当前位置:首页 >> 环境科学/食品科学 >>

Chlorpyrifos exposure of developing zebrafish- effects on survival and long-term effects


Neurotoxicology and Teratology 25 (2003) 51 – 57 www.elsevier.com/locate/neutera

Chlorpyrifos exposure of developing zebrafish: effects on survival and long-term effects on response latency and spatial discrimination
Edward D. Levina,*, Elizabeth Chrysanthisa, Kari Yacisinb, Elwood Linneyb
Neurobehavioral Research Laboratory, Department of Psychiatry and Behavioral Sciences, Box #3412, Duke University Medical Center, Durham, NC 27710, USA b Department of Molecular Genetics and Microbiology, Duke University Medical Center, Durham, NC, USA Received 5 July 2002; received in revised form 13 August 2002; accepted 23 September 2002
a

Abstract Chlorpyrifos (CPF) is a widely used insecticide, which has been shown to interfere with neurobehavioral development. Rat models have been key in demonstrating that prenatal CPF exposure causes choice accuracy deficits and motor alterations, which persist into adulthood. Complementary nonmammalian models can be useful in determining the molecular mechanisms underlying the persisting behavioral effects of developmental CPF exposure. Zebrafish with their clear chorion and extensive developmental information base provide an excellent model for assessment of molecular processes of toxicant impacted neurodevelopment. To facilitate the use of the zebrafish model and to compare it to the more typical rodent models, the behavioral phenotype of CPF toxicity in zebrafish must be well characterized. Our laboratory has developed methods for assessing spatial discrimination learning in zebrafish, which can differentiate response latency from choice accuracy in a three chambered fish tank. Low and high doses of CPF (10 and 100 ng/ml on days 1 – 5 postfertilization) both had significant persisting effects on both spatial discrimination and response latency over 18 weeks of testing. The high, but not the low dose, significantly accelerated mortality rates of the fish during the study from 20 – 38 weeks of age. Developmental exposure to either 10 or 100 ng/ml of CPF caused significant spatial discrimination impairments in zebrafish when they were adults. The impairment caused by 10 ng/ml was seen during early but not later testing, while the impairment caused by 100 ng/ml became more pronounced with continued testing. The higher dose caused a more pervasive impairment. The 10 and 100 ng/ml doses had opposite effects on response latency. The low 10 ng/ml dose significantly slowed response latency, while the high 100 ng/ml dose significant increased response latency. Both of these effects diminished with continued testing. CPF exposure during early development caused clear behavioral impairments, which lasted throughout adulthood in zebrafish. The molecular mechanisms by which early developmental CPF exposure produces these behavioral impairments expressed in adulthood can now be studied in the zebrafish model. D 2002 Elsevier Science Inc. All rights reserved.
Keywords: Zebrafish; Danio rerio; Behavior; Learning; Chlorpyrifos

1. Introduction Chlorpyrifos (CPF) is one of the most widely used insecticides in the world. CPF exposure during development causes persisting neurobehavioral effects in rats [2,12,13,15,19], leading to CNS neural cell loss and abnormalities of synaptic function [5,7,19] and behavior [12,13].

* Corresponding author. Tel.: +1-919-681-6273; fax: +1-919-6813416. E-mail address: edlevin@duke.edu (E.D. Levin).

CPF is a putative neuroteratogen in humans [11]. Pre- or postnatal CPF administration in rats causes behavioral disruption of motor activity and learning and memory function [4,6,12,13], which persist well after the end of CPF administration. Prenatal CPF exposure in rats has been found in our laboratory to cause a learning impairment that lasts throughout adulthood. Determining the sequence of molecular mechanisms of the CPF-induced behavioral impairments in mammals is a challenge because of the limited accessibility of mammals during prenatal development. Complementary nonmammalian models such as fish can provide continuous visual access to the embryo

0892-0362/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. doi:10.1016/S0892-0362(02)00322-7

52

E.D. Levin et al. / Neurotoxicology and Teratology 25 (2003) 51–57

throughout development and subsequently have an intact organism for behavioral study. Determining the behavioral phenotype of toxicant exposure in these new models will open new avenues for determining the molecular mechanisms of neurobehavioral teratology. Zebrafish (Danio rerio) are becoming a model of choice for studying the molecular bases of neurodevelopment [8]. The clear chorion (egg sack) of zebrafish allows continual visualization of the developmental process. Rapid development and accessibility to genetic analysis make the zebrafish an excellent model system for studies of neurodevelopment. The wide variety of genetic mutants available in zebrafish offers the promise of determining the molecular mechanisms of neurobehavioral function. Use of zebrafish for neurodevelopment studies has led to the identification of a variety of genes affecting various aspects of neural development and function [18]. For example, these types of studies have shown the importance of acetylcholinesterase for neurodevelopment [3]. To understand the meaning of molecular events during neurodevelopment, it is essential to have reproducible and validated methods for determining behavioral function. Learning in zebrafish is beginning to be studied. Our laboratory and other’s have developed methods to characterize discrimination learning. Williams and Messer [21] have reported that zebrafish would learn to swim to one side of the tank in response to a tapping on that side that signaled the delivery of food. All of the fish learned this response within three weeks of training. They tested fish in groups of five to six making it difficult to determine learning of individual fish. Suboski [20] found that learning in fish often involves stimulus substitution in which a releasing stimulus becomes linked with a previously neutral stimulus in controlling the release of behavior. Recognition learning in fish involves primarily the extension to new stimuli of control over the releasing of preorganized innate responses. Hall and Suboski [9,10] showed in zebrafish that visual or olfactory stimuli can elicit escape responses in a learning paradigm. Some studies have begun to identify molecular factors important for memory in zebrafish [16,17]. Our laboratory has developed methods, which demonstrate spatial and nonspatial escape and avoidance discrimination learning [1]. These tests use a three-chambered tank in which the choice behavior and response latency of zebrafish can be assessed in the context of spatial and nonspatial discrimination tests. The advantage of this procedure over previous behavioral tasks is that individual zebrafish can be repeatedly tested. Zebrafish reliably learn and remember the response contingencies in these tests. In the current study, a three-chamber spatial discrimination test was used to monitor the effect of early developmental CPF exposure on later choice accuracy performance in zebrafish. The development of learning and memory tests for zebrafish is essential for determining the molecular mechanisms of cognitive function.

2. Methods 2.1. Subjects and CPF exposure Zebrafish (D. rerio) embryos were exposed to CPF starting when embryos were 2 –16 cells and continuing for days 1 –5 postfertilization. Embryos were collected from adult AB-star zebrafish and rinsed four times in fresh egg water. Healthy looking and developing embryos between the 2- and 16-cell stages were placed in 25 ml of 10 or 100 ng/ml CPF (Supelco, PS-674) or in water with addition of the vehicle alone, which was 0.2 ml/ml dimethyl sulfoxide (Mallinckrodt). Embryos were kept in an incubator at approximately 28.5 °C. Water was replaced daily until day 5 when embryos were rinsed in fresh water and placed in 150 ml of egg water (60 mg/ml Instant Ocean). Embryos were fed Tetrahymena (Carolina Biological) from days 5 until 17 and day-old brine shrimp (INVE) from days 8 to 17. Larvae were graduated into 2-l tanks on day 17 where they were fed brine shrimp twice a day and flake food (TetraMin) once a day. Fish were kept at approximately 28.5 °C on a 14:10-h light – dark cycle until about 9 weeks of age when they were transferred to the behavioral testing laboratory. The fish were kept in tanks in the behavioral testing laboratory until they began behavioral testing at 20 weeks of age. At the beginning of the behavioral study, there were n’s of 13, 16 and 12 for the CPF 0, CPF 10 and CPF 100 groups, respectively. These group sizes were not significantly different by chi-square analysis. The adult fish were maintained in 9-l tanks measuring 15 cm wide ? 30 cm long ? 20 cm high. Each tank was on a 12-h light – dark cycle, and all behavioral testing took place during the light phase between 8:00 a.m. and 5:00 p.m. Half the volume of water was changed weekly utilizing de-ionized H2O and sea salts (Instant Ocean, 1.2 g/20 l of water). The tanks with the adult fish were maintained at approximately 28.5 °C with constant filtration and aeration. Fish were fed daily after testing with TetraMin flakes. 2.2. Fish identification labeling Before the onset of behavioral testing the fish were individually labeled with a fluorescent dye marker. The zebrafish were individually anesthetized with MS-222 (tricaine methane sulfonate, 40 mg/473 ml H2O). Visible Fluorescent Elastomer (Northwest Marine Technology, Shaw Island, WA) was injected subcutaneously as a permanent marker. The labeled zebrafish were then placed in their original tanks and allowed to recover for 1 week. 2.3. Behavioral testing A Plexiglas maze was fitted into a 29-l fish tank (22 cm wide ? 44 cm long ? 30 cm high) as illustrated in Fig. 1. The test tank was divided into three equal-sized compartments, one central and two side compartments. There were

E.D. Levin et al. / Neurotoxicology and Teratology 25 (2003) 51–57

53

Fig. 1. The three-compartment zebrafish maze was made of clear Plexiglas inserted in a 29-l fish tank. The middle chamber measured 22 ? 18 cm. There were vertically sliding doors on either side of this central start area leading to two 13 ? 22 cm choice areas at either end of the tank.

vertically sliding doors, which were 12 cm high and 10 cm wide on either side of this central start area leading to two 13 ? 22 cm choice areas. The partitions were mounted on rails so that it could be moved to 2 cm from the end wall of the tank to restrict the movement of the fish as the aversive contingency. A dark panel covered one long side of the tank, covering one side of each of the three compartments to provide an axis of orientation for right – left discrimination. To start a trial, each fish was placed in the central compartment. After 60 s, the doors to each of the side compartments were simultaneously opened. If the fish swam into the correct side compartment, the door was closed and the fish was left alone in that compartment for 30 s. If the fish swam into the incorrect side compartment, the door was closed and the partition to that side was moved to be 2 cm from the end wall for 10 s. If the fish did not make a choice within 20 s, a fishnet was moved across the central compartment parallel to the doors. Half of the fish were trained right and half were trained left. Choice accuracy and response latency were recorded. Performance was assessed with six-session blocks averaged for analysis. Trial and response times were monitored by an electronic stopwatch. 2.4. Statistical analysis The data were assessed for significance by analysis of variance (ANOVA) for between subjects factors and withinsubjects repeated measures. Between subjects factors were CPF dose and correct location for training. The repeated factor was six-session block of training. Significant interactions were followed up by tests of the simple main effects. Dunnett’s tests (two-tailed) were used to compare treated groups vs. controls. Chi-square analysis was used to exam-

ine CPF effects on survival. P-values less than .05 were considered to be significant.

3. Results Survival was significantly affected by the higher dose of CPF (Table 1). The early developmental CPF exposure, which only lasted through day 5 after fertilization did not cause immediate effects on survival, but did impair later long-term survival. The survival of the fish in each group over the next 18 weeks was analyzed. At 26 weeks of age, there were 13, 15 and 7 fish alive. Chi-square analysis showed this to be a significant CPF treatment effect on survival ( P < .01). In particular, there was a significant difference between controls and the CPF 100 group (chisquare: P < .01), but not controls and the CPF 10 group. The significant CPF 100 treatment effect on survival persisted. On week 32, there was still a significantly (chi-square: P < .05) greater survival rate in controls (n = 12) than in the CPF 100 group (n = 7). The CPF 10 group (n = 15) still did not differ from controls. Finally, after the end of the
Table 1 Number of zebrafish surviving at different stages during the behavioral study Exposure Age (weeks) 20 CPF 0 CPF 10 CPF 100 13 16 12 26 13 15 7* * 32 12 15 7* 38 9 12 6

* Chi-square test: P < .05 vs. CPF 0. ** Chi-square test: P < .01 vs. CPF 0.

54

E.D. Levin et al. / Neurotoxicology and Teratology 25 (2003) 51–57

Fig. 2. Zebrafish spatial discrimination, escape/avoidance choice accuracy (% correct) average over 18 sessions (mean ± S.E.M.).

behavioral study at 38 weeks of age, some of the control fish began to die so that there was no longer a significant CPF treatment effect on survival as determined by chi-square analysis.

Choice accuracy (% correct) was significantly affected by developmental CPF treatment. The main effect of CPF treatment was significant [ F(2,27) = 6.49, P < .005]. Dunnett’s tests showed that both the low 10 ng/ml CPF dose ( P < .05) and the high 100 ng/ml dose ( P < .01) caused significantly lower percent correct scores averaged over the 18-session test sequence (Fig. 2). The Treatment ? Session Block interaction was also significant [ F(4,54) = 3.03, P < .05]. As shown in Fig. 3, Dunnett’s tests within the simple main effects of CPF at each session block showed that during the initial session block the 10 ng/ml dose caused a significant impairment relative to controls ( P < .05). A significant impairment in the 100 ng/ml CPF group emerged during the middle session block ( P < .05) and became more pronounced during the final session block ( P < .01). There was also a significant CPF Treatment ? Training Side interaction [ F(2,27) = 3.41, P < .05] with a more pronounced CPF effect with training to the left side than to the right. Developmental CPF exposure caused substantial effects on response latency. The main effect of CPF treatment was quite significant [ F(2,27) = 26.52, P < .0001]. The CPF effect was biphasic with the Dunnett’s tests showing that lower dose of 10 ng/ml caused a significant increase in response latency ( P < .01) relative to controls while the higher dose of 100 ng/ ml caused a significant decrease ( P < .01) in response latency (Fig. 4). There was a significant interaction of CPF treatment

Fig. 3. Zebrafish spatial discrimination, escape/avoidance learning percent correct three blocks of six sessions of testing (mean ± S.E.M.).

E.D. Levin et al. / Neurotoxicology and Teratology 25 (2003) 51–57

55

ng/ml CPF dose group and in contrast there was a significant ( P < 0.01) decrease in response latency in the high 100 ng/ml CPF dose group. Smaller, though still significant, effects were seen during the middle session block. The 10 ng/ml group had a significant ( P < .01) increase of response latency while the 100 ng/ml CPF group had a significant ( P < .05) decrease in response latency. In the third session block, the CPF effect had diminished to the point where it was no longer significant.

4. Discussion CPF exposure during early development in zebrafish caused significant, long-lasting behavioral effects. CPF exposed zebrafish, like rats in earlier studies, showed significant CPF effects on motor activity and choice accuracy function. Both low (10 ng/ml) and high (100/ml) CPF doses during embryonic development had significant persisting effects. Both doses caused significant impairments in spatial discrimination. There was a biphasic effect on response latency with the lower CPF dose significantly slowing response and the higher CPF dose significantly reducing response latency. The higher dose of 100 ng/ml significantly increased mortality rates in the fish during adulthood, but the 10 ng/ml dose was subthreshold for this effect. Short-term CPF exposure during early development

Fig. 4. Zebrafish spatial discrimination, escape/avoidance response latency (seconds per trial) average over 18 sessions (mean ± S.E.M.).

with session block [ F(4,54) = 9.33, P < .0001]. As shown in Fig. 5, the CPF effects on response latency were pronounced during the early part of testing and diminished later. In the initial block, Dunnett’s tests showed that there was a significant ( P < .01) increase in response latency in the low 10

Fig. 5. Zebrafish spatial discrimination, escape/avoidance response latency (seconds per trial) three blocks of six sessions of testing (mean ± S.E.M.).

56

E.D. Levin et al. / Neurotoxicology and Teratology 25 (2003) 51–57

was found to cause disruptions in choice accuracy and motor function in zebrafish when they were tested in adulthood. Rats have also been found to show changes in choice behavior and locomotor activity in adulthood after developmental CPF exposure [12,13]. The zebrafish model offers an excellent opportunity to determine the molecular mechanisms for CPF impacts on neurodevelopment. The CPF effects on response latency were biphasic. The low 10 ng/ml CPF dose caused an increase in response latency, while the high 100 ng/ml CPF dose caused a decrease in response latency. Both effects were very robust with P-values less than .0001. Both the increase and decrease in response latency were readily apparent at the beginning of testing but became attenuated over the course of testing so that by the last block of sessions it was no longer apparent. The biphasic nature of the CPF effect on response latency suggests possible multiple mechanisms of CPF toxicity. Nonmonotonic dose-effects of prenatal CPF on radial-arm maze performance in rats have recently been documented in our laboratory [12]. The CPF effect on choice accuracy was, in contrast to response latency, univalent in expression. Both CPF doses caused significant impairments in choice accuracy relative to controls. The high dose caused a more pronounced impairment than the low dose. The expression over time of the effects of the two doses on choice accuracy differed. Over the course of testing the low dose had an early effect that diminished over time, while the high dose had an effect that became more pronounced over time. The significant effect of the low dose in the first block of test sessions may have been overcome by additional training. The later emergence of the impairment in the high dose group is still a phenomenon that needs further research to explain. There was a significant CPF Treatment ? Training Side interaction with a more pronounced CPF effect with training to the left side than to the right. The side preference may have been related to the position of the tank in the room. This is a good example highlighting the importance of balancing the training conditions in each treatment group. The high 100 ng/ml CPF dose caused a significant premature mortality in adult fish. More general health impairing effects in the high dose group may have also influenced behavior in this group. The low 10 ng/ml CPF dose group had no premature mortality; however, this dose still had significant effects on both response latency and choice accuracy. Similar spatial learning impairments were seen in our laboratory in adult rats after short-term prenatal exposure to low doses of CPF [12]. The zebrafish model should considerably help in the discovery of the molecular mechanisms of persisting effects of developmental CPF exposure. Zebrafish have become a central model for characterizing the molecular biology of neurodevelopment. The clear chorion and rapid development of zebrafish facilitates study of the molecular basis of neurobehavioral toxicity. The zebrafish model provides a means for visualizing devel-

oping nervous system [14]. To understand the molecular mechanisms of developmental neurotoxicity relevant to behavioral function, it is critical to efficiently and reliably determine the behavioral expression of behavioral function of zebrafish. Our laboratory has developed such methods [1]. In the current study, a short-term early CPF treatment was shown to cause long-lasting behavioral impairments that are expressed well into adulthood. This paradigm can now be used to determine the molecular underpinnings of CPF-induced developmental behavioral toxicity. Techniques well developed for zebrafish such as molecular microarrays and morphalinos can be used to map the molecular mechanisms of altered neurodevelopment responsible for these long-lasting CPF-induced behavioral impairments.

Acknowledgements The authors thank Paul Blackwelder, Jennifer Blackwelder and Jennifer Song for their assistance with behavioral testing and Dr. Theodore Slotkin for his advice concerning chlorpyrifos toxicology. This research was supported by NIH grants #ES10387, ES10356 and ES9808. These experiments were conducted under experimental protocols approved by the Duke University Animal Care and Use Committee in compliance with the laws and regulations of the United States of America.

References
[1] D. Arthur, E.D. Levin, Spatial and non-spatial discrimination learning in zebrafish, Anim. Cogn. 4 (2001) 125 – 131. [2] S. Barone, K.P. Das, T.L. Lassiter, L.D. White, Vulnerable processes of nervous system development: a review of markers and methods, Neurotoxicology 21 (2000) 15 – 36. [3] M. Behra, X. Cousin, C. Bertrand, J. Vonesch, D. Biellmann, A. Chatonnet, U. Strahle, Acetylcholinesterase is required for neuronal ¨ and muscular development in the zebrafish embryo, Nat. Neurosci. 5 (2002) 111 – 118. [4] R.L. Carr, H.W. Chambers, J.A. Guarisco, J.R. Richardson, J. Tang, J.E. Chambers, Effects of repeated oral postnatal exposure to chlorpyrifos on open-field behavior in juvenile rats, Toxicol. Sci. 59 (2001) 260 – 267. [5] K. Dam, S.J. Garcia, F.J. Seidler, T.A. Slotkin, Neonatal chlorpyrifos exposure alters synaptic development and neuronal activity in cholinergic and catecholaminergic pathways, Dev. Brain Res. 116 (1999) 9 – 20. [6] K. Dam, F.J. Seidler, T.A. Slotkin, Chlorpyrifos exposure during a critical neonatal period elicits gender-selective deficits in the development of coordination skills and locomotor activity, Dev. Brain Res. 121 (2000) 179 – 187. [7] K.P. Das, S. Barone Jr., Neuronal differentiation in PC12 cells is inhibited by chlorpyrifos and its metabolites: is acetylcholinesterase inhibition the site of action? Toxicol. Appl. Pharmacol. 160 (1999) 217 – 230. [8] J.R. Fetcho, K.S. Liu, Zebrafish as a model system for studying neuronal circuits and behavior, Ann. NY Acad. Sci. 860 (1998) 333 – 345. [9] D. Hall, M.D. Suboski, Sensory preconditioning and second-order conditioning of alarm reactions in zebra danio fish (Brachydanio rerio), J. Comp. Psychol. 109 (1995) 76 – 84.

E.D. Levin et al. / Neurotoxicology and Teratology 25 (2003) 51–57 [10] D. Hall, M.D. Suboski, Visual and olfactory stimuli in learned release of alarm reactions by zebra danio fish (Brachydanio rerio), Neurobiol. Learn. Mem. 63 (1995) 229 – 240. [11] P.J. Landrigan, L. Claudio, S.B. Markowitz, G.S. Berkowitz, B.L. Brenner, H. Romero, J.G. Wetmur, T.D. Matte, A.C. Gore, J.H. Godbold, M.S. Wolff, Pesticides and inner-city children: exposures, risks, and prevention, Environ. Health Perspect. 107 (Suppl. 3) (1999) 431 – 437. [12] E. Levin, N. Addy, A. Baruah, A. Elias, N. Christopher, F. Seidler, T. Slotkin, Prenatal chlorpyrifos exposure in rats causes persistent behavioral alterations, Neurotoxicol. Teratol., 2002 (in press). [13] E.D. Levin, N. Addy, A. Nakajima, N.C. Christopher, F.J. Seidler, T.A. Slotkin, Persistent behavioral consequences of neonatal chlorpyrifos exposure in rats, Dev. Brain Res. 130 (2001) 83 – 89. [14] H.-C. Park, C.-H. Kim, Y.-K. Bae, S.-Y. Yeo, S.-H. Kim, S.-K. Hong, J. Shin, K.-W. Yoo, M. Hibi, T. Hirano, N. Miki, A.B. Chitnis, T.-L. Huh, Analysis of upstream elements in the HuC promoter leads to the establishment of transgenic zebrafish with fluorescent neurons, Dev. Biol. 227 (2000) 279 – 293.

57

[15] C.N. Pope, Organophosphorus pesticides: do they all have the same mechanism of toxicity? J. Toxicol. Environ. Health 2 (1999) 161 – 181. [16] G. Pradel, M. Schachner, R. Schmidt, Inhibition of memory consolidation by antibodies against cell adhesion molecules after active avoidance conditioning in zebrafish, J. Neurobiol. 39 (1999) 197 – 206. [17] G. Pradel, R. Schmidt, M. Schachner, Involvement of L1.1 in memory consolidation after active avoidance conditioning in zebrafish, J. Neurobiol. 43 (2000) 389 – 403. [18] A.F. Schier, Genetics of neural development in zebrafish, Curr. Opin. Neurobiol. 7 (1997) 119 – 126. [19] T.A. Slotkin, Developmental cholinotoxicants: nicotine and chlorpyrifos, Environ. Health Perspect. 107 (Suppl. 1) (1999) 65 – 69. [20] M.D. Suboski, Acquisition and social communication of stimulus recognition by fish, Behav. Processes 16 (1988) 213 – 244. [21] F.E. Williams, W.S. Messer Jr., Memory function and muscarinic receptors in zebrafish, Soc. Neurosci. Abstr. 24 (1998) 182.


相关文章:
更多相关标签: