Glutathione depletion associated with rose bengal-photosensitized mortality in the housefly Musca domestica.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 5:245-254 (1987) Glutathione Depletion Associated With Rose Bengal-Photosensitized Mortality in the Housefly, Musca domestica John M. Wages, Jr. and James R. Heitz Deparfment of Biochemistry, Mississippi State University, Mississippi State, Mississippi Glutathione, pyridine nucleotides, and lipid peroxides were measured in adult houseflies following various regimens of dye treatment and light exposure. Comparisons were made between dark control and light control flies to judge the effect of light exposure alone; between dark control and dark, dye-treated flies to evaluate the effects of dye-feeding in the dark; and between dark, dye-treated and light, dye-treated flies t o measure the effect of photodynamic action. N o significant effect was observed in levels of NADC, NADH, or NADPC. However, a decrease (-16.7%) in NADPH during photodynamic treatment was measured. Relatively small inductions of glutathione were observed in light controls and dark, dye-treated flies. Depletion of both CSH and total glutathione (the sum of GSH and GSSG, expressed as GSH equivalents) occurred in light, dye-treated flies as compared t o dark, dye-treated flies. Depletion of NADPH, when related t o GSH depletion, suggested that GSH is being utilized to conjugate some products of photooxidation or that it is being directly oxidized t o GSSG. However, the observation of a reduction in total glutathione also suggests that a fraction of GSH is being either oxidized t o a product other than GSSG or irreversibly conjugated. No significant effects from photodynamic treatment o n peroxidative potential or lipid hydroperoxides were observed. Key words: lipid hydroperoxidation, photodynamic action, pyridine nucleotides INTRODUCTION Halogenated derivatives of fluorescein (eg, rose bengal and erythrosin B) have been shown to elicit mortality in a wide range of insect species [l]. Acknowledgments: This research was supported by funds from the Mississippi Agricultural and Forestry Experiment Station (MAFES). The authors are indebted to Dr. Fay Hagan of the Department of Experimental Statistics, MAFES, for her assistance in performing the statistical analysis and to Drs. Marvin Salin, Maurice Kennedy, and Sonny Ramaswamy for their critical review of the manuscript. This is MAFES publication No. 6350. Received July 10,1986; accepted March 16,1987. Address reprint requests to Dr. James R. Heitz, Department of Biochemistry, Mississippi State University, P. 0. Drawer BB, Mississippi State, MS 39762. 01987 Alan R. Liss, Inc. 246 Wages and Heitz Molecular mechanisms of photosensitization have been extensively reviewed [2,3]. Xanthene dyes are believed to generate singlet oxygen  and also to be capable of free radical-type reactions . Although the literature abounds with toxicological  and photochemical [7-91 investigations, relatively few studies have been performed on intact organisms to determine the mechanism of photodynamic action. Previously, Callaham et al.UO] observed that newly emerged boll weevils (Anthonornus grundis B.), when fed a diet containing rose bengal, exhibited no net weight gain or net protein synthesis, whereas total lipid levels decreased by greater than 50%. These results were interpreted as showing metabolic energy depletion in the insect. Perturbation of the cellular energy charge as reflected in the balance between the oxidized and reduced forms of NADf and NADP+ could result in such an energy stress. The tripeptide GSH is used in vivo as a cellular reservoir of cysteine, as a radiation-protective antioxidant, in maintenance of enzymic thiol groups, and in the detoxification of pesticides, which contributes to development of resistance [ll]. Both free cysteine and the cysteinyl residue in GSH have been shown to be directly photooxidized in vitro by various dyes [12,13]. In addition, oxidation of sulfhydryl moieties of enzymes might be reflected in GSH levels as a measure of thio1:disulfide status. Such an effect on enzymes could be disastrous to the insect, decreasing the efficiencies of metabolic pathways. Photosensitized hydroperoxidation of membrane lipids was suggested to be a major biological effect of singlet oxygen by experiments with photooxidation of erythrocyte [14,15] and chloroplast  membrane lipids. Hydroperoxidation of membrane lipids in midgut epithelial cells would be expected to lead to increased fragility and eventual lysis, which has been observed in insects [17,18]. GSH, pyridine nucleotides, and unsaturated lipids were deemed likely targets for the reactive products of photosensitization. The work reported here was undertaken to determine whether photodynamic action affects levels or redox states of these potential targets. MATERIALS AND METHODS The houseflies (Muscu dornesticu L.) used in this study were a susceptible strain maintained in the Department of Biochemistry, Mississippi State University . Adult insects were not sexed but were used in the proportions of males to females as emerged. Photodynamic treatments were performed on flies less than 24 h posteclosion. Photodynamic treatment was by exposure to General Electric 40 W Cool White fluorescent lights, whose spectral output  coincides with the absorption spectrum of rose bengal. The insects, in cylindrical, screen-mesh tubes, were surrounded by a vertical array of six fluorescent lamps. The dimensions of each exposure tube were 34.0 cm in height and 8.9 cm in diameter. One such exposure tube was constructed for each of four treatment Photodynamically-Induced Clutathione Depletion 247 groups: DT,* DC, LC, and LT. The construction of the tube for LT contained a clear glass funnel taped to the bottom to facilitate collection of moribund insects. The tubes containing LC and LT were placed in the center of the array of fluorescent lights. The dark treatment groups were kept in the dark in the same room in tubes similar to those used for LC. Four experimental groups were designated as noted above. A population of insects was collected and separated into two treatment groups. One, a control, was fed 10% sucrose in deionized water. The second group was fed 1mM rose bengal in 10% sucrose in deionized water. Both were allowed to feed undisturbed in the dark for 24 h. At the end of this feeding time, each of the two populations was divided into two subpopulations to yield the four treatment groups. The DC and DT were kept in the dark for the same time period for which the LT and LC were maintained under the lights. The LT flies that dropped into the funnel during the first half hour were discarded. As the moribund LT flies dropped into the funnel (during this time, only rarely did any mortality occur in DT, DC, or LC), they were collected by removing a cotton plug in the stem of the funnel, allowing the insects to fall into liquid N2. The time of light exposure varied between 90 and 120 min, the time required for a sufficiently large sample of moribund flies to be collected. At the end of this time, LC, DC, and DT flies were collected by immersion of their respective tubes in a canister of liquid N2. The insects were stored at liquid N2 temperature until samples were weighed for homogenization. Whole flies from each treatment group were ground with a mortar and pestle precooled with liquid N2. Samples were weighed into glass tissue grinder tubes. Extraction in the appropriate medium was by motor-driven grinding with a ground glass pestle. Rose bengal (88% pure color) was obtained as the sodium salt (HiltonDavis Chemical Company, Cincinnati, OH). All other chemicals were of reagent grade or better and were purchased from standard sources. All absorbance measurements were made using a Gilford 2600 spectrophotometer. Fluorescence measurements were carried out on a Perkin-Elmer W F 44B fluorescence spectrophotometer. Statistical treatment of the data was by analysis of variance for a split-plot experimental design. Two treatments (control or dye-fed) were used, with two factors (light or dark) within each; the means from each experiment were used as replicates within each of the four resulting treatment groups. Probability values (p) were calculated for variation within treatments, factors (light and dark), and replicates, as well as for the interaction between treatments and factors, using the Statistical Analysis System (SAS Institute, Inc., Cary, NC) on a Data General R/Iv/6000 computer. LSDs were also calculated, and probabilities were obtained for differences between the means. *Abbreviations: DC = nonlight-exposed, nondye-treated controls; DT = nonlight-exposed, dye-treated flies; DTNB = 5,5'-dithiobis (Znitrobenzoic acid); HPLC = high-performance liquid chrornatrography; LC = light-exposed, nondye-treated controls; LSD = least significant difference; LT = light-exposed, dye-treated flies; mBBr = rnonobromobimane; MTT = 3-(4,5-dirnethylthiazolyl-2)-2,5-diphenyltetrazolium bromide; PES = phenazine ethosulfate; TBA = thiobarbituric acid. 248 Wages and Heitz Three a priori comparisons were made, on the basis of possible competing induction and depletion, using these LSD values. These comparisons were suggested by consideration that the final four treatment groups were treated separately at two levels, those of dye-feeding for 24 h and those of light exposure for approximately 2 h. This allowed more opportunity for ambient conditions to increase experimental variation between, for example, LC and LT than between DT and LT. Any difference between LC and LT may have been generated over a time course of about 26 h; DT and LT have diverged, however, over only 2 h. Any difference between DT and LT is thus potentially more important as a consequence of photodynamic action. Similar considerations prompted comparisons between DC and LC to see the effect of light exposure alone, and between DC and DT to see the effect of dyefeeding alone, as well as between DT and LT to observe the effects accompanying photodynamic action. NADH and NADPH were extracted by homogenization in boiling 0.1 N NaOH; NADf and NADPf were extracted in boiling 0.1 N HCl. The resulting extracts were then assayed by the continuous enzyme cycling microassay of Matsumura and Miyachi . Percent recoveries of NADPf and NADH added to acidic or basic extracts, respectively, of both dye-fed and control insects, were reproducibly near 100%. This demonstrated that neither rose bengal nor an endogenous inhibitor was able to affect the assay. For assay of total glutathione and thiol, extraction was by homogenization in 20 mM EDTA. The extracts were kept at 100°C. for 3 min to denature proteins. Centrifugation at 25,0009 for 5 min was sufficient to sediment precipitated material. Total thiols were estimated using Ellman’s reagent . Total glutathione was measured in the same extract using the cycling assay of Owens and Belcher . GSH, in abdomens from which the chitinous exoskeleton had been removed, was measured, using mBBr derivatization and HPLC separation as described by Fahey and colleagues [24,25]. Standards were prepared by derivatization of GSH, GSSG, Cys, cysteic acid, and CoASH, as described by Fahey et al. . HPLC separations were carried out using two chromatography pumps from Waters Associates (Milford, MA) equipped with an automatic injector (WISP 710A, Waters), a Perkin-Elmer LS-4 fluorescence spectrometer as detector, a data module (Waters Assoc.), and a programmable system controller (Waters Assoc.). An Anspec Ultrasphere ODS (5 pm, C-18) column was employed at ambient temperature and a flow rate of 1.5 mllmin, utilizing the buffer system described by Newton et al. . The GSH, Cys, and CoASH derivatives eluted at 11.0 f 0.5 min, 7.0 k 0.5 min, and approximately 30 min, respectively. Cys peaks were typically much smaller than those for GSH, and many were not accurately quantifiable. These data were omitted from analyses. CoASH peaks were barely detectable. No attempt was made to quantitate CoASH. Experiments were designed to ascertain whether induction of total glutathione occurred in LC and DT. In the experiment to measure total glutathione induction in LC, unsexed flies were fed ad libitum on 10% sucrose for 24 h Photodynamically-Induced Glutathione Depletion 249 in the dark. Samples were collected as a function of illumination time. Extracts were assayed for thiol and total glutathione as described above. In the experiment designed to measure GSH induction in DT, two populations of insects were fed ad libitum on either 1mM rose bengal in 10% sucrose or 10% sucrose alone in the dark. Samples were collected as a function of feeding time and were assayed for GSH by derivatization with mBBr and HPLC analysis as previously described. S-n-Butyl homocysteine sulfoximine (buthionine sulfoximine), an inhibitor of glutathione synthesis , was employed to deplete glutathione in a population of insects to determine if glutathione depletion is a factor in inducing mortality or in enhancing the phototoxicity of rose bengal. One population was fed ad libitum on 10 mM buthionine sulfoximine in 10% sucrose for 24 h in the dark; a second group was fed on 10% sucrose alone under the same conditions. Total glutathione was estimated in these two treatment groups as previously described. Samples of the remaining flies were separated into clear plastic cups covered with screen wire. Each cup contained 20 flies, with five cups per treatment. Four treatments were designated as follows: flies fed 10% sucrose for 24 h, then 10% sucrose for a second 24 h; flies fed 10% sucrose for 24 h, then 50 pM rose bengal in 10% sucrose for 24 h; flies fed 1 O m M buthionine sulfoximine for 24 h, then 10% sucrose for 24 h; and flies fed 10 mM buthionine sulfoximine for 24 h, then 50 pM rose bengal in 10% sucrose for 24 h. At the end of the feeding period, all the cups were placed under horizontal arrays of the fluorescent lights previously described, and mortality was determined as a function of illumination time. Mortality was compared between treatment groups to ascertain whether any enhancement of toxicity by total glutathione depletion had occurred. To measure lipid peroxidative potential, the TBA method adapted by Sohal et al.  to measure lipid peroxidation in insect lipid extracts was used. Recently, a fluorescence method  was described for direct determination of hydroperoxides. This method was used to determine whether photodynamic treatment produced peroxidation products not readily detectable by the TBA method. Extraction in ice-cold methanol, homogenization, and reaction with dichlorofluorescin were carried out under an N2 atmosphere in a polyethylene glove bag. Hematin dichlorofluorescin reagent was prepared as described by Cathcart et al. . For assay, 2.9 ml of the reagent was mixed with 200 pl of the extract in a glass test tube, which was sealed and incubated at 50°C. for 50 min. Fluorescence of dichlorofluorescein was read at an emission wavelength of 523 nm and an excitation wavelength of 503 nm. The relative fluorescence in each reaction mixture was divided by the wet weight of the tissue sample to obtain relative fluorescence per gram. This value was employed in statistical analyses for comparison without extrapolation to actual lipid hydroperoxide concentrations. RESULTS Data from NADP+ and NADPH assays are given in Table 1.No significant difference in NAD+ or NADH levels was observed (data not shown). Significance (p < 0.05) is found only in the difference between DT and LT with 250 Wages and Heitz TABLE 1. Effect of Rose Bengal-Photosensitized Oxidation on NADP+ and NADPH Levels in the Houseflv* Coenzyme treatment NADP' Controls Dark (DC) Light (LC) Treated Dark (DT) Light (LT) NADPH Controls Dark (DC) Light (LC) Treated Dark (DT) Light (LT) Coenzyme levels (nmollg wet weight; mean f SD) Statistical comparisons (probability) DC-+LC DC-+DT DT-+LT 0.2804 0.0799 0.1542 0.5075 0.3820 0.0085 12.4 f 2.1 11.5 f 1.5 10.8 f 0.7 12.1 f 1.6 15.4 f 2.1 14.9 rt 1.4 16.1 f 2.6 13.4 zt 1.2 *N = five separate experiments. One homogenate was prepared per treatment in each of five experiments on 5 different days. Each homogenate was assayed once. TABLE 2. Effect of Rose Bengal-Photosensitized Oxidation on Total Glutathione (GSH + GSSG) and Total Nonprotein Thiol Levels in the Housefly* Thiol treatment group Total glutathione Controls Dark (DC) Light (LC) Treated Dark (DT) Light (LT) Total thiol Controls Dark (DC) Light (LC) Treated Dark (DT) Light (LT) Glutathione and thiol levels (nmollg wet weight; mean f SE) Statistical comparisons (probability) DC-+LC DC-+DT DT-+LT 0.0197 0.0043 0.0003 0.2627 0.0506 0.0007 1,290 f 14 1,350 f 31 * 1,380 47 1,190 f 42 998 f 137 1,030 f 148 1,060 f 138 858 f 143 *N = Three separate experiments. Three homogenates per treatment were prepared in each of three experiments on 3 different days. Each homogenate was assayed in triplicate. respect to NADPH levels. The mean reduction in NADPH levels from DT to LT is 16.7%. No concomitant, significant increase in NADP+ levels is observed, although it should be noted that the mean level in NADP' in LT is 11.6%greater (p < 0.15) than that in DT. Total glutathione and total thiol were assayed (Table 2). Discrepancy between these values for total glutathione in the housefly and those obtained by Saleh et al.  is probably due to the different extraction and assay Photodynamically-Induced Glutathione Depletion 251 methods. Comparisons between treatment groups show significant differences in total glutathione between DC and LC, between DC and DT, and between DT and LT and in total nonprotein thiol between DC and DT, and between DT and LT. Mean levels of GSH and Cys as determined by mBBr derivatization are listed in Table 3. GSH titers, determined by this method, are lower than those determined by the enzyme cycling method. Since care was taken to maintain samples at liquid N2 temperature and to derivatize immediately, it is doubtful that this discrepancy can be explained by oxidation of GSH to GSSG. Another possible explanation is that extraction was less complete in mBBr reagent than in boiling EDTA. A significant difference in GSH was found between DT and LT. Differences in Cys content between DC and LC, and between DC and DT, were also measured. The magnitude of depletion of GSH in the LT from the levels in DT was approximately 55.7%. Exposure of DC to light for varying time periods resulted in a general, though small, increase (- 4.7%) in both total glutathione and thiol over time (data not shown). The results from the treatment with rose bengal in the dark indicated a generally higher total glutathione level ( 5-50% higher) in DT than in DC (data not given). A reduction of total glutathione to 19.5% of the control value was effected by feeding 10 mM buthionine sulfoximine in the dark for 24 h. No effect on mortality from total glutathione depletion was observed. No statistically significant effects were found from photodynamic treatment by either the TBA assay or the dichlorofluoresceinassay (data not given). - TABLE 3. Effect of Rose Bengal-Photosensitized Oxidation on Reduced Glutathione and Cysteine Levels in the Housefly* Thiol treatment group GSH Controls Dark (DC) Light (LC) Treated Dark (DT) Light (LT) CYS Controls Dark (DC) Light (LC) Treated Dark (DT) Light (LT) Levels of GSH and Cys (nmolig wet weight; mean f SE) Statistical comparisons (probability) DC+LC DC-tDT DT-tLT 0.5448 0.4377 0.0344 0.0008 0.0010 0.2379 825 & 11 711 f 130 973 f 309 431 f 103 210 f 32 105 +_ 21 110 k 16 94 f 18 *N = Three experiments. At least three homogenates were prepared per treatment. Each homogenate was assayed at least two times, in each of three experiments on 3 different days, Replication of cysteine assays was variable; data consisting of HPLC peaks that could not be accurately measured were discarded. 252 Wages and Heitz DISCUSSION Lipid peroxidation has been observed to be associated with GSH depletion [31,32]. Failure to observe an effect on lipid peroxidation by the TBA method implies that photodynamic action leaves the oxidation state of housefly lipids relatively untouched. That no effect is observed by either the TBA assay or the dichlorofluorescein method suggests that photodynamic mortality in the housefly does not involve lipid hydroperoxidation, which is well documented as an effect of in vitro xanthene dye-photosensitized oxidation. The mass of literature on photooxidation of unsaturated fatty acids and cholesterol makes it apparent that some hydroperoxidation of membrane lipids probably occurs. Failure to detect it in these experiments could be due to one of the following: 1)Metabolism of lipid hydroperoxides rapidly follows their formation such that only a negligible fraction remains until assay. 2) The methods are insufficiently sensitive to detect hydroperoxidation of some crucial lipid components, which represent only a small fraction of the total lipid. 3) The insects expired from some other lesion before initial antioxidation barriers could be breached and detectable lipid hydroperoxidation could occur. Yu et al.  showed that photooxidation of histidine residues led to inactivation of membrane Ca2+ transport proteins before detectable lipid hydroperoxidation occurred. A significant GSH depletion was observed as a result of in vivo photodynamic action. The pool of total glutathione was also depleted, implying either that GSH synthesis was inhibited or that GSH itself was oxidized to a product other than the disulfide. NADPH depletion in the photodynamically treated insects suggested oxidation of some amount of GSH to GSSG. Assuming a relation between depletion of NADPH and GSH, at least some of the GSH is being oxidized to GSSG, since the major enzyme involved in maintaining the equilibrium between GSH and GSSG, GSSG reductase, utilized NADPH as a donor of electrons for the reduction of GSSG. Irreversible conjugation of GSH with electrophilic products of photooxidation could bring about the observed decrease in total glutathione. As a complete hypothesis, however, this is not consistent with the NADPH data. Lysis of midgut epithelial cells  would bring intracellular GSH and the dye into proximity in the lumen of the alimentary canal. Here, GSH would be expected to be photooxidized at a rate similar to that observed in vitro. Potentially, two separate mechanisms are operating under photodynamic conditions: an oxidation of GSH to GSSG by oxidative detoxification mechanisms, which explains the depletion in NADPH levels, and an oxidation of GSH to a product other than GSSG, possibly to the sulfonic acid or to an irreversible conjugate. A minor induction of total glutathione occurred within 2 h of light exposure. Given the general radioprotectant role of GSH, this is not surprising. Induction of GSH also occurred with dye-feeding, which may be related to the known role of GSH in detoxification of xenobiotics. That a general GSH deficiency cannot account for observed mortality data is shown by the inability of buthionine sulfoximine-mediated depletion of total glutathione either to kill the insects or to synergize dye-sensitized Photodynamically-InducedGlutathione Depletion 253 killing. The possibility remains that GSH depletion is occurring in response to photooxidation of vital systems. In this case, depletion that would seem insignificant in the context of normal experimental variation might, in fact, be fatal. Unless localized depletion occurs, it appears that GSH depletion is incident to, rather than the cause of, photodynamic toxicity. LITERATURE CITED 1. Robinson JR: Photodynamic insecticides: A review of studies on phtosensitizing dyes as insect control agents, their practical application, hazards, and residues. Residue Rev, 88, 69 (1983). 2. Spikes JD, Livingston R: The molecular biology of photodynamic action: Sensitized photoautoxidations in biological systems. Adv Radiat Biol3, 29 (1969). 3. Jori G, Spikes JD: Photosensitized oxidations in complex biological structures. In: Oxygen and Oxy-Radicals in Chemistry and Biology. Rodgers MAJ, Powers EL, eds. Academic Press, New York, pp 441-459 (1981). 4. Midden WR, Wang SY: Singlet oxygen for solution kinetics: clean and simple. J Am Chem SOC105, 4129 (1983). 5. Jefford CW, Boschung AG: The reaction of biadamantylidine with singlet oxygen in the presence of dyes. Helv Chim Acta 60, 2673 (1977). 6. Heitz JR: Xanthene dyes as pesticides. In: Insecticide Mode of Action. Coats J, ed. Academic Press, New York, pp 429-457 (1982). 7. Doleiden FH, Fahrenholtz SR, Lamola AA, Trozzolo AM: Reactivity of cholesterol and some fatty acids toward singlet oxygen. Photochem Photobiol20, 519 (1974). 8. Bellin JS, Yankus CA: Influence of dye binding on the sensitized photooxidation of amino acids. Arch Biochem Biophys 223, 18 (1968). 9. Nieumint AWM, Aubry JM, Arwert F, Kortbeek H, Herzberg S, Joenje H: Inability of chemically generated singlet oxygen to break the DNA backbone. Free Rad Res Commun 2, 1(1985). 10. Callaham ME, Broome JE, Poe WE, Heitz JR: Time dependence of light-independent biochemical changes in the boll weevil, Anthonornus gmndis, caused by dietary rose bengal. Environ Entomol6, 669 (1977). 11. Yang RSH: Enzymatic conjugation and insecticide metabolism. In: Insecticide Biochemistry and Physiology. Wilkinson CF, ed. Plenum Press, New York, pp 177-225 (1976). 12. Sone K, Koyanagi T: Photooxidation of glutathione sensitized by the riboflavin. Vitamins (Japan) 33, 349 (1966). 13. Spikes JD, Livingston R: Photosensitization. In: The Science of Photobiology. Smith KC, ed. Plenum Press, New York, pp 87-112 (1977). 14. Lamola AA, Yamane T, Trozzolo AM: Cholesterol hydroperoxide formation in red cell membranes and photohemolysis in erythropoieticprotoporphyria. Science 179, 1131(1973). 15. Valenzeno DP: Photohemolytic lesions: Stoichiometry of creation by phloxine B. Photochem Photobiol40, 681 (1984). 16. Percival MP, Dodge AD: Photodynamic damage to chloroplast membranes, photosensitized oxidation of chloroplast acyl lipid. Plant Sci Lett 29, 255 (1983). 17. Schildmacher H: Concerning photosensitizationof mosquito larvae with fluorescent dyes (in German). Biol Zentralbl69, 468 (1950). 18. Yoho TP: The photodynamic effect of light on dye-fed houseflies, Musca dornestica L. PhD dissertation, West Virginia University, p 49 (1972). 19. Respicio NC, Heitz JR: Comparative toxicity of rhodamine B and rhodamine 6G to the housefly (Musca dornestica L.). Bull Environ Contam Toxic01 27, 274 (1981). 20. Iscotables: A Handbook of Data for Biological and Physical Scientists, 3rd ed. Instrumentation Specialties Company, Lincoln, NE, p 33 (1970). 21. Matsumura H, Miyachi S: Cycling assay for nicotinamide adenine dinucleotides. Meth Enzymol69, 465 (1980). 22. Ellman GL: Tissue sulfhydryl groups. Arch Biochem Biophys 82, 70 (1959). 254 Wages and Heitz 23. Owens CWI, Belcher RV: A colorimetric micro-method for the determination of glutathione. Biochem J 94, 705 (1965). 24. Fahey RC, Newton GL: Occurrence of low molecular weight thiols in biological systems. In: Functions of Glutathione: Biochemical, Physiological, Toxicological, and Clinical Aspects. Larsson A, Orrenius s, Holmgren A, Mannervik 0, eds. Raven Press, New York, pp 251-260 (1983). 25. Fahey RC, Newton GL, Dorian R, Kosower EM: Analysis of biological thiols: quantitative determination of thiols at the picomole level based upon derivatization with monobromobimanes and separation by cation-exchange chromatography. Anal Biochem 111, 357 (1981). 26. Newton GL, Dorian R, Fahey RC: Analysis of biological thiols: Derivatization with monobromobimane and separation by reverse-phase high-performance liquid chromatography. Anal Biochem 114, 383 (1981). 27. Griffith OW, Meister A: Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (s-n-butyl homocysteine sulfoximine). J Biol Chem 254, 7558 (1979). 28. Sohal RS, Donato H, Biehl ER: Effect of age and metabolic rate on lipid peroxidation in the housefly Muscu domesticu L. Mech Aging Dev 26, 159 (1981). 29. Cathcart R, Schwiers E, Ames BN: Detection of picomole levels of lipid hydroperoxides using a dichlorofluorescein fluorescent assay. Meth Enzymol 105, 352 (1984). 30. Saleh Ma, Motoyama N, Dauterman WC: Reduced glutathione in the housefly: Concentration during development and variation in strains. Insect Biochem 8, 311 (1978). 31. Hogberg J, Orrenius S, Larson RE: Lipid peroxidation in isolated hepatocytes. Eur J Biochem 50, 595 (1975). 32. Reiter R, Wendel A: Chemically-induced glutathione depletion and lipid peroxidation. Chem Biol Interact 40, 365 (1982). 33. Yu BP, Masoro EJ, Bertrand HA: The functioning of histidine residues of sarcoplasmic reticulum in Ca2+transyurt and related activities. Biochemistry 13, 5083 (1974).