 5- and 70-d median lethal concentrations (LC50s) were 510 6 36 mg ammonium perchlorate/L and 223 6 13 mg ammonium perchlorate/L, respectively. Ammonium perchlorate did not cause any concentration-related developmental abnormalities at concentrations below the 70-d LC50. Ammonium perchlorate inhibited metamorphosis in a concentration-dependent manner as evident from effects on forelimb emergence, tail resorption, and hindlimb growth. These effects were observed after exposure to ammonium perchlorate concentrations in the parts-per-billion range, at or below concentrations reported in surface waters contaminated with ammonium perchlorate. Ammonium perchlorate significantly inhibited tail resorption after a 14-d exposure in the U.S. Environmental Protection Agency (U.S. EPA) Endocrine Disruptor Screening and Testing Committee (EDSTAC) Tier I frog metamorphosis assay for thyroid disruption in amphibians. We believe that ammonium perchlorate may pose a threat to normal development and growth in natural amphibian populations. Keywords—Thyroid

Anuran

Perchlorate

Amphibian decline

Metamorphosis

USA [8]. In contrast, two other studies found no link between AP-contaminated drinking water and neonatal hypothyroidism [9,10]. Ionic perchlorate competitively inhibits thyroidal iodide uptake in mammals, thus preventing synthesis of thyroid hormone (TH) [2,11]. The loss of negative feedback due to a decrease in serum thyroxine (T4) levels leads to stimulation of the thyroid follicular epithelium by elevated blood levels of thyroid-stimulating hormone (TSH) [12]. Sustained exposure to perchlorate leads to hypertrophy and hyperplasia of follicular cells, resulting in an increased thyroid weight and the development of goiter in rodents [13]. In addition to blocking iodide uptake, perchlorate also induces an efflux of internal thyroid iodide [2,11]. Perchlorate also disrupts thyroid accumulation of iodide in nonmammals, including amphibians and lampreys [14–16]. Perchlorate inhibits TH-dependent aspects of development and metamorphosis in anuran [14,15,17] and urodele [18] larvae, leading to oversize larvae if exposure is continued for long periods of time [14]. Although perchlorate salts, principally NaClO4 and KClO4, have been used experimentally for years to inhibit amphibian metamorphosis, the concentrations used to block metamorphosis are generally greater (250–1,000 mg/L) [15,18,19] than concentrations of perchlorate reported in contaminated surface and groundwaters [20]. In a recent study of surface waters and sediments at Longhorn Army Ammunition Plant (LHAAP) in Karnack, East Texas, USA [21], perchlorate levels as high as 31.2 6 0.21 mg perchlorate/L were reported. Whole-tissue perchlorate levels were 1,130 to 2,567 mg perchlorate/L in American bullfrog (Rana catesbeiana) tadpoles collected from an AP-contaminated pond at LHAAP [21]. Although the per-

INTRODUCTION

Perchloric acid and its salts, such as ammonium perchlorate (AP), are strong oxidizers and can ignite violently when combined with combustibles [1]. This characteristic has led to the use of AP in pyrotechnics, explosives, and jet and rocket fuels. Perchlorates also induce iodide discharge from the thyroid, leading to their clinical utility in treating certain types of hyperthyroidism [2,3]. Perchlorate occurs in ground- and surface waters [4], most likely as the result of AP discharge from rocket fuel manufacturing facilities or from the demilitarization of missiles. The AP is highly water-soluble, and perchlorate absorbs weakly to most soil minerals. Because reduction of the central chlorine atom in perchlorate occurs very slowly [4], this ion can persist in the environment for decades. Perchlorates also occur in natural mineral deposits used as nitrogen fertilizers, such as Chilean sodium nitrate [5,6] and New Mexican langbenite [7]. Although not a major component, a recent study of several fertilizers [7] suggests that the mineral deposits used in the manufacturing process of some fertilizers may be a source of perchlorate accumulation in the food chain. Whether perchlorate in drinking water represents a threat to humans is still hotly debated, even though AP contamination has been detected in a number of public water supply wells in California and surface and groundwaters in Arkansas, Maryland, New York, Texas, and Utah, USA [4]. A recent study found abnormal thyroid function in newborns from Arizona, USA, exposed to AP-contaminated water from the Colorado River, * To whom correspondence may be addressed ([email protected]). 424

Environ. Toxicol. Chem. 21, 2002

Developmental effects of ammonium perchlorate

chlorate levels found in contaminated surface waters at LHAAP [21] and elsewhere [20] are well below those traditionally known to prevent metamorphosis in experimental settings, the effects of environmentally relevant concentrations of AP on anuran metamorphosis are generally unknown. The present study has two goals. The first goal was to examine the effects of environmentally relevant AP concentrations on metamorphosis in Xenopus laevis, a standard laboratory model for studying anuran metamorphosis. The second goal was to compare the recently developed U.S. Environmental Protection Agency (U.S. EPA) Endocrine Disruptor Screening and Testing Committee (EDSTAC) Tier I screening test for frog metamorphosis [22] with longer-term exposure scenarios. Portions of this work were reported previously in abstract form [23]. METHODS

Test material Ammonium perchlorate (CAS 7790-98-9) with a purity of 99.999% was purchased from Aldrich Chemical (Milwaukee, WI, USA).

Animals Sexually mature male and female X. laevis were purchased from Xenopus Express (Homosassa, FL, USA). Adults were maintained in dechlorinated water on a 12:12-h light:dark regime (lights on 0700) and fed frog brittle (Nasco, Ft. Atkinson, WI, USA) three times weekly immediately following a water change. Water was changed three times each week. Prior to breeding, adults were allowed to acclimate in 45L glass aquaria containing 18 L Frog Embryo Teratogenesis Assay—Xenopus (FETAX) medium [24]: NaCl, 10.7 mM; NaHCO3, 1.14 mM, KCl, 0.4 mM; CaCl2, 0.14 mM; CaSO4, 0.35 mM, and MgSO4, 0.62 mM. The FETAX medium was prepared using deioinzed water passed through a 1.2-ft3 carbon filter immediately before use. Depending on the experiment, naturally fertilized eggs were obtained from one or three to five pairs of adults. Spawning was induced artificially by injecting adult X. laevis with human chorionic gonadotropin (Sigma Chemical, St. Louis, MO, USA) dissolved in 0.9% NaCl. The injections were administered via the dorsal lymph sac in a volume of 0.3 ml (males) or 0.5 ml (females). Males were injected with 150 IU followed by a second injection 24 h later. Females were injected with 250 IU followed by an injection of 500 IU 8 h later. Immediately following the second injection, males and females were placed together in 21-L glass breeding tanks with false bottoms made of silicone-coated 0.5-in hardware mesh. They were left there overnight to breed. Fertilized eggs were identified by visual observation with a binocular dissecting microscope. Eggs and tadpoles up to 5 d old were kept in 100 ml of FETAX medium or test solution in 250-ml glass beakers maintained in a standing water bath at 228C (628C) on a 12:12-h L:D regime. Five-day-old larvae were transferred to 21-L glass tanks containing 8 L of FETAX medium; the stocking density was seven tadpoles per liter. Tadpoles were fed 0.4 g of powdered frog brittle (Nasco) mixed in 2 ml of FETAX solution per tank every 72 h immediately following a 50% water change. All procedures involving X. laevis were approved by the Texas Tech Animal Care and Use Committee (Lubbock, TX, USA).

425

Analytical procedures Ammonia, pH, and specific conductance of the tank water were monitored every 7 d. Dissolved oxygen was monitored every 2 d, and water temperature was monitored daily. A YSIt (Yellow Springs, OH, USA) model 85 meter was used to monitor water temperature, percentage dissolved oxygen, specific conductivity, and salinity for each tank. Free ammonium ion and pH levels of the water in each tank were determined with a Hacht spectrophotometer model DR/2000 (Loveland, CO, USA) and an Oaktont pH meter (Vista, CA, USA), respectively. The Cooperative Extension Service of the University of Georgia (Athens, GA, USA) performed a chemical analysis of the food and water. Adult and larval frog brittle contained no detectable pesticide residues (limits of detection ranged from 0.01–0.2 ppm for 22 different pesticides). Powdered frog brittle (larvae food) contained low levels of barium (12 ppm), arsenic (0.39 ppm), and selenium (0.45 ppm); adult frog brittle contained barium (7.6 ppm), arsenic (0.44 ppm), and selenium (1.03 ppm). Water analysis reveled no detectable pesticides or heavy metals. Verification of AP in diluted source solutions was performed by ion chromatography [21,25]. Analyses of perchlorate ions was carried out using a Dionext DX-500 Ion Chromatography System equipped with a GP50 gradient pump, a CD20 conductivity detector, and an AS40 automated sampler (Dionex, Sunnyvale, CA, USA). PeakNett chromatography software (Dionex) was used to control the system. A Dionex IonPac AS16 (250 3 4 mm) analytical column was used for ion separation. Conditions for the system were as follows: run time 5 12:00 min, flow rate 5 1.0 ml/min, eluent 5 100 mM sodium hydroxide, and injection volume 5 1,000 ml. Ion detection was by suppressed conductivity in the external water mode. Using the analytical method described previously, the detection limit for perchlorate anion in water was 1 ppb. The lower limit of quantitation was 2.5 ppb.

Experiment 1 We determined the concentration-dependent effects of AP on X. laevis lethality, developmental abnormalities, and THdependent aspects of metamorphosis during a 70-d exposure. One range-finding and two definitive concentration experiments were performed. For the range-finding experiment, groups of approximately 50 fertilized eggs (Nieuwkoop–Faber [NF] stages 4–10 [26]) were exposed to one of seven concentrations of AP diluted in FETAX medium (target concentrations ranged from 1.175 ppb to 1,175 ppm in 10-fold incremental steps) or FETAX medium alone for 70 d, beginning ,24 h after oviposition. All treatments were performed in duplicate. A 50% change of test and control solutions was performed every 72 h. Contents of the test chambers were aerated continuously throughout the experiment. Based on the results of the range-finding experiment, two additional AP concentrations were added to better define the upper lethal range. For the definitive concentration experiments, groups of approximately 50 eggs were exposed to one of nine concentrations of AP diluted in FETAX medium. In this experiment, target concentrations of 1.18 ppb (5 6 2 ppb measured), 11.8 ppb (18 6 3 ppb measured), 118 ppb (147 6 6 ppb measured), 1175 ppb (1,412 6 32 ppb measured), 11.8 ppm (14.4 6 0.07 ppm measured), 118 ppm (133 6 2.5 ppm measured), 351 ppm (425 6 45 ppm measured), 585 ppm (nominal), and 1,175 ppm (nominal) were tested, as was FE-

426


TAX medium alone for 70 d (68 d posthatch), beginning ,24 h after oviposition. All treatments were performed in duplicate, and the entire experiment was replicated once. All AP dilutions were prepared from a single stock solution. Beginning on the day of hatch, hatching success (no. unhatched eggs/total no. eggs), deformities (no. showing bent tails/total hatched), edema (no. showing distention of body with fluid/total hatched), and abnormal swimming (no. showing abnormal swimming/total) were noted daily for each test and reference solution. For free-swimming larvae, percentage mortality (no. dead larvae/no. hatched), percentage showing deformities, percentage displaying abnormal swimming behavior, percentage demonstrating forelimb emergence (FLE; no. showing FLE/no. hatched), and percentage metamorphosed animals (no. showing complete tail resorption/no. hatched) were noted every day. Time to tail resorption for each froglet was recorded. Dead animals were removed and preserved in 10% neutral-buffered formalin. Snout-vent length (SVL), hindlimb length (HLL), tail length, and NF stage were determined every 5 d from 10 tadpoles per tank beginning 16 d after hatching. Animals were rapidly transferred by dip net to a petri dish containing ice-chilled FETAX under a dissecting microscope. Tail height and HLL (until NF stage 63) were measured using an ocular micrometer. A metric ruler was used to measure SVL and HLL after NF stage 63. Animals were returned to their home tank after measurements were completed. All experiments were terminated after 70 d. Unmetamorphosed tadpoles were euthanized by immersion in 3-aminobenzoic acid ethyl ester (MS-222, 1 g/L in distilled water), weighed, staged, and measured for SVL and HLL. Animals completing metamorphosis (NF stage 66) prior to posthatch day 68 (treatment day 70) were removed, euthanized in MS222, weighed, and measured for SVL and HLL. Measurements of SVL and HLL from these animals were included in the posthatch day 68 measurements. Animals were stored in 10% neutral-buffered formalin.

Experiment 2 We examined the ability of AP to inhibit tail resorption in X. laevis with the 14-d EDSTAC Tier I frog metamorphosis assay [22], based on methodology used by Fort and Stover [27]. Larvae were raised in plain FETAX medium as described in experiment 1 until they reached NF stage 60. Then the larvae were transferred to 21-L glass aquaria containing 9 L of premixed AP diluted in FETAX or FETAX medium alone. A 50% medium change was performed every 72 h. All AP dilutions were prepared from a single stock solution. The AP concentration (target concentration of 14,040 ppb) was selected based on AP concentrations in effluent from the burning ground number 3 groundwater treatment plant at LHAAP measured in April 1999 [28]. Time to metamorphosis for each froglet was recorded. The NF stage, SVL, HLL, tail length, and tail height were measured as described previously and recorded prior to exposure (day 0) and at the end of exposure (day 14) for each tadpole/froglet. On completion of the exposure, the animals were euthanized in MS-222 and stored in 10% neutral-buffered formalin.

Data analysis The LC50s were calculated independently for each of two definitive concentration experiments by the moving-angle-average method using SoftToxy software (ChemSW, Fairfield, CA, USA). The mean inhibitory concentration (IC50s) were

W.L. Goleman et al.

calculated from four-parameter logistic equations using SigmaPloty (SPSS, Chicago, IL, USA). Data were analyzed using Student’s two-tailed t test or by analysis of variance (ANOVA) followed by the Tukey–Kramer multiple comparison test. Data were tested for homogeneity of variance prior to ANOVA using Bartlett’s test. Data were compared with an F test for equal standard deviations prior to Student’s t test. Means with significantly different standard deviations were compared using the Mann–Whitney nonparametric t test. Data expressed as a proportion (FLE, tail resorption) were transformed (arcsine of the square root) before ANOVA. RESULTS

Experiment 1 Concentration-related effects of AP on hatching and mortality are shown in Figure 1. Five- and 70-d LC50s for AP averaged 510 6 36 and 223 6 13 ppm, respectively. Survival of larvae averaged 87 to 93% between 5 ppb and 133 ppm but decreased to 6 to 7% in the 425-ppm treatment group. Hatching success was reduced above 1,000 ppm (Fig. 1). Data on abnormalities in AP-treated larvae are shown in Table 1. The percentage of animals with bent tails, edema, and abnormal swimming increased at concentrations $425 ppm. The AP affected larval growth, although the effects on hindlimb growth were particularly robust. Exposure to 425 ppm reduced SVL compared to controls at posthatch days 16 (F7,24 5 9.7, p , 0.001), 21 (F7,24 5 6.6, p , 0.01), 31 (F7,24 5 3.7, p , 0.05), and 36 (F7,24 5 4.3, p , 0.05) (Fig. 2). No significant effects of AP on SVL were observed at concentrations below 425 ppm. In contrast, AP exposure reduced hindlimb growth in a concentration-dependent manner at posthatch days 55 (F7,24 5 7.9, p , 0.0001), 60 (F7,24 5 7.846, p , 0.0001), 65 (F7,24 5 7.765, p , 0.0001), and 68 (F7,24 5 10.1, p , 0.0001) (Fig. 2). The IC50 for AP inhibition of hindlimb growth (at day 68 posthatch) was 16.7 ppb. Body weight and total length measurements were not included in analysis of growth affects because significant reductions in both parameters occur during metamorphosis that would confound analysis of any potential treatment effects. Strong concentration-dependent effects of AP were observed on the percentage of animals reaching FLE and the percentage of animals completing metamorphosis (Fig. 3). Forelimb emergence is a thyroid-hormone-dependent process that marks the beginning of metamorphic climax in X. laevis and other anuran species, whereas tail resorption marks the completion of metamorphosis. One-way ANOVA revealed a significant inhibitory effect of AP on FLE when data from both definitive concentration experiments were combined (Fig. 3; F7,24 5 38.0, p , 0.0001). Under the rearing conditions described here, the lowest concentration of AP capable of inhibiting FLE was 5 ppb (Fig. 3). At concentrations of 1 ppm and above, AP completely prevented FLE (Fig. 3). The IC50 for AP inhibition of FLE (at day 68 posthatch) was 33 ppb. Likewise, AP significantly reduced tail resorption in a concentration-related fashion (F7,24 5 13.3, p , 0.0001; Fig. 3). The lowest concentration of AP inhibiting tail resorption was 18 ppb. At AP concentrations of 147 ppb and above, none of the animals completed tail resorption.

Experiment 2 Results of the EDSTAC Tier I frog metamorphosis test are shown in Table 2. The data presented in Table 2 are pooled


Developmental effects of ammonium perchlorate

427

Table 1. Percentage deformities in surviving Xenopus laevis tadpoles exposed to ammonium perchlorate (AP) for 70 da Trial 2

Trial 1 Treatment FETAXb AP (ppm) 0.005 0.018 0.147 1.41 14.4 133 425

Bent tails

Abnormal Edema swimming

Bent tails

Abnormal Edema swimming

0

1

0

1

0

3

2 4 2 1 2 2 26

4 3 5 5 4 1 20

6 5 5 6 3 4 88

0 1 3 5 4 0 3

6 1 5 7 6 2 14

6 6 5 11 14 15 55

a

Values are the mean of two replicates. Sample size per replicate was 38 to 50/replicate for trial 1 and 25 to 36 for trial 2 in all but the highest concentration, which had a sample size of less than five for both trials. b FETAX 5 Frog Embryo Teratogenesis Assay—Xenopus.

Fig. 1. Percentage mortality (●) and hatching (n) in Xenopus laevis larvae exposed to ammonium perchlorate for 70 d. Fifty-one embryos were placed in frog embryo teratogenesis assay—Xenopus (FETAX) medium or one of nine concentrations of ammonium perchlorate within 24 h after fertilization. Data from two independent trials are presented. Values are the means of duplicate treatments.

from two independent experiments. Mean tail lengths were identical between the two treatment groups at the start of each experiment. The AP reduced the percentage of tadpoles completing tail resorption. Mean tail length was significantly greater in AP-treated versus control tadpoles at the end of the 14d exposure (Table 2). It is important to note that 14-d exposure of NF stage 60 tadpoles to AP did not completely prevent tail resorption, as mean tail length in AP-treated tadpoles at the end of the experiment was significantly less than tail length on day 0 (Table 2). DISCUSSION

A preliminary study on the lethality of AP [29] reported an LC50 of 496 mg/L for AP in the FETAX assay, a 4-d assay

[30] that does not include the later stages of metamorphosis. Five-day LC50 values (510 6 36 mg AP/L) from the definitive concentration experiments performed in the present study are in close agreement with this value. The LC50 values obtained for the entire 70-d study were lower (223 6 13 mg AP/L), probably reflecting the low 70-d survivorship observed in the 425-ppm treatment group. This finding also suggests that short exposure windows, such as those used in FETAX-like assays, may not accurately predict the effects of chronic exposure. Although an increased incidence of abnormalities (percentage bent tails, percentage edema, percentage abnormal swimming) was evident in larvae exposed to 425 ppm AP, this concentration is well above the 70-d LC50. The apparent lack of abnormalities below this concentration suggests that AP is not teratogenic in X. laevis. We did not directly assess the contribution of ammonium ions to the lethality of AP, but preliminary data from our laboratory indicate a 5-d LC50 of 189 mg/ L for ammonium chloride and an LC50 of 1,590 mg/L for sodium perchlorate using the same protocol described in the present study. This preliminary LC50 for ammonium chloride is comparable to the 4-d LC50 of 127.5 mg/L reported for ammonium chloride in X. laevis [31]. The fact that the 5-d LC50 calculated for AP is closer to that of ammonium chloride than to that of sodium perchlorate suggests that ammonium ions may significantly contribute to the lethality of AP. Although the FETAX assay is suitable for examining early embryonic effects of xenobiotics, it cannot be used to gauge the activity of chemicals that disrupt thyroid activity because thyroid gland development occurs well after the 96-h window of the FETAX assay. In the present study, we used two different approaches to address the effects of environmentally relevant concentrations of AP on TH-sensitive indices of development. First, we exposed developing larvae to environmentally relevant AP concentrations over a 70-d period, from egg to metamorphosis. Second, we exposed tadpoles to AP using the 14-d EDSTAC Tier I frog metamorphosis assay. To our knowledge, this is the first study to examine the concentration-related effects of AP throughout the larval period in any amphibian species. Although perchlorate salts have been used for more than 50 years to inhibit metamorphosis, the concentrations used are generally orders of magnitude greater (250–1,000 mg/L; see [15,18,19]) than even the highest perchlorate values reported in surface waters. Our results indicate

428


Fig. 2. Snout-vent (A) and hindlimb (B) lengths in Xenopus laevis larvae. Measurements were made every 5 d starting 16 d posthatch. Values are means of four replicates per treatment pooled from two independent trials. Sample size per replicate was 25 to 50 in all but the 425-ppm group, which had a sample size of less than five. Standard error bars are omitted to reduce clutter. Variance around means was compared by one-way analysis of variance. Asterisks indicate significantly different from frog embryo teratogenesis assay—Xenopus (FETAX) controls as determined by the Tukey–Kramer multiple comparisons test (*, p , 0.05; **, p , 0.01; ***, p , 0.001). Stacked asterisks in Figure 2B indicate 147 ppb to 425 ppm AP significantly different from FETAX control (p , 0.01). FETAX control (●), 5 ppb AP (n), 18 ppb (m), 147 ppb (.), 1.41 ppm (V), 14.4 ppm (□), 133 ppm (n), and 425 ppm (,).

that AP inhibits TH-dependent aspects of metamorphosis (FLE, tail resorption) and growth (HLL) at concentrations in the micrograms-per-liter range, well within the range of concentrations reported for AP levels in contaminated surface waters. Although we cannot yet eliminate the possibility that ammonium ions contributed directly to the effects of AP on

W.L. Goleman et al.

Fig. 3. Percentage forelimb emergence (A) and percentage completing tail resorption (B) in Xenopus laevis larvae exposed to ammonium perchlorate for 70 d. Values are the means 6 standard error of means of measurements made on treatment day 70 from four replicates per treatment pooled from two independent trials. Sample size per replicate was 25 to 50 in all but the 425-ppm group, which had a sample size of less than five. Variance around means was compared by analysis of variance. Asterisks indicate significantly different from frog embryo teratogenesis assay—Xenopus (FETAX) controls as determined by the Tukey–Kramer multiple comparisons test (*, p , 0.05; **, p , 0.01; ***, p , 0.001).

growth and metamorphosis, the fact that AP specifically affected TH-sensitive aspects of growth and metamorphosis argues against nonspecific effects of AP due to ammonium-ion toxicity. For example, AP in the parts-per-billion concentration range inhibited hindlimb growth but did not affect overall somatic growth, as evidenced by no affect on SVL (Fig. 2). Anuran hindlimbs are exquisitely sensitive to changes in circulating TH and require TH for normal development [32]. The


Developmental effects of ammonium perchlorate Table 2. Mean tail length standard error of means (6 SE) and percentage completing tail resorption in Nieuwkoop–Faber stage 60 Xenous laevis tadpoles exposed to ammonium perchlorate (AP) for 14 da Mean tail length (mm) AP (ppm)

n

Day 0

Day 14

Tail resorption (%)

0 19.8 6 6.70

24 24

37.4 6 0.74 36.6 6 0.63

0.02 6 0.02 6.12 6 1.70b

96 17

a

The experiment was carried out using recommendations based on the Endocrine Disrupter Screening and Testing Committee Tier I frog metamorphosis assay. b Significantly different from control, p , 0.0001, based on Mann– Whitney nonparametric t-test.

only effect of AP on SVL occurred early in the developmental period in the 425-ppm treatment group, a concentration that also resulted in 94 to 95% mortality over the 70-d exposure period. The frog metamorphosis assay [22] has been recommended as one of two Tier I assays for screening affects of potential endocrine-disrupting chemicals on the thyroid axis. This test was not designed to detect agents interfering with metamorphosis: It was designed to detect only agents affecting the thyroid axis. The ability of the frog metamorphosis test to screen for effects on TH synthesis has not been addressed previously. Our data suggest that the EDSTAC frog metamorphosis assay may not be sensitive enough to detect agents acting directly on TH synthesis, at least as the test is presently designed. For example, exposure to a concentration of AP (19.8 6 6.7 mg perchlorate/L) that completely inhibited tail resorption in the 70-d assay (Fig. 3) reduced but did not prevent tail resorption in the 14-d test period. One possible reason for the difference in sensitivity between the 70-d assay and the EDSTAC 14-d frog metamorphosis assay is the fact that the latter test employs NF stage 60 tadpoles. By NF stage 60, blood levels of TH are already rising, reaching maximal levels around NF stage 62 [32,33]. Thyroid hormone can be stored in the thyroid gland as colloid or bound to transport proteins in the blood. Thus, inhibition of TH synthesis may not affect TH availability to target tissues in the short exposure duration available in the EDSTAC frog metamorphosis test. Consistent with this idea are recent data from Lawrence et al. [34], showing that perchlorate (10 mg/d) administered in drinking water over 14 d had no effect on circulating TH or TSH concentrations in human adult males, even though thyroid 123I uptake was decreased by 38% at termination of dosing. This issue was avoided in our 70-d exposure studies because tadpoles were exposed to AP before differentiation of the thyroid. In fact, development of surviving larvae exposed for 70 d to AP concentrations greater than 117.5 mg/L was halted at NF stages 51 to 52 (data not shown), immediately preceding the stage when thyroid differentiation is completed (NF stage 53) in X. laevis [26]. Perchlorate is the most potent known inhibitor of the mammalian sodium-dependent iodide symporter [35]. The mechanism of AP action on frog development almost certainly involves a direct inhibition of iodide uptake by the thyroid gland, although indirect effects on iodide transport by the gastrointestinal tract and possibly other tissues cannot be ruled out. Sodium-dependent iodide transport has been demonstrated in a number of extrathyroidal tissues in mammals, including cho-

429

roid plexus, salivary glands, and renal tubules [36]. Nonetheless, the suite of effects that occur in response to AP (Figs. 2 and 3) strongly implicates disruption of TH synthesis early in X. laevis development during the 70-d exposure studies. It is generally believed that TH synthesis in amphibians is identical to that in mammals [12], although direct similarities in sequence and function between the mammalian and amphibian transport proteins await the sequencing and cloning of the amphibian transporter. Two key features of sodium-dependent iodide transport are relevant to the possible mechanism of AP action on frog development as observed in the present study. First, substrate binding to the iodide transporter is saturable [35]. This may explain the lack of a concentration response above 147 mg/L perchlorate for effects on tail resorption, percentage FLE, and HLL. Second, the degree to which perchlorate inhibits iodide transport is likely to depend not only on the concentration of AP and duration of AP exposure but also on the degree of iodide uptake. In our studies, the only source of iodide available to the developing larvae was in the food supply. This may explain the low IC50s for the effects of AP on percentage FLE and HLL. The fact that the IC50s for AP inhibition of FLE and hindlimb growth are relatively similar also argues for a common mechanism of AP action, such as inhibition of iodide uptake. Xenopus laevis is a useful lab model but is not native to North American surface waters (isolated feral populations have been reported in Southern California and Florida, USA). Thus, it remains to be determined if AP represents a threat to natural amphibian populations. Given the recent findings of AP in surface waters and sediments at LHAAP, at concentrations ranging from 70 mg perchlorate/L to greater than 30 mg perchlorate/L at one site [21], the possibility of threat seems plausible. The LHAAP site containing AP at 30 mg/L perchlorate serves as a breeding site for at least two ranid species, R. catesbeiana and the bronze frog, Rana clamitans clamitans (personal observations). The range of perchlorate concentrations found in surface waters and sediments at LHAAP are well within the range of those that prevent metamorphosis during a 70-d exposure in X. laevis (Fig. 3). In addition to the gross developmental affects of AP noted in the present study, delayed consequences associated with inhibiting TH may occur that are not manifest until sexual maturity. For example, lack of TH at critical stages during development can prevent masculinization of laryngeal muscles involved in calling [37]. The possibility that AP has long-term effects on reproductive success is a topic for future research.

Acknowledgement—We would like to acknowledge the help of Catherine Bens, Cynthia McMurry, Elizabeth Maull, Reynaldo Patin˜o, and Nick Parker. This research was supported by funding from the U.S. Department of Defense, through the Strategic Environmental Research and Development Program, under a cooperative agreement with the U.S. Air Force. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of the 311 Human Systems Wing/Air Force Institute for Environment, Safety, and Occupational Health Risk Analysis of the U.S. government. REFERENCES 1. Von Burg R. 1995. Perchlorates. J Appl Toxicol 15:237–241. 2. Stanbury JB, Wyngaarden JB. 1952. Effect of perchlorate on the human thyroid gland. Metabolism 1:533–539. 3. Bartalena L, Brogioni S, Grasso L, Bogazzi F, Bureli A, Martino E. 1996. Treatment of amiodarone-induced thyrotoxicosis, a dif-

430

4. 5. 6. 7. 8.

9.

10. 11. 12. 13.

14.

15. 16.

17. 18. 19.

20. 21.

Environ. Toxicol. Chem. 21, 2002 ficult challenge: Results of a prospective study. J Clin Endocrinol Metab 81:2930–2933. Urbansky ET. 1998. Perchlorate chemistry: Implications for analysis and remediation. Bioremediation J 2:81–95. Tollenaar H, Martin CB. 1972. Perchlorate in Chilean nitrate as the cause of leaf rugosity in soybean plants in Chile. Phytopathology 62:1164–1166. Levens L. 1960. Biological action of perchlorates. In Schumacher JC, ed, Perchlorates: Their Properties, Manufacture and Uses. Reinhold, New York, NY, USA, pp 168–186. Susarla S, Collette TW, Garrison AW, Wolfe NL, McCuthcheon SC. 1999. Perchlorate identification in fertilizers. Environ Sci Technol 33:3469–3472. Brechner RJ, Parkhurst GD, Humble WO, Brown MB, Herman WH. 2000. Ammonium perchlorate contamination of Colorado River drinking water is associated with abnormal thyroid function in newborns in Arizona. J Occup Environ Med 42:777–782. Xiao Li F, Byrd DM, Deyhle GM, Sesser DE, Skeels MR, Katkowsky SR, Lamm SH. 2000. Neonatal thyroid-stimulating hormone level and perchlorate in drinking water. Teratology 62:429– 431. Crump C, et al.2000. Does perchlorate in drinking water affect thyroid function in newborns or school-age children? J Occup Environ Med 42:603–612. Wolff J. 1998. Perchlorate and the thyroid gland. Pharmacol Rev 50:89–105. Norris DO. 1997. The hypothalamo-hypophysial-thyroid axis of mammals. In Vertebrate Endocrinology, 3rd ed. Academic, San Diego, CA, USA, pp 242–267. Capen CC. 1996. Toxic responses of the endocrine system. In Klaassen CD, ed, Casarett and Doull’s Toxicology: The Basic Science of Poisons, 5th ed. McGraw-Hill, New York, NY, USA, pp 617–640. Buscaglia M, Leloup J, De luze A.1985. The role and regulation of monoiodination of thyroxine to 3,5,39-triiodothyronine during amphibian metamorphosis. In Balls M, Bownes M, eds, Metamorphosis. Oxford University Press, New York, NY, USA, pp 273–294. Miranda LA, Pisano A, Casco V. 1996. Ultrastructural study of thyroid glands of Bufo arenarum larvae kept in potassium perchlorate solution. Biocell 20:147–153. Manzon RG, Youson JH. 1997. The effects of exogenous thyroxine (T4) on triiodothyronine (T3), in the presence or absence of potassium perchlorate, on the incidence of metamorphosis and on serum T4 and T3 concentrations in larval sea lampreys (Petromyzon marinus L). Gen Comp Endocrinol 106:211–220. Galton VA. 1988. The role of thyroid hormone in amphibian development. Am Zool 28:309–318. Brown DD. 1997. The role of thyroid hormone in zebrafish and axolotl development. Proc Natl Acad Sci USA 94:13011–13016. Rollins-Smith LA, Flajnik MF, Blair PJ, Davis AT, Green WF. 1997. Involvement of thyroid hormones in the expression of MHC class I antigens during ontogeny in Xenopus. Dev Immunol 5: 133–144. Gullick RW, LeChevallier MW, Barhorst TS. 2001. Occurrence of perchlorate in drinking water sources. J Am Water Works Assoc 93:66–77. Smith PN, Theodorakis CW, Anderson TA, Kendall RJ. 2001. Preliminary assessment of perchlorate in ecological receptors at

W.L. Goleman et al.

22. 23.

24.

25. 26.

27.

28.

29. 30.

31.

32. 33. 34. 35. 36. 37.

the Longhorn Army Ammunition Plant (LHAAP), Karnack, Texas. Ecotoxicology 10:305–313. Federal Register. 1998. Endocrine disruptor screening program: Statement of policy and priority-setting workshop notice. Fed Reg 63:71542–71568. Goleman WL, Urquidi LJ, McMurry CS, Anderson TA, Kendall RJ, Carr JA. 2001. Effects of environmentally relevant concentrations of ammonium perchlorate on survival and metamorphosis in Xenopus laevis. Am Zool 40:1030–1031. Dawson DA, Bantle JA. 1987. Development of a reconstituted water medium and preliminary validation of the frog embryo teratogenesis assay—Xenopus (FETAX). J Appl Toxicol 7:237– 244. Dionex. 1998. Analysis of low concentrations of perchlorate in drinking water and groundwater by ion chromatography. Application Note 121. Sunnyvale, CA, USA. Nieuwkoop PD, Faber J, eds. 1994. Normal Table of Xenopus laevis (Daudin): A Systematical and Chronological Survey of the Development from the Fertilized Egg till the End of Metamorphosis, 3rd ed. Garland, New York, NY, USA. Fort D, Stover E. 1997. Development of short-term, whole-embryo assays to evaluate detrimental effects on amphibian limb development and metamorphosis using Xenopus laevis. In Dwyer FJ, Doane TR, Hinman ML, eds, Environmental Toxicology and Risk Assessment: Modeling and Risk Assessment, Vol 6. STP 1317. American Society for Testing and Materials, Philadelphia, PA, pp 376–390. U.S. Army Corps of Engineers. 1999. Results of a chemical test of water samples, groundwater treatment plant, Longhorn Army Ammunition Plant—Tulsa District. SWF-EV-D Report 16967-1. Fort Worth, TX. Bantle JA, Dumont JN, Harvey JG, Mattie DR. 1999. FETAX assay of ammonium perchlorate. Toxicologist 48:528. Dumont JN, Schultz TW, Buchanan MV, Kao GL. 1983. Frog embryo teratogenesis assay: Xenopus (FETAX)—A short assay applicable to complex environmental mixtures. In Waters MD, Sandhu SS, Lewtas J, Claxton L, Chernoff N, Nesnow S, eds, Symposium on the Application of Short-Term Bioassays in the Analysis of Complex Environmental Mixtures: III. Plenum, New York, NY, USA, pp 393–405. Schuytema GS, Nebeker AV. 1999. Comparative effects of ammonium and nitrate compounds on Pacific treefrog and African clawed frog embryos. Arch Environ Contam Toxicol 36:200– 206. Shi YB. 2000. Amphibian Metamorphosis: From Morphology to Molecular Biology. Wiley-Liss, New York, NY, USA. Leloup J, Buscaglia, M. 1977. La triiodothyronine: Hormone de la me´tamorphose des amphibiens. CR Acad Sci Ser III Sci Vie 284:2261–2263. Lawrence JE, Lamm SH, Pino S, Richman K, Braverman LE. 2000. The effect of short-term low-dose perchlorate on various aspects of thyroid function. Thyroid 10:659–663. Eskandari S, Loo DD, Dai G, Levy O, Wright EM, Carrasco N. 1997. Thyroid Na1/I-symporter: Mechanism, stoichiometry, and specificity. J Biol Chem 272:27230–27238. Carrasco N. 1993. Iodide transport in the thyroid gland. Biochim Biophys Acta 1154:65–82. Robertson JC, Kelley DB. 1996. Thyroid hormone controls the onset of androgen sensitivity in the developing larynx of Xenopus laevis. Dev Biol 176:108–123.