* Department of Biochemistry and Molecular Biology, The University of Queensland, St. Lucia, QLD 4072 Australia; and
Queensland Department of Primary Industries, 80 Meiers Road, Indooroopilly, QLD 4068 Australia
Received October 2, 2002; accepted January 23, 2003
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: methyl bromide; oxidative stress; oxygen toxicity; ROS; mitochondrial dysfunction; EMS mutagenesis; clk-1; insecticide; fumigant; hyperoxia.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Phosphine has two overriding chemical properties that could have a significant bearing on its toxicity. The first of these is that it is a strongly reducing compound that may interrupt many biological redox systems (Chaudhry and Price, 1991). Secondly, phosphine readily complexes with metal ions and may act as a substrate or as an inhibitor of metal-containing enzymes (Lam et al., 1991
). Chaudhry and Price (1990)
have combined these two features in a model which proposes that electron donation by phosphine is facilitated by its interaction with transition metals, such as iron. In addition, phosphine toxicity is absolutely dependent on oxygen (Kashi, 1981
). However, the nature of a potential interaction between phosphine, transition metals, and oxygen in vivo is completely unknown. The end result of phosphine exposure is oxidative damage to lipids, proteins, and DNA, any of which could be the direct cause of death following phosphine exposure (Hsu et al., 1998
). As expected, given the proposed oxidative stress mechanism of toxicity, antioxidants have been shown to reduce or eliminate the oxidative damage to macromolecules observed in phosphine-exposed rats (Hsu et al., 2000
).
It is widely stated that mitochondria are the likely source of reactive oxygen species (ROS) responsible for phosphine-mediated oxidative damage to macromolecules. Indeed, mitochondria are known sources of significant quantities of ROS due to errors in electron transport. Phosphine is also able to interact with the reactive metal containing center of cytochrome c oxidase, disrupting electron flow through the mitochondrial electron transport chain (Chaudhry and Price, 1990; Kashi and Chefurka, 1976
). There is, however, no evidence of sustained production of ROS from mitochondria during prolonged phosphine exposure. In addition, complex IV is only slightly inhibited by phosphine in vivo (Price, 1980
), and only a small fraction of the oxidation products of phosphine are actually found in the mitochondrial compartment (Robinson and Bond, 1970
). Furthermore, strongly increased activity of the protective enzymes normally expected in response to ROS generation (e.g., superoxide dismutase and peroxidases including catalase) has not been observed in response to phosphine exposure (Bolter and Chefurka, 1990
). Therefore, an alternative model of toxicity has been proposed recently in which phosphine is directly involved in the chemical reactions that result in oxidative damage (Quistad et al., 2000
).
More recently, a molecular genetic study revealed two loci responsible for high level resistance in the lesser grain borer, Rhyzopertha dominica (Schlipalius et al., 2002). Identification of the genes conferring phosphine resistance for this species is complicated by the lack of molecular genetic tools. A productive alternative to studying phosphine resistance in a pest organism is to use a genetically well-characterized model organism such as the free-living soil nematode, Caenorhabditis elegans. The advantages offered by this organism are its small size and its rapid reproduction as a self-fertile hermaphrodite. The latter facilitates creation of many thousands of inbred mutant lines for genetic selection of resistance to toxicants. Analysis of mutants is facilitated by molecular genetic tools such as the published genome sequence and techniques for either overexpression or suppression of genes. In addition, the nematodes grow on the surface of a solid medium, facilitating the application of either gases or dissolved chemicals for toxicity analysis. In contrast, quantitative application of chemicals to pest insects that burrow within grain is much more difficult.
In the present study, we created and characterized three mutant lines of C. elegans that express significant levels of resistance to the fumigant phosphine. We explored the synergistic effect of oxygen exposure on phosphine toxicity and have extended this analysis to include the link between stress resistance and longevity. Additionally, we assessed the developmental rate and reproductive capacity of each mutant line.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
C. elegans mutagenesis.
Adult wild-type nematodes were allowed to lay eggs for 3 h on NGM plates seeded with E. coli strain OP50, after which they were washed from the plates with M9 buffer (0.6% Na2HPO4, 0.3% KH2PO4, 0.5% NaCl and 0.025% MgSO4). The plates were placed at 25°C for 40 h allowing the nematodes to mature to the fourth larval stage (L4). The L4 nematodes were washed free of bacteria by repeated suspension in M9 buffer, followed by gentle centrifugation and removal of the wash solution. Fifty mM ethylmethane sulfonate (EMS) was added to the nematodes in 1.5 ml M9 buffer, and mutagenesis was allowed to proceed for 4 h with periodic gentle agitation. Following exposure to EMS, the nematodes were washed 4 times with M9 buffer and allowed to recover on NGM agar plates seeded with OP50 for 4 h. Four mutated hermaphrodites were then transferred to each of 39 NGM agar plates, again seeded with OP50. After 17 h, the adults were removed and the ~200 eggs per plate were allowed to hatch, grow into adults, and reproduce for 7 days with supplementary feeding with an OP50 slurry as required. Nematodes of mixed ages were transferred to fresh plates and selected at 1 mg/l phosphine in air at 25°C for 24 h. After recovery, individual survivors were transferred to fresh, seeded NGM plates and were left to lay eggs for several days. The adults were removed and the offspring were allowed to grow and reproduce for 7 to 10 days, followed by selection at 1 mg/l phosphine, as before. Selection was repeated for a total of 3 to 5 cycles to allow lines to be assessed for a reliable and robust resistance phenotype. Lines with a less robust resistance phenotype sometimes failed to produce offspring after recovery, so up to 5 survivors were transferred with each round of selection following the initial round in which survivors were subcultured individually.
Phosphine and oxygen exposure.
Phosphine gas was generated by exposing aluminum phosphide tablets to a solution of 5% sulfuric acid. The concentration of the collected gas was measured by gas chromatography, using nitrogen (N2) as a standard and Freon-24 as the carrier. Phosphine exposures were performed in desiccation chambers fitted with a septum through which phosphine could be injected to achieve the desired concentration. To achieve the hyperoxic environments used to study phosphine-oxygen synergism, pure oxygen was flushed into the chamber prior to phosphine injection. The final oxygen concentration was determined using a GPR-20F oxygen analyzer (Advanced Instruments, Inc.). Exposures were conducted at 25°C for 24 h.
Analysis of phosphine toxicity.
Developmentally synchronized nematodes were obtained by treating mixed-age nematodes with alkaline hypochlorite solution (1.4% sodium hypochlorite, 0.7 N sodium hydroxide) to cause release of eggs. The eggs were rinsed with M9 buffer three times and allowed to hatch in M9 buffer overnight at 20°C. Development was then initiated in the newly hatched L1 nematodes by transfer to NGM plates seeded with OP50 bacteria; 24-h treatment of nematodes with the desired concentrations of phosphine was started at exactly 42 h after initiation of feeding. The mortality of the wild-type line, N2, was determined at phosphine concentrations ranging from 0.1 to 1.0 mg/l, whereas the phosphine-resistant mutants were tested up to 5.0 mg/l. Nematodes exposed to concentrations of phosphine greater than 0.1 mg/l were allowed to recover at 20°C for 48 h. Because individuals treated with lower concentrations recovered more quickly, it was necessary to assess survival at 24 h prior to production of progeny. Individuals were assumed to be dead if they did not initiate movement in response to flooding the plates with M9 buffer. Experiments on all strains were replicated at least twice, except that mortality of pre-7 and pre-33 was determined three times. Results were corrected for control mortality (10% or less in all cases) using Abbotts formula (Abbott, 1925). Probit, logit, and complementary log-log (CLL) transformations were tested for fit of the data to concentration-response curves using Genstat 6 (Genstat, 2002
). CLL transformation provided the best fit to the data, as residual values were least with this transformation. Data fitted the general formula:
![]() |
LC50 values and their 95% fiducial limits were calculated from the regression analyses for each mutant population. Nonoverlap of 95% fiducial limits was used as the test for significance.
Analysis of phosphine and oxygen synergism.
Phosphine-induced mortality under hyperoxic conditions was determined as above, except that the phosphine concentration was maintained at 0.1 mg/l and the oxygen concentration was varied from 20.9% to 80%. The effect of hyperoxia without phosphine was simultaneously determined for each strain. The entire experiment, hyperoxia with and without phosphine, was replicated, and analysis of variance was used to compare treatment means. The data were analyzed as a two-way ANOVA with no blocking using Genstat 6 (Genstat 2002). Significant differences between treatments were determined using a least significant difference (LSD) test.
Lifespan.
Synchronized hermaphrodites of all four strains, pre-1 (n = 61), pre-7 (n = 66), pre-33 (n = 75), and N2 (n = 101) were grown on normal NGM plates at 20°C for 48 h, after which they were transferred to plates containing 40 µM 5-fluoro-2'-deoxyuridine (Sigma), which disrupts DNA synthesis, thereby preventing progeny production. Adults are not affected, because cell division no longer occurs in any of the somatic cells of the adult. Mortality was scored daily with dead individuals being identified by lack of response to physical touch, and dead nematodes were removed from the plates each day. Lifespan is reported as mean ± SE. Significance was assessed by Students t-test and a p value less than 0.05 was considered to be significant.
Developmental rates and fecundity.
Individual hermaphrodites were allowed to mature on NGM plates seeded with OP50 bacteria until egg laying began, after which time they were transferred to fresh plates every 24 h. Offspring were allowed to mature prior to being counted. For each strain, two independent experiments were conducted and 911 nematodes were used in each replicate. To estimate developmental rate, median reproductive age, the age at which 50% of progeny have been produced (Sokal and Rohlf, 1981) was measured and a mean value calculated for each nematode line. Fecundity was reported as the average number of progeny produced per individual ± SE for each line. Significance was assessed by Students t-test and a p value less than 0.05 was considered to be significant.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Mutants were selected as individuals that were able to survive and produce offspring following exposure to phosphine for 24 h at 1 mg/l in aira discriminating concentration above which the wild-type nematode cannot survive (Fig. 1). A single mutant was retained from each of ten of the 39 plates established from the nematodes originally mutated using EMS. Each of these nematodes was used to found a separate mutant line. Three (pre-1, pre-7, and pre-33) of these were very robust and could recover quickly from selection at 1 mg/l phosphine for 24 h at 25°C in air, and were used in this study. In general, mobility of these nematodes was only slightly inhibited when they were viewed immediately following exposure. Maturation also appeared to be only slightly impeded relative to nonphosphine-treated controls. Notably, egg development was prevented during phosphine exposure, but egg laying recovered within two days once phosphine exposure was terminated. The three mutant lines were out crossed to the wild-type strain twice each before being used for toxicology studies, thereby reducing the possibility of secondary mutations interfering with phenotypic analysis of the phosphine resistance mutations. The seven remaining mutant lines were found to be less resistant to phosphine at 1 mg/l and were not studied further.
|
Phosphine synergism with oxygen.
Without phosphine, even high concentrations of oxygen are completely nontoxic for both wild-type and mutant nematodes during the 24-h period of this experiment (data not shown). Similarly, a concentration of 0.1mg/l phosphine is not toxic when administered in normal air (Fig. 1). When exposed to 0.1 mg/l phosphine; however, the wild-type line, N2, showed dose-dependent mortality up to 95% at 60% oxygen, which increased to 100% at 80% oxygen (Fig. 2
). In contrast, pre-33 was completely resistant to the synergistic effect of oxygen under the conditions employed. Mutants pre-7 and pre-1 showed 30% and 66% mortality, respectively, when exposed to 0.1 mg/l phosphine under 80% oxygen. According to the LSD tests, mortality results between each of the four lines are significantly different at oxygen concentrations of 60% or above, under 0.1mg/l of phosphine. The rank order of resistance of the four strains toward hyperoxia, in the presence of minimal phosphine, is identical to their rank order of resistance toward phosphine in normal air (Fig. 1
). Furthermore, the effect of hyperoxia under elevated phosphine (0.6 mg/l) on pre-33 was also determined, and no significant increase in mortality was observed, even under 80% oxygen.
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We observed that phosphine toxicity is dose-dependent for not only wild-type but for all three mutant lines of C. elegans as well. The mutants, however, show LC50 values that are 2-, 5-, and 9-fold greater than wild type. Interestingly, 99% mortality was not achieved for two of the mutants (pre-7 and pre-33), even after 24 h at the extremely high concentration of 5 mg/l phosphine (Fig. 1). This situation is similar to that observed with insects that have acquired high level phosphine resistance (Daglish, 2002
), as significant survival occurs even at very high concentrations of phosphine unless the exposure time is extended. This reflects the fact that phosphine has an unusual mode of action that does not obey Habers rule, which states that a constant level of mortality should be achieved if the concentration of fumigant multiplied by the time of exposure is constant (Du Bois and Geiling 1959
). Therefore, a low concentration of phosphine applied for an extended period of time disproportionately enhances the toxicity of phosphine. This unusual interaction is consistent with an indirect mode of action of phosphine such as the initiation of oxidative stress, which, when propagated over time by free radicals, results in the accumulation of damage to macromolecules to a lethal threshold. The rapid development of C. elegans (3 days for the complete life cycle) makes it difficult to test extended exposure periods much beyond the 24 h that we used in this study because of the confounding influences of cuticular weakening during larval molting and the increase in metabolic rate associated with reproductive development.
Previous work on insects has shown that phosphine toxicity is absolutely dependent on a concentration of oxygen of at least 1% (Bond and Monor 1967; Kashi 1981
). This work has never been extended to determine whether hyperoxic conditions (>20.9% O2) enhance phosphine toxicity. Using wild-type nematodes, we have now demonstrated a dose-dependent, synergistic enhancement of phosphine toxicity with increasing oxygen concentration. This synergistic action of oxygen is consistent with phosphine toxicity being the direct result of induced oxidative stress. We also tested whether the mutations that conferred phosphine resistance also conferred resistance to the synergistic interaction between phosphine and oxygen as observed in the wild-type C. elegans. We not only observed cross-resistance to the synergistic effect of oxygen with one mutant, pre-33, showing complete resistance to this synergism, but the rank order of this resistance was identical between the oxygen and phosphine dose response experiments. The apparent coupling of resistance toward both phosphine and the synergistic effect of oxygen indicates that the mutations either enhance the ability of the organisms to cope with oxidative stress or they suppress the initiation or propagation of oxidative damage.
Hyperoxic conditions similar to those used in our experiments have also been observed to decrease lifespan in C. elegans (Honda et al., 1993) as a direct result of increased oxidative stress. Accordingly, the relationship between oxidative stress and both phosphine toxicity and lifespan determination suggest that phosphine-resistant mutants could be expected to exhibit a long-lived phenotype. Indeed, all three of the mutants show a lifespan extension (Fig. 3
) and the average lifespan is greatest for pre-33, the mutant with the greatest resistance to phosphine (Fig. 1
). The pleiotropy of all three mutations, which exhibit not only phosphine resistance but also longevity, matches the previous observation that long-lived C. elegans mutants are resistant toward a range of environmental stressors such as oxidative stress, heat shock, and UV exposure (Lithgow and Walker, 2002
).
Despite the importance of the reaction between phosphine and oxygen in phosphine toxicity, mixtures of phosphine and molecular oxygen are relatively stable in vitro under normal temperature and pressure (Fluck, 1973). Oxidation of phosphine, however, occurs quite readily in vivo (Lam et al., 1991
; Robinson and Bond, 1970
). This implies that oxidation of phosphine by molecular oxygen is facilitated by a cellular redox catalyst such as proteins that contain heme iron, some of which are known to interact with both phosphine and molecular oxygen (Chaudhry and Price, 1990
; Kashi and Chefurka, 1976
).
Pratt (in press) recently found that phosphine uptake is directly coupled to phosphine oxidation. The essential role of molecular oxygen in phosphine toxicity suggests that the oxidation of phosphine during uptake involves molecular oxygen. Because neither phosphine in the absence of molecular oxygen (Kashi, 1981
) nor the oxidation products of phosphine are themselves toxic (Van Wazer, 1958
), it must be that the redox partner of phosphine oxidation is a key component of phosphine toxicity. By this model, a redox catalyst directly facilitates the molecular interaction between oxygen and phosphine at the point of phosphine uptake resulting in the cogeneration of oxidized phosphine and a redox active partner. This hypothesis is supported by our observation that while hyperoxia itself is not acutely lethal under our experimental conditions, it significantly enhances phosphine-induced mortality in wild-type nematodes and, to a lesser extent, in two out of the three mutants. The fact that phosphine toxicity is not enhanced by elevated oxygen concentrations in pre-33, even at a dose of phosphine that is somewhat toxic in normal air, indicates that a unique toxicity mechanism is disrupted or that pre-33 is a null mutant in the same pathway identified by the other mutants. Genetic analysis is currently underway to clarify the relationships between the three mutations.
The most prevalent model for phosphine action invokes the misdirection of electrons from the mitochondrial electron transport chain to the production of ROS (Chaudhry, 1997). If this model were correct, a possible mechanism of resistance would be to reduce the flow of electrons through the mitochondrial electron transport chain, thereby limiting the potential for ROS production. The resulting reduction in metabolic output, however, would be expected to inhibit a wide range of developmental and physiological processes. This is indeed observed in the C. elegans mutation clk-1, which causes a reduction in mitochondrial electron transport as well as a two-fold delay in post embryonic development (Wong et al., 1995
). Because C. elegans clk-1 mutations confer an extended lifespan, as do the pre mutations, we thought it worthwhile to measure the development rate of the mutant strains. The fact that none of the pre mutations affects developmental rate makes it unlikely that the resistance displayed by the three mutants is the result of a clk-1-like mitochondrial effect.
Conclusions and future directions.
Our results are consistent with the established hypothesis that phosphine toxicity is mediated through oxidative stress. This is supported by both the synergistic enhancement of phosphine toxicity by oxygen and the coselection of a longevity phenotype in the mutants, along with phosphine resistance. Our results are also consistent with an emerging view that uptake and oxidation of phosphine are directly coupled to the initiation of oxidative stress in cells. This is an attractive alternative to the hypothesis that phosphine disrupts mitochondrial function, resulting in the indirect generation of reactive oxygen species that subsequently cause damage to macromolecules. Our results suggest that C. elegans is a useful model for the study of phosphine toxicology. The availability of the genome sequence, together with a range of powerful genetic tools, will allow the genes responsible for phosphine resistance in the three mutants to be identified readily. Subsequent gene-based analysis will help us to understand the phosphine resistance mechanism at the molecular level and will provide insight into the factors responsible for aging.
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
NOTES |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Banjaj, R., and Wasir, H. S. (1988). Epidemic aluminium phosphide poisoning in northern India. Lancet 1, 820821.[CrossRef][Medline]
Bolter, C. J., and Chefurka, W. (1990). The effect of phosphine treatment on superoxide dismutase, catalase, and peroxidase in the granary weevil, Sitophilus granarius. Pestic. Biochem. Physiol. 36, 5260.[CrossRef][ISI]
Bond, E. J., and Monor, H. A. U. (1967). The role of oxygen in the toxicity of fumigant to insects. J. Stored Prod. Res. 3, 295310.[CrossRef]
Chaudhry, M. Q. (1997). A review of the mechanisms involved in the action of phosphine as an insecticide and phosphine resistance in stored-product insects. Pestic. Sci. 49, 213228.[CrossRef][ISI]
Chaudhry, M. Q., and Price, N. R. (1990). A spectral study of the biochemical reactions of phosphine with various haemproteins. Pestic. Biochem. Physiol. 36, 1421.[CrossRef][ISI]
Chaudhry, M. Q., and Price, N. R. (1991). Comparative toxicity of arsine and stibine to susceptible and phosphine-resistant strains of Rhyzopertha dominica (F.) (Coleoptera: Bostrychidae). Comp. Biochem. Physiol. C Comp. Pharmacol. Toxicol. 99, 4145.[CrossRef]
Daglish, G. J., Collins, P. J., Pavic, H., and Kopittke, R. A. (2002). Effects of time and concentration on mortality of phosphine-resistant Sitophilus oryzae fumigated with phosphine. Pest Manag. Sci. 58, 17.[CrossRef][ISI]
Du Bois, K. P., and Geiling, E. M. K. (1959). Text Book of Toxicology. Oxford University Press, New York.
Finkel, T., and Holbrook, N. J. (2000). Oxidants, oxidative stress, and the biology of ageing. Nature 408, 239247.[CrossRef][ISI][Medline]
Fluck, E. (1973). The chemistry of phosphine. In Topics in Current Chemistry (A. Davison, M. J. S. Dewar, K. Hafner, E. Heilbronner, U. Hofmann, K. Niedenzu, K. L. Schafer, and G. Wittig, Eds.), Vol. 35, pp. 164. Springer-Verlag, Berlin.
Genstat (2002). Genstat for Windows. VSN International, Ltd., Oxford.
Honda, S., Ishii, N., Suzuki, K., and Matsuo, M. (1993). Oxygen-dependent perturbation of life span and aging rate in the nematode. J. Gerontol. 48, B5761.[ISI][Medline]
Hsu, C., Han, B., Liu, M., Yeh, C., and Casida, J. E. (2000). Phosphine-induced oxidative damage in rats: Attenuation by melatonin. Free Radic. Biol. Med. 28, 636642.[CrossRef][ISI][Medline]
Hsu, C. H., Quistad, G. B., and Casida, J. E. (1998). Phosphine-induced oxidative stress in Hepa 1c1c7 cells. Toxicol. Sci. 46, 204210.[Abstract]
Kashi, K. P. (1981). Response of five species of stored-product insects to phosphine in oxygen-deficient atmospheres: Wheat flour. Pestic. Sci. 12, 111115.[ISI]
Kashi, K. P., and Chefurka, W. (1976). The effect of phosphine (grain fumigant) on the absorption and circular dichroic spectra of cytochrome c and cytochrome oxidase. Pestic. Biochem. Physiol. 6, 350362.[ISI]
Lam, W. W., Toia, R. F., and Casida, J. E. (1991). Oxidatively initiated phosphorylation reactions of phosphine. J. Agric. Food Chem. 39, 22742278.[ISI]
Lessenger, J. E. (1999). Case report: Multiple system illness in a woman exposed to aluminum phosphide. J. Agromed. 6, 2531.
Lithgow, G. J., and Walker, G. A. (2002). Stress resistance as a determinate of C. elegans lifespan. Mech. Ageing Dev. 123, 765771.[CrossRef][ISI][Medline]
Pratt, J. S. (in press). A new measure of uptake: Desorption of unreacted phosphine from susceptible and resistant strain of Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae). J. Stored Prod. Res.
Price, N. R. (1980). The effect of phosphine on respiration and mitochondrial oxidation in susceptible and resistant strains of Rhyzopertha dominica. Insect Biochem. 10, 6571.[CrossRef][ISI]
Quistad, G. B., Sparks, S. E., and Casida, J. E. (2000). Chemical model for phosphine-induced lipid peroxidation. Pest Manage. Sci. 56, 779783.[CrossRef][ISI]
Robinson, J. R., and Bond, E. J. (1970). The toxic action of phosphinestudies with 32PH3; terminal residues in biological materials. J. Stored Prod. Res. 6, 133146.[CrossRef][ISI]
Schlipalius, D. I., Cheng, Q., Reilly, P. E., Collins, P. J., and Ebert, P. R. (2002). Genetic linkage analysis of the lesser grain borer Rhyzopertha dominica identifies two loci that confer high-level resistance to the fumigant phosphine. Genetics 161, 773782.
Sokal, R. R., and Rohlf, F. J. (1981). Biometry: The Priciples and Practice of Statistics in Biological Research. W. H. Freeman, San Francisco.
Thomas, W. B. (1996). Methyl bromide: Effective pest management tool and environmental threat. J. Nematol. 28, 586589.[ISI]
Van Wazer, J. R. (1958). Phosphorus and its compounds. In Chemistry, Vol. 1, pp. 187367. Interscience, New York.
Wong, A., Boutis, P., and Hekimi, S. (1995). Mutations in the clk-1 gene of Caenorhabditis elegans affect developmental and behavioral timing. Genetics 139, 12471259.
|