ENVIRONMENTAL TEMPERATURE CHANGES UPTAKE RATE AND BIOCONCENTRATION FACTORS OF BISPHENOL A IN TADPOLES OF RANA TEMPORARIA

Environmental Toxicology and Chemistry, Vol. 25, No. 10, pp. 2804–2808, 2006 䉷 2006 SETAC Printed in the USA 0730-7268/06 $12.00 ⫹ .00

ENVIRONMENTAL TEMPERATURE CHANGES UPTAKE RATE AND BIOCONCENTRATION FACTORS OF BISPHENOL A IN TADPOLES OF RANA TEMPORARIA JANI O. HONKANEN* and JUSSI V.K. KUKKONEN Laboratory of Aquatic Ecology and Ecotoxicology, Department of Biology, University of Joensuu, P.O. Box 111, FI-80101 Joensuu, Finland ( Received 16 January 2006; Accepted 4 May 2006) Abstract—Toxicokinetics of radiolabeled (14C) bisphenol A was studied in the common frog (Rana temporaria) at two experimental temperatures (7 and 19⬚C). The growth rate of the tadpoles during the 96-h experiment was very slow at 7⬚C, but the weight of tadpoles almost tripled at 19⬚C. At all tested exposure concentrations (0.2, 1.5, 10, and 100 ␮g/L), conditional uptake rate constants (ku) were 69 to 82%, and elimination rates (ke) 79 to 90% lower, at 7⬚C than at 19⬚C. On the contrary, bioconcentration factors (BCFs) were higher at 7⬚C than at 19⬚C. Total accumulated bisphenol A per individual was higher at 19⬚C, which is in agreement with higher ku at 19⬚C. Exposure concentrations did not have any constant effect on BCFs at the two temperatures. The results of the current experiment suggest that higher temperature increases uptake and total amount of chemical in frog tadpoles but does not necessarily lead to higher BCFs. High temperature may have increased the growth rate more than the uptake rate, resulting in a net dilution of bisphenol A in tadpole tissues. The observed difference in BCFs also could be a result of temperature-induced changes in allometric relationships (increased surface area to volume ratio) and/or more effective elimination in more developed tadpoles at high temperature. Keywords—Bioconcentration factor

Rana temporaria

Bisphenol A

Temperature

Growth rate

model compound, because it is moderately water soluble (120– 300 mg/L) and accumulates relatively well in lipid-rich animal tissues (log Kow ⫽ 2.20–3.82) [6]. In addition, it is used for many purposes, such as the manufacture of plastics and as a fungicide, antioxidant, flame retardant, and rubber chemical [7], which creates a potential for its dispersion in the environment.

INTRODUCTION

Body size is an important factor affecting toxicokinetics of chemicals in aquatic animals [1]. In aquatic poikilothermic animals, such as fish and frogs, environmental temperature may change activity and growth rate [2], which are important factors affecting the accumulation of waterborne chemicals. In addition, temperature-induced changes in permeability of biological membranes and activity of enzymes could affect both accumulation and elimination processes of chemicals [3]. Temperature-induced changes in growth rate may lead to further changes in allometric relationships (e.g., ratios of surface area to volume), which could modify accumulation and elimination of chemicals. Over short time periods (hours to days), body size usually is relatively stable in adult anurans, and a change in environmental temperature should not affect their growth [4]. However, growth rate during early life stages could be very fast at optimal temperature regimes, and a change in environmental temperature could modify growth rate even over short periods [5]. Fast growth during early life stages of anurans has several implications for toxicokinetic processes, because allometric relationships of developing frog tadpoles change when they increase their size. Currently, relationships among temperature, growth, and accumulation of organic chemicals in aquatic poikilotherms are not well-described in the scientific literature. The objective of the present paper is to examine the temperature dependence of toxicokinetic processes of bisphenol A in tadpoles of the common frog (Rana temporaria). The results presented are based on experiments in which newly hatched tadpoles were exposed to several bisphenol A concentrations at two temperatures. Bisphenol A was used as a

MATERIALS AND METHODS

Animals The frog egg masses were collected from temporary pools near the Joensuu City Center (Hasanniemi, Joensuu, Finland) during two consecutive springs, when the short-term experiments were performed. In both years, three batches of newly laid eggs were added to 10-L containers of water and acclimated to experimental temperatures for 5 d. Newly hatched tadpoles (1–24 h from hatching) were used in the experiments. The Animal Care and Use Committee of the University of Joensuu approved all procedures. The permission to collect frog spawn was given by North Karelia Regional Environment Center (Joensuu, Finland).

Water and chemicals The water used in experiments was Joensuu City tap water (pH 7–7.5; total hardness, 0.4–0.6 mmol/L; total organic carbon, ⬍0.3 mg/L), which is groundwater and is treated only with limestone filters before distribution. The exposure waters were aerated for 24 to 48 h and, thereafter, spiked with 14Clabeled bisphenol A ([propyl-2-14C]; specific activity, 2,074 MBq/mmol; radiochemical purity, 99.9%; Moravek Biochemicals, Brea, CA, USA) and/or with nonlabeled bisphenol A (purity, ⬎99%; Sigma-Aldrich, Gillingham, UK). The lowest concentration (0.2 ␮g/L) was spiked purely with 14C-labeled

* To whom correspondence may be addressed ([email protected]). 2804

Temperature changes BCFs in frog tadpoles

Environ. Toxicol. Chem. 25, 2006

2805

bisphenol A and the other concentrations with both labeled and nonlabeled bisphenol A in the following mixing ratios: Concentration in exposure water, 1.5 ␮g/L (33% of 14C-labeled bisphenol A from total amount of bisphenol A), 10.5 ␮g/L (4.8%), and 100.5 ␮g/L (0.50%). After spiking, the exposure waters were mixed continuously for 5 min in 5-L glass flasks, and then 250 ml of the spiked water were added to each 250ml beaker.

Experimental setup Semistatic exposures were made at two temperatures, 7 ⫾ 1 and 19 ⫾ 1⬚C. The first experiment was performed at 19⬚C in the year 1999, and the second was performed at 7⬚C in the year 2000. The exposure at 7⬚C was made in a cold chamber and the exposure at 19⬚C in a laboratory at room temperature. The light cycle at each exposure temperature was 14:10-h light: dark. Tadpoles were not fed during the experiment, because newly hatched tadpoles (hatchlings) use their internal yolk reserves as a nutritional source until active feeding begins. At the beginning of the experiments, five frog tadpoles (sampling times at 6 and 12 h) or four tadpoles (all other sampling times) were added to each beaker. Each concentration at each sampling time included three replicate beakers. Sampling intervals were 6, 12, 24, 48, and 96 h. When samples were taken, the tadpoles in each beaker were weighed separately and pooled for radiolabel analysis. Weights were determined on a wet-weight basis, and before measuring the weight, each tadpole was rinsed in clean water to remove nonaccumulated bisphenol A from the surface. Tadpoles were dried gently on the rim of the beaker. Weighed and pooled tadpoles were placed into glass scintillation vials and dissolved in 0.5 ml of tissue solubilizer (Soluene威 350; Packard, Groningen, The Netherlands) at 50⬚C for 24 h, after which 12 ml of scintillation cocktail (Ultima Gold; Packard) were added. Dissolved samples were kept at room temperature for one week before analysis using a liquid scintillation counter (WinSpectral 1414; Wallac, Turku, Finland). At each sampling time, 6 ml of exposure water were removed from each beaker and placed into a plastic scintillation counter vial, and 6 ml of scintillation cocktail (Insta-Gel Plus; Packard) were added. Radioactivities in the water samples were analyzed in the same way as the animal samples. The measured counts per minute were converted to mass equivalents using the specific activities of compounds. Therefore, all values for bisphenol A are reported as mass equivalents. The final bisphenol A concentrations in animals and waters were calculated from the mixing ratios of labeled and nonlabeled bisphenol A.

Data analysis and kinetic models Data analyses were performed with the SigmaPlot 8.0 (Systat Software, Richmond, CA, USA) and SPSS威 11.0 (SPSS, Chicago, IL, USA) statistical programs. Wet-weight data of tadpoles were pooled at both temperatures, because the increase of wet weights did not vary among the bisphenol A treatments (analysis of variance, p ⫽ 0.0001). Wet weights of tadpoles at both temperatures were fitted to a linear-growth model to calculate the growth rate. Accumulation data were fitted to two first-order kinetic models (Eqns. 1 and 2), as described by Landrum et al. [8]. The suitability of the models was tested on the entire data set to describe and test the importance of growth in data fitting.

Fig. 1. Increase of tadpole (Rana temporaria) wet weights in 96-h experiments at 7⬚C (solid circles) and 19⬚C (open circles). Growth rate did not differ among various bisphenol A treatments; therefore, different exposures were pooled to two groups according to their exposure temperatures. Each dot represents the average wet weight of four to five tadpoles in replicate beakers at different treatments. Significant increase of wet weights was observed at 19⬚C (r2 ⫽ 0.954, p ⫽ 0.000) but not at 7⬚C (r2 ⫽ 0.004, p ⫽ 0.264).

Ca ⫽

ku Cw (1 ⫺ e⫺ke t ) ke

(1)

Ca ⫽

ku Cw (1 ⫺ e⫺(ke⫹g)t ) ke ⫹ g

(2)

where Ca is the concentration of bisphenol A in the animal (␮g/g), ku is the conditional uptake clearance coefficient (ml/ g/h), ke is the conditional elimination rate constant (1/h), Cw is the concentration of bisphenol A in water (␮g/ml), and t is the time (h). In Equation 2, g is the growth rate (g/g/h). Bioconcentration factors (BCFs) were calculated in three ways to verify possible differences in estimation methods: BCFku/ke ⫽ ku/ke, in which the uptake rate (ku) was divided by the elimination rate (ke); BCF96h ⫽ Ca(96h)/Cw, in which the bisphenol A concentration in the animal at the end of the experiment (Ca(96h)) was divided by the average water concentration of bisphenol A during the experiment (Cw); and BCFss ⫽ Ca(ss)/Cw, in which the bisphenol A concentration in the animal at steady state (Ca(ss)) was divided by the average water concentration of bisphenol A during the experiment (Cw). Bisphenol A concentrations at various sampling times were considered to be at steady state when three or more consecutive sampling times did not differ statistically from each other (analysis of variance). RESULTS

Temperature had a clear effect on tadpole growth at 19⬚C, and the average fresh weight tripled during the 96-h experiment (Fig. 1). However, the growth rate of the tadpoles was insignificant at 7⬚C, and tadpoles did not increase their mass during the 96-h experiment. Uptake and elimination rates of bisphenol A clearly were lower at 7⬚C than at 19⬚C (Table 1). The graphical presentation of measured accumulation at two temperatures shows that uptake rate was faster, and that steady state was achieved sooner, at 19⬚C (Fig. 2). However, steady state was not reached at any exposure concentration in the 96-h exposure time at 7⬚C. Total

2806

Environ. Toxicol. Chem. 25, 2006

J.O. Honkanen and J.V.K. Kukkonen

Table 1. Uptake (ku) and elimation (ke) rate constants of bisphenol A in frog (Rana temporaria) tadpolesa Growth rate excluded in the kinetic model (Eqn. 1)

ku Temperature (⬚C) 19

7

a

Growth rate included in the kinetic model (Eqn. 2)

kc

ku

ke

Exposure concn. (␮g/L)

L/kg/h

⫾SE

1/h

⫾SE

r2

L/kg/h

⫾SE

1/h

⫾SE

r2

0.2 1.5 10 100 0.2 1.5 10 100

20.90 21.34 16.53 32.82 4.69 4.86 5.19 5.92

4.38 2.90 2.79 17.80 0.554 0.708 0.355 0.328

0.17 0.21 0.15 0.46 0.026 0.025 0.032 0.328

0.039 0.031 0.028 0.26 0.005 0.006 0.003 0.004

0.75 0.89 0.81 0.74 0.92 0.88 0.97 0.96

29.5 24.1 24.7 33.0 — — — —

8.88 3.69 6.44 17.39 — — — —

0 0 0 0.23 — — — —

0 0 0 0.25 — — — —

0.70 0.89 0.74 0.74 — — — —

The table shows estimated rates for four different exposure concentrations at two experimental temperatures. Growth rate excluded refers to Equation 1, and growth rate included refers to Equation 2. SE ⫽ standard error of the mean.

amounts of bisphenol A in tadpole tissues (calculated on an individual basis) clearly were higher at 19⬚C than at 7⬚C (Fig. 2). Bisphenol A concentrations (14C equivalents) in exposure

waters remained almost constant at both experimental temperatures throughout the experiments. Concentrations decreased by up to 4% from the original concentrations at various sampling times at 7⬚C and by 10.2% at 19⬚C.

Fig. 2. Weight-normalized (left) and individual (right) accumulation of bisphenol A (14C equivalents) in frog (Rana temporaria) tadpoles exposed to four different concentrations of chemical at 7⬚C (solid circles) and 19⬚C (open circles). Each dot in the weight-normalized data represents the average bisphenol A concentration of four to five tadpoles in a beaker (three replicate beakers per sampling time). Individual data show the average amount of bisphenol A in tadpoles from three parallel exposure beakers.

Temperature changes BCFs in frog tadpoles

Environ. Toxicol. Chem. 25, 2006

2807

Estimated accumulation rates were 69 to 82% slower at 7⬚C than at 19⬚C. However, elimination rates were more affected by low temperature, because the calculated elimination rates were 79 to 90% slower at 7⬚C than at 19⬚C. However, toxicokinetic constants (uptake and elimination rates) at 19⬚C varied depending on the models used. When the growth rate (g) was excluded (Eqn. 1) from the toxicokinetic model, the uptake rates were, at all exposure concentrations, lower than the estimates where growth was included (Eqn. 2) in the model (Table 1). The elimination rates estimated when the growth rate was not included in the model varied from 0.15 to 0.46/ h, but when the growth rate was included in the toxicokinetic model (Eqn. 2), all elimination rates except the highest exposure concentration decreased to zero (Table 1). The BCFs were higher at 7⬚C than at 19⬚C (Fig. 3). Exposure concentration did not have any clear effect on BCFs at 7 or 19⬚C. The BCFs calculated from the kinetic estimates and the tissue concentrations gave similar results (Fig. 3). However, when tissue concentrations at steady state (only at 19⬚C) were used to calculate BCFs, the values were similar at all exposure concentrations; only the BCFs at the highest concentration remained lower than at other concentrations. DISCUSSION

The results of the current experiment demonstrated that the higher incubation temperature increased the conditional uptake rate constants in frog tadpoles but did not necessarily lead to higher BCFs. The observed four- to sixfold higher uptake rates at 19⬚C in comparison to 7⬚C can be explained by increased activity and oxygen consumption of tadpoles [9], which increase the uptake rate of chemicals [10]. Another reason could be growth dilution and/or more effective elimination rate at the higher temperature. The effect of growth dilution can be minimized, as described by Landrum et al. [8], by adding a growth rate constant to the first-order kinetic model (Eqn. 1). If the growth rate was not used in the toxicokinetic models, elimination rates were overestimated (Table 1), suggesting that growth dilution had a significant role in decreasing bisphenol A concentration in tadpole tissues at 19⬚C. However, our results show that very fast growth rates (factor g in Eqn. 2) of frog tadpoles decrease estimated elimination rates to zero or close to it (Table 1). This was a relatively good estimate of net elimination of bisphenol A from tadpole tissues to water in steady state but did not represent real elimination capability of tadpoles, because their capacity to eliminate chemicals should not disappear because of fast growth. Fast growth also leads to allometric changes in tadpoles, such as decreased ratios of surface area to volume (i.e., larger animals have less body surface per total body volume), which would decrease accumulation per mass unit and, therefore, lead to lower BCFs at 19⬚C. From a developmental point of view, tadpoles at 19⬚C were larger in size than tadpoles at 7⬚C, which means that their organs (e.g., liver) were further developed and, possibly, more efficient at eliminating bisphenol A. In addition, net elimination of bisphenol A may be more efficient at high temperatures, especially within the optimal temperature regime, as has been shown in freshwater clam [11], but not above the optimal temperature regime in developing salmon embryos [12]. In the present study, higher exposure temperature was close to optimal temperature for development of frog tadpoles and could have caused more efficient elimination and lower BCFs at 19⬚C. Temperature dependence of BCFs was clear,

Fig. 3. Bioconcentration factors (BCFs) of bisphenol A in frog (Rana temporaria) tadpoles at 7⬚C (solid bars) and 19⬚C (open bars). Three methods were used to calculate the BCFs: Kinetic constants (top), chemical concentrations at the end of experiment (middle), and steady-state concentrations (bottom). Numbers inside the bars refer to times when the tissue concentrations were considered to be statistically ( p ⬍ 0.05) in steady state. Vertical lines represent the standard deviations of means.

2808

Environ. Toxicol. Chem. 25, 2006

because the values were 24 to 48% higher at low than at high temperature (Fig. 3). Changes in BCFs of bisphenol A with temperature also have been reported in salmon embryos [12] and freshwater clam [11]. In salmon embryos, BCFs increased with increasing temperature, but in freshwater clams, no real temperature relationship was observed. The relationship between temperature and BCFs seems to be more complicated than usually is expected. This could mean that in poikilothermic animals, the relationship between temperature and dependent factors, such as BCFs, is not necessarily linear, especially outside the optimal temperature regime. Interestingly, exposure concentration did not have any clear relationship to BCFs, as has been observed for bisphenol A– exposed salmon yolk-sac fry [13]. However, in the present study, the lowest BCFs were observed at the highest bisphenol A concentration (100 ␮g/L) at both temperatures. This could mean that BCFs do not follow any consistent relationship with exposure concentrations below the limits of acute toxicity, perhaps because of the nonexistent effect on animal activity or other factors affecting accumulation. Acute toxicity of bisphenol A in water exposures commonly is observed above 1,000 ␮g/L [6,14], and it is suggested that BCFs could follow a different pattern at exposure concentrations close to acute toxicity. Weight-normalized accumulation data clearly suggest that the bisphenol A concentration reaches steady state in tadpole tissues at 19⬚C, which means that bisphenol A is accumulated at the same rate as it is eliminated from the tissues. However, the total amount of bisphenol A in individual tadpoles increased throughout the experiment at 19⬚C (Fig. 2). The toxicological importance of continuous increase in total amount is not known. A recent study in our laboratory has shown that dry weight of tadpoles does not increase during prefeeding stages (P. Koponen, University of Joensuu, Department of Biology, Joensuu, Finland, personal communication). In the present study, the dry weights of the tadpoles were not measured, but constant increase of bisphenol A in individual tadpoles at 19⬚C indicates that steady state was not reached on a dryweight basis. Because chemical analyses in this study were based on measuring the amount of radiolabeled bisphenol A in tadpole tissues (14C equivalents), we cannot give any exact numbers for tissue levels of parent compound. Because bisphenol A contains two hydroxyl groups, it is readily available for conjugation by phase II enzymes, mainly by glucuronide conjugation [15]. This suggests that bisphenol A is present in tadpole tissues mainly as parent compound and as glucuronide conjugates. However, we cannot exclude the presence of other forms of metabolites. The observation of steady state indicates elimination of radiolabeled compound from tadpole tissues— at least as long as the bisphenol A concentration in exposure waters remains constant. Higher BCFs at lower temperatures, as the present study shows, could mean that tadpoles developing at cold temperatures accumulate more chemicals from water than do those at warmer temperatures. However, it is not known how accumulation of chemicals is affected in long-term exposures. Earlier studies have shown that temperature may modify the toxicity of chemicals during early development of frogs [16,17], but it is not known whether differences in toxic responses result from changed tissue concentrations or sensitivities (or even both). In future studies, the toxicological significance of changes in BCFs with temperature and possible

J.O. Honkanen and J.V.K. Kukkonen

changes in threshold concentrations of toxic responses should be verified. In addition, when toxicokinetics of chemicals in developing animals at various temperatures are compared, it would be more informative to follow accumulation until a certain developmental stage rather than using the same exposure times at various temperatures. This would mean that exposure times were longer at low temperatures and that achieved data were better for comparisons between different developmental stages. Acknowledgement—The authors want to thank Marja Noponen, Katja Tauriainen, and Peter V. Hodson. This study was funded by the Finnish Graduate School in Environmental Science and Technology and the Academy of Finland (projects 206071 and 208430). REFERENCES 1. Newman MC. 1994. Quantitative Methods in Aquatic Ecotoxicology. CRC, Boca Raton, FL, USA, pp 59–118. 2. Rombough PJ. 1996. The effects of temperature on embryonic and larval development. In Wood CM, McDonald DG, eds, Global Warming: Implications for Freshwater and Marine Fish. Society for Experimental Biology Series 61. Cambridge University Press, Cambridge, UK, pp 177–223. 3. Kennedy CJ, Walsh PJ. 1996. Effects of temperature on xenobiotic metabolism. In Wood CM, McDonald DG, eds, Global Warming: Implications for Freshwater and Marine Fish, Society for Experimental Biology Series 61. Cambridge University Press, Cambridge, UK, pp 303–324. 4. Duellman WE, Trueb L. 1986. Biology of Amphibians. McGrawHill, New York, NY, USA. 5. Ultsch GR, Bradford DF, Freda J. 1999. Physiology—Coping with the environment. In McDiarmid RW, Altig R, eds, Tadpoles— The Biology of Anuran Larvae. The University of Chicago Press, Chicago, IL, USA, pp 189–214. 6. Staples CA, Dorn PB, Klecka GM, O’Block ST, Harris LR. 1998. A review of the environmental fate, effects, and exposures of bisphenol A. Chemosphere 36:2149–2173. 7. Takahashi O, Oishi S. 2001. Testicular toxicity of dietary 2,2bis(4-hydroxyphenyl)propane (bisphenol A) in F344 rats. Arch Toxicol 75:42–51. 8. Landrum PF, Lee HI, Lydy MJ. 1992. Toxicokinetics in aquatic systems: Model comparisons and use in hazard assessment. Environ Toxicol Chem 11:1709–1725. 9. Feder ME. 1982. Effect of developmental stage and body size on oxygen consumption of Anuran larvae: Reappraisal. J Exp Zool 220:33–42. 10. Yang R, Brauner C, Thurston V, Neuman J, Randall DJ. 2000. Relationship between toxicant transfer kinetic processes and fish oxygen consumption. Aquat Toxicol 48:95–108. 11. Heinonen J, Honkanen J, Kukkonen JVK, Holopainen IJ. 2002. Bisphenol A accumulation in the freshwater clam Pisidium amnicum at low temperatures. Arch Environ Contam Toxicol 43: 50–55. 12. Honkanen JO, Heinonen J, Kukkonen JVK. 2001. Toxicokinetics of waterborne bisphenol A in landlocked salmon (Salmo salar m. sebago) eggs at various temperatures. Environ Toxicol Chem 20:2296–2302. 13. Honkanen JO, Holopainen IJ, Kukkonen JVK. 2003. Bisphenol A induces yolk-sac edema and other adverse effects in landlocked salmon (Salmo salar m. sebago) yolk-sac fry. Chemosphere 55: 187–196. 14. Alexander HC, Dill DC, Smith LW, Guiney PD, Dorn P. 1988. Bisphenol A: Acute aquatic toxicity. Environ Toxicol Chem 7: 19–26. 15. Yokota H, Miyashita N, Yuasa A. 2002. High glucuronidation activity of environmental estrogens in the carp (Cyprinus carpino) intestine. Life Sci 71:887–898. 16. Boone MD, Bridges CM. 1999. The effect of temperature on the potency of carbaryl for survival of tadpoles of the green frog (Rana clamitans). Environ Toxicol Chem 18:1482–1484. 17. Broomhall SD. 2004. Egg temperature modifies predator avoidance and the effects of the insecticide endosulfan on tadpoles of an Australian frog. J Appl Ecol 41:105–113.