ToxSci Advance Access originally published online on December 2, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Toxicological Sciences 77, 341-346 (2004)
Copyright © 2004 by the Society of Toxicology
NEUROTOXICOLOGY |
Structure-Activity and Interaction Effects of 14 Different Pyrethroids on Voltage-Gated Chloride Ion Channels
MRC Applied Neuroscience Group, School of Biomedical Sciences, University of Nottingham, Nottingham NG7 2UH, United Kingdom
Received July 31, 2003; accepted October 14, 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have proposed that since the type II pyrethroids deltamethrin and cypermethrin, but not the type I pyrethroid cismethrin act on chloride channels, this could contribute to the bimodal nature of pyrethroid poisoning syndromes. We now examine a wider range of pyrethroid structures on the activity of these calcium-independent voltage-gated maxi-chloride channels. Excised inside-out membrane patches from differentiated mouse neuroblastoma cells were used, and mean channel open probabilities calculated. For single dosing at 10 µM, bioallethrin, ß-cyfluthrin, cypermethrin, deltamethrin, and fenpropathrin were all found to significantly decrease open channel probability (p < 0.05). Bifenthrin, bioresmethrin, cispermethrin, cisresmethrin, cyfluthrin isomers 2 and 4,
-cyhalothrin, esfenvalerate, and tefluthrin, did not significantly alter open channel probability (p > 0.05). Since the type II pyrethroids, esfenvalerate, and -cyhalothrin were ineffective, we must conclude that actions at the chloride ion channel target cannot in themselves account for the differences between the two types of poisoning syndrome. Sequential dosing with type II pyrethroids caused no further chloride ion channel closure. The type I pyrethroid cisresmethrin did however prevent a subsequent effect by the mixed type pyrethroid fenpropathrin. In contrast, the type I pyrethroid cispermethrin did not prevent a subsequent effect due to the type II pyrethroid deltamethrin. The difference in effect may be the result of differences in potency, as deltamethrin had a greater effect than fenpropathrin. It therefore appears clear that in some combinations the type I and type II pyrethroids can compete and may bind to the same chloride channel target site. Key Words: pyrethroid; chloride ion channel; patch clamp; neuroblastoma.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It has been previously established that all pyrethroids have a common mechanism of action upon the voltage-gated membrane sodium channel (Chinn and Narahashi, 1986
). Their toxicology is dominated by acute pharmacological actions on excitability originating from this common mechanism (Ray, 2001). However this acute toxicity is manifested in different ways, and most pyrethroids divide into two distinct classes which produce very different poisoning syndromes (type I and II) in both insects and mammals (Ray, 2001; Verschoyle and Aldridge, 1980). Although some pyrethroids produce a combination of type I and type II syndromes, the classes can be unambiguously separated by a number of electrophysiological indices that have been developed in the rat (Wright et al., 1988). When examined at the level of the sodium channel however, there is a continuous spectrum of effect across pyrethroid structures. There is also continuous variability in some other secondary electrophysiological indices such as prolongation of hippocampal paired pulse inhibition (Joy et al., 1990). This continuous variability across structures contrasts markedly with the bimodal effect seen in the whole animal. It appears likely that this bimodal effect is the result of either heterogeneity within the sodium channel targets (Soderlund et al., 2002) or to additional actions of the type II pyrethroids on some target (or targets) other than the sodium channels.
One additional target of type II pyrethroids is the membrane chloride ion channel. The ligand-gated chloride channel was first proposed as a target (Gammon et al., 1982), but does not appear to be sufficiently sensitive to have an influence at doses relevant to systemic poisoning. The voltage-gated chloride channel was then proposed as a target (Forshaw et al., 1993; Forshaw and Ray, 1990), and for deltamethrin at least, this does appear to be sufficiently sensitive, both on the basis of in vitro potency (Ray et al., 1997), and in vivo effects (Forshaw and Ray, 1990). Indeed it is possible to antagonize both the salivation and choreoathetosis, which are the prominent characteristics of type II pyrethroid poisoning, with chloride channel agonists (Forshaw et al., 2000). Furthermore the open channel probability of chloride channels decreases with the type II pyrethroids deltamethrin and cypermethrin, but not the type I pyrethroid cismethrin (Ray et al., 1997). All this suggests that type II poisoning results from a combined action on the voltage-gated sodium and chloride channels, whereas type I poisoning probably results from an action on the sodium channels alone. Other additional primary targets have also been proposed, such as calcium channels (Hagiwara et al., 1988) and these also appear to play a part in defining the poisoning syndromes (Symington et al., 1999). To investigate the chloride channel effect further we have studied the action of 14 different pyrethroids on the activity of voltage-gated chloride channels to better evaluate the hypothesis that the type II syndrome is related to chloride channel action. We also tried sequential dosing with one type of pyrethroid followed by another to investigate whether there is any interaction between pyrethroids at this chloride channel target site, as has been found for sodium channels (Song et al., 1996). Pyrethroids were selected on the basis of commercial use, and hence the pyrethroids most likely to lead to human exposure were investigated.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell culture.
Mouse neuroblastoma cells (N1E-115) were grown in DMEM culture medium supplemented with glutamine ("Glutamax-1," Glibco 21885-025) and 7.5% fetal calf serum. Stock flasks were kept at 37°C and 5% CO2 in a humidified incubator; the medium being refreshed every 2–3 days and subcultures made when growth became confluent. Subcultured cells were grown on 58 mm dishes to allow easy access for experimentation on the microscope stage (Zeiss Axiovert 100). All cells had previously undergone 20–50 passages, and "patching" was undertaken 2–6 days after subculturing, i.e., when expressing neurite outgrowths.
Patch-clamp technique.
Microelectrode pipettes were made from thin-walled inner filament borosilicate glass capillary tubing. Tips were forged on a two-stage vertical pipette puller and fire-polished. Tips had a final diameter of 1–2 µm, giving a resistance of 5–20 M
. All experiments were carried out at room temperature (19–25°C), and excised inside-out patches were used. While maneuvering the pipette prior to "patching" a positive pressure was applied to prevent the tip from clogging with debris. When the tip was adjacent to a suitable cell membrane the positive pressure was stopped. Gentle suction (10–30 mm H2O) was applied if a spontaneous seal did not form. Successful seal resistances varied from 1–33 G between experiments. The electrode tip potential was manually corrected to 0 pA at 0 mV before each experiment. This correction current remained constant (±3 pA) within an experiment. Both seal resistance and correction current were monitored before the first recording, at dosing, and after the last recording. Patches were held at 0 mV and positive or negative potentials applied in 20 mV steps as required. Clear gating activity with marked sub-states and a unitary conductance of around 200–400 pS, confirmed that the channels under investigation were of the maxi-chloride subtype (Forshaw et al., 1993).
Solutions.
The pipette and bath dish contained the same solution (144 mM NaCl; 3 mM CsCl; 1 mM EGTA; 0.87 mM CaCl2 and 5 mM HEPES Na, all buffered to pH 7.2 with 1 mM NaOH). A trace amount of calcium equivalent to 1 µM of free Ca2+ was needed to facilitate patch seals, but this level was low enough to ensure that only the calcium-independent maxi-chloride channels that have already been shown to be sensitive to selective block by type II pyrethroids would be active (Forshaw et al., 1993). The bath solution was passed through a 0.2-µm filter to remove debris immediately prior to use.
Pyrethoid delivery.
Pyrethroids (Table 1) were dissolved in dimethylsulphoxide (DMSO) at concentrations to ensure that with each dose the DMSO added would consistently contribute 0.1% (5 µl) to the total bath volume of 5 ml. Each pyrethroid was administered to give a nominal bath concentration of 10 µM. The fixed dose was set deliberately high, well above the deltamethrin threshold (Ray et al., 1997), in order to ensure that any positives were clear and that any negatives were unambiguous.
|
Doses were delivered to the intracellular membrane surface by injection into the bath, within 5 mm of the microelectrode pipette tip, over a period of approximately 10 s, by graded application from a 5 µl Hamilton syringe. Deltamethrin was utilized for positive control experiments. Negative control experiments were carried out with DMSO alone. Pyrethroid doses were delivered in the same way as in previous work with deltamethrin, and this yielded a fine precipitate, pyrethroids only remaining in true solution below 10-9 M (Ray et al., 1997
). Hence the 10 µM concentration used in this study must be considered as nominal.
Patch clamp data acquisition and analysis.
Hardware comprised an Axon Instruments Axopatch-1D, with CV4-1/100 Headstage, and Axon Digidata 1200 data acquisition digitizing board, routed through a personal computer. Acquisition software was Clampex (version 8.0.2.113; Axon Instruments). Once a patch was obtained a continuous negative holding potential of -40 mV was applied. Most patches initially showed no activity at -40 mV, and so the voltage was gradually stepped up to -140 mV over a 20-min period, or until clear gating activity of channels was observed. The patch was discarded if no activity was evident. As soon as clear activity was observed the holding potential was then reversed and dropped to the lowest voltage (+20 or +40 mV) at which activity continued. Ten pre-dose recordings (each of 10 s duration) were made at 1-min intervals. Within patches, the holding potential was held at the same level for all recordings and was 0 mV between each recording. The preparation was then dosed (as described above). Ten post-dose recordings (again each of 10 s duration) were made at 1-min intervals. Recordings were continued at 2-min intervals for a further 10 min. In the case of sequential dosing, the second dose was then administered and recording resumed at 1-min intervals as before. Only stable patches with scope for both increase and decrease in activity upon dosing were included.
Software analysis utilized Clampfit (version 8.0.2.113; Axon Instruments), Fetchan (version 6.0.6.01; Axon Instruments), and Pstat (version 6.0.5.07; Axon Instruments). The degree of chloride channel activity was determined by open channel probability (Po) analysis. Po was calculated from the mean current recorded over the 10 s for each recording. Within any one experiment, the patch was held at a fixed voltage once stable activity developed. Open channel probability has previously been shown not to change in a time-dependent manner for these channels (Forshaw et al., 1993).
Appropriate paired two-tailed statistical tests were used to compare Po values for different treatments. F-tests revealed that none of the paired groups had significantly unequal variance. The normality of data distribution (as assessed by skew, kurtosis, and Shapiro-Wilk test) determined whether either a type I T test or a Wilcoxon sign-ranked test for two related samples was used.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Single Pyrethroids
For all included patches (n = 100), the estimated single channel conductance was 286 ±68 pS (mean ± SD), while the number of channels in a patch was 4 ± 2 (mean ± SD). Dosing with DMSO alone caused no change in open channel probability (Tables 2
and 3). Bioallethrin, ß-cyfluthrin, cypermethrin, deltamethrin, and fenpropathrin all significantly decreased open channel probability (p < 0.05, Fig. 1 shows an example of recordings). In contrast, bifenthrin, bioresmethrin, cispermethrin, cisresmethrin, cyfluthrin isomers 2 and 4, cyhalothrin, esfenvalerate, and tefluthrin did not significantly alter open channel probability (p > 0.05). Thus, not all of the type II pyrethroids tested affected chloride ion channels (Table 2). Deltamethrin had the most efficacious effect on open channel probability (Table 2 and Fig. 2A).
|
|
|
|
Pyrethroid Combinations
Our null hypothesis was that the second dose would be additive. The combination of two active pyrethroids (cypermethrin followed by cyfluthrin) resulted in no significantly greater fall in Po than was seen after cypermethrin alone (Table 3
). The combination of pairs of inactive pyrethroids (bifenthrin and -cyhalothrin, cyfluthrin isomer 2 and bioresmethrin, cyfluthrin isomer 4 and tefluthrin, -cyhalothrin and bifenthrin, bioresmethrin and cyfluthrin isomer 2) produced no significant combined effect. The effects of deltamethrin, cypermethrin, bioallethrin, and fenpropathrin at first dose did not continue to be significant into the post second dose period. This appeared to be the result of variability over time, rather than antagonism. Where an active pyrethroid followed an inactive one, the effect was variable, deltamethrin still being active when preceded by cispermethrin (although less so; Fig. 2B), but the less potent fenpropathrin failing to produce any effect when preceded by cisresmethrin (Fig. 2D). Hence the only pyrethroid tested that resulted in a significant fall in open channel probability after the second dose, was deltamethrin after cispermethrin (Table 3 and Fig. 2B). However, this decrease was not as significant as with deltamethrin alone (Tables 2 and 3). Where an inactive pyrethroid followed an active one, dosing with cispermethrin after deltamethrin resulted in a partial recovery of open channel probability, abolishing the significance of the fall due to deltamethrin, (Table 3 and Fig. 2A). Similarly, dosing with cisresmethrin after fenpropathrin resulted in a partial recovery of open channel probability, abolishing the significance of the fall due to fenpropathrin (Table 3 and Fig. 2C). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In accordance with our previous studies (Forshaw et al., 1993
; Ray et al., 1997), both deltamethrin and cypermethrin were found to effectively suppress the open state of voltage-gated chloride channels, the nature and time course of the changes being precisely as previously described. We have however markedly extended the range of pyrethroid structures previously examined. One difference from our previous results was that the present dosing vehicle, DMSO appeared not to have any effect on the action of the chloride channel. This was in contrast to the membrane stabilizing effect previously seen with the solvent glycerinformal (also at 0.1%; Forshaw et al., 1993). Overall, 3/6 of the type II pyrethroids, 2/2 of the mixed type pyrethroids, and 0/3 of the type I pyrethroids were active at the chloride ion channel. 3/14 pyrethroids were of unknown type.
Although our original observations with deltamethrin and cypermethrin (Ray et al., 1997) were reproduced, esfenvalerate and -cyhalothrin, which are type II pyrethroids (Wright et al., 1988), did not reduce the activity of chloride channels. No type I pyrethroids were active but bioallethrin, which has been described as a type I pyrethroid (Gammon et al., 1982), but may perhaps be of mixed type (Soderlund et al., 2002), did produce a small but significant effect. This demonstrates that not all type II pyrethroids act at chloride channels, while (possibly) one type I pyrethroid does act at chloride channels. One unusual result is that ß-cyfluthrin significantly reduced open channel probability, but isomers 2 and 4 (which constitute more than 90% of ß-cyfluthrin; see Table 1) were individually ineffective. This implies that either one or both of the minor isomers of ß-cyfluthrin caused the observed closure of chloride ion channels, or that there is an interaction between the individual isomers in order to cause the effect.
Overall the results were unexpected (Forshaw et al., 1993, 2000) and lead us to conclude that an additional action at chloride channels cannot in itself explain the dichotomy in symptoms between type I and type II poisoning. Nevertheless, action at chloride channels would still be expected to synergize sodium channel actions by increasing membrane resistance (Ray et al., 1997). Our results further indicate that the closer a pyrethroid structure resembles that of deltamethrin, the more likely it will have an effect on chloride channels. There were three exceptions (Fig. 3): Bioallethrin was effective and yet lacks a cyano group; while -cyhalothrin and esfenvalerate have cyano groups but were not effective. It appears that an alcohol moiety identical (or very similar) to that of deltamethrin is preserved in pyrethroid structures that are effective at chloride channels; whereas pyrethroids with a larger acid moiety were less effective are at closing chloride channels.
|
Song et al.(1996)
, studying the sodium channel response of dorsal root ganglion cells to the type II pyrethroid fenvalerate, found that subsequent addition of the type I pyrethroid tetramethrin abolished the fenvalerate response and replaced it by a response very similar to that generated by tetramethrin alone. Although in the case of chloride channels there is no type I response, we also investigated the effect of some pyrethroid combinations. Although the effect of fenpropathrin was not significantly reduced by subsequent treatment with the inactive cisresmethrin, pre-treatment with cisresmethrin did prevent the fenpropathrin response. Thus a type I pre-treatment is antagonizing the effect of a mixed type pyrethroid at the chloride channel, in an analogous way to that described by Song et al.(1996) for the sodium channel. In contrast, although the effect of our most potent pyrethroid deltamethrin was again not significantly reduced by subsequent treatment with the inactive cispermethrin, pretreatment with cispermethrin failed to prevent a significant deltamethrin response. Hence our results provide only limited support for antagonism between pyrethroids at the chloride channel. It appears clear that while an ineffective pyrethroid (cispermethrin) did not interact in any way with our most potent pyrethroid (deltamethrin), another ineffective pyrethroid (cisresmethrin) did interact with a less potent pyrethoid (fenpropathrin). This provides evidence for a negative allosteric interaction between pyrethroids. We also tested several pairs of ineffective pyrethroids (e.g., bifenthrin and -cyhalothrin), and one pair of effective pyrethroids (cypermethrin and cyfluthrin). None of these combined to produce a significant additive effect. However we know that the dose-response relationship for deltamethrin in our system is shallow (Ray et al., 1997), possibly due to the hydrophobic nature of pyrethroids limiting their bioavailability via saline. This in turn may explain why some of the combinations of pyrethroids that we used showed no additional effect of the second administration.
|
ACKNOWLEDGMENTS |
|---|
This research was supported by the Pyrethroid Working Group, a consortium of firms that market pyrethroid insecticides in the United States. COI: The authors acknowledge that they have a grant from the Pyrethroid Working Group to do research in this area; the funding organization does not have control over the resulting publication.
|
NOTES |
|---|
1 To whom correspondence should be addressed. E-mail: steven.burr{at}nottingham.ac.uk.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Chinn, K., and Narahashi, T. (1986). Stabilization of sodium channel states by deltamethrin in mouse neuroblastoma cells. J. Physiol. 380, 191–207.
Forshaw, P. J., and Ray, D. E. (1990). A novel action of deltamethrin on membrane resistance in mammalian skeletal muscle and non-myelinated nerve fibres. Neuropharmacology 29, 75–81.
Forshaw, P. J., Lister, T., and Ray, D. E. (1993). Inhibition of neuronal voltage-dependent chloride channel by the type II pyrethroid, deltamethrin. Neuropharmacology 32(2), 105–111.[CrossRef][Web of Science][Medline]
Forshaw, P. J., Lister, T., and Ray, D. E. (2000). The role of voltage-gated chloride channels in type II pyrethroid insecticide poisoning. Toxicol. Appl. Pharmacol. 163, 1–8.[CrossRef][Web of Science][Medline]
Gammon, D. W., Lawrence, L. J., and Casida, J. E. (1982). Pyrethroid toxicology: Protective effects of diazepam and phenobarbital in the mouse and cockroach. Toxicol. Appl. Pharmacol. 66, 290–296.[CrossRef][Web of Science][Medline]
Hagiwara, N., Irisawa, H., and Kameyama (1988). Contribution of two types of calcium currents to the pacemaker potentials of rabbit sino-atrial node cells. J. Physiol. 395, 233–253.
Joy, R. M., Lister, T., Ray, D. E., and Seville, M. P. (1990). Characteristics of the prolonged inhibition produced by a range of pyrethroids in the rat hippocampus. Toxicol. Appl. Pharmacol. 103, 528–538.[CrossRef][Web of Science][Medline]
Ray D. E. (2001). Pyrethroid insecticides: Mechanisms of toxicity, systemic poisoning syndromes, parasthesia, and therapy. In Handbook of Pesticide Toxicology (R. Kreiger, Ed.), pp. 1289–1303. Academic Press, London.
Ray, D. E., Sutharsan, S., and Forshaw, P. J. (1997). Actions of pyrethroid insecticides on voltage-gated chloride channels in neuroblastoma cells. Neurotoxicology 18(3), 755–760.[Medline]
Soderlund, D. M., Clark, J. M., Sheets, L. P., Mullin, L. S., Piccirillo, V. J., Sargent, D., Stevens, J. T., and Weiner, M. L. (2002). Mechanisms of pyrethroid neurotoxicity: Implications for cumulative risk assessment. Toxicology 171, 3–59.[CrossRef][Web of Science][Medline]
Song, J. H., Nagata, K., Tatebayashi, H., and Narahashi, T. (1996). Interactions of tetramethrin, fenvalerate and DDT at the sodium channel in rat dorsal root ganglion neurons. Brain Res. 708, 29–37.[CrossRef][Medline]
Symington, S. B., Zhang, A., Karstens, W., Van Houten, J., and Clark J. M. (1999). Characterization of pyrethroid action on ciliary calcium channels in Paramecium tetraurelia. Pestic. Biochem. Physiol. 65, 181–193.[CrossRef]
Verschoyle, R. D., and Aldridge, W. N. (1980). Structure-activity relationships of some pyrethroids in rats. Arch. Toxicol. 45, 325–329.[CrossRef][Web of Science][Medline]
Wright, C. D. P., Forshaw, P. J., and Ray, D. E. (1988). Classification of the actions of ten pyrethroid insecticides in the rat, using the trigeminal reflex and skeletal muscle as test systems. Pestic. Biochem. Physiol. 30, 79–86.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Imamura, M. Yasuda, K. Kuramitsu, D. Hara, A. Tabuchi, and M. Tsuda Deltamethrin, a Pyrethroid Insecticide, Is a Potent Inducer for the Activity-Dependent Gene Expression of Brain-Derived Neurotrophic Factor in Neurons J. Pharmacol. Exp. Ther., January 1, 2006; 316(1): 136 - 143. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||