Heme Peroxidase Activity Measured in Single Mosquitoes Identifies Individuals Expressing an Elevated Oxidase for Insecticide Resistance

Home , Peroxidase

Journal of the American Mosquito Control Association, 13(3):233-237, 1997 Copyright @ 1997 by the American Mosquito Control Association, Inc.

HEME PEROXIDASE ACTIVITY MEASURED IN SINGLE MOSQUITOES IDENTIFIES INDIVIDUALS EXPRESSING AN ELEVATED OXIDASE FOR INSECTICIDE RESISTANCE

WILLIAM G. BROGDON,' JANET C. McALLISTERI rNr JOHN VULULE?

ABSTRACT. Optimum conditions are described for a simple, rapid, microplate-based assay that indirectly measures the differences in oxidase levels between individual susceptible, resistant, or induced mosquitoes. A small proportion (0.01-0.1) of a single mosquito is used, allowing multiple replicates of the oxidase assay. Cytochrome C is used as a positive control. The levels of oxidase found in sample populations of pyrethroid- susceptible, pyrethroid-resistant, and phenobarbital-indtced Anopheles albimanus mosquitoes are characterized with the assay.

INTRODUCTION al. 1977); aldrin epoxidation to dieldrin (Yu et al. 1971, Feyereisen 1983); or aryl hydrocarbon hy- Xenobiotic detoxification in organisms is accom- droxylase (Dehnen et al. 1973). All of these meth- plished by families of enzymes that hydrolyze, ox- ods suffer from the drawback that pooled micro- idize, and conjugate injurious compounds into less somal fractions from a number of insects must be toxic, more water-soluble products that can be used; none of these methods can be used to mea- readily excreted. Among the most important detox- sure differences in oxidase levels in single mos- iflcation enzymes are the mixed-function, or P450, quitoes. oxidases (also termed oxygenases). These enzymes An indirect alternative is to measure the level of are present throughout the tissues of plants and an- heme-containing enzymes (including the cyto- imals and catalyze an enorrnous range of biochem- chrome oxidase enzymes) in single insects. The ical reactions essential to life (Agosin 1985, Hodg- levels of these enzymes should be correlated with son 1985). Our particular interest in oxidase en- the peroxidase activity of the heme group. Such a zymes involves detoxification of insecticides by technique would provide a useful means for mea- disease vectors. In biochemical detection of insec- suring large-scale differences in oxidase levels ticide resistance (Brogdon 1989), it is necessary characteristic of insecticide resistance and oxidase that enzyme levels be determined in single mos- induction, provided that the heme peroxidase levels quitoes, because populations often consist of are low but measurable in susceptible, noninduced mixtures of susceptible and resistant individuals. mosquitoes. Methods have been developed for the measurement The present study had 3 objectives. The first was of levels of esterases and glutathione s-transferases to develop a method for measuring basal levels of (Brogdon and Dickinson 1983, Brogdon and Barber heme peroxidase activity, thus, hemoprotein levels, l99oa) in single mosquitoes. A simple method for in susceptible populations of anopheline mosqui- detecting elevated levels of oxidase in single insects toes. Second, sample populations were compared has not been available. using the assay to determine if enzyme levels are There have been 2 approaches to measurement enhanced by induction or by the presence of insec- and study of oxidase activity in insects (Agosin ticide resistance. Third, reliability of the technique 1976). The first approach has been to evaluate spec- was evaluated through analysis of within-run and tral and catalytic properties of oxidases associated between-run variability of results. The means for with the membrane, using techniques such as the running a standard curve of enzyme activity and an determination of carbon monoxide inhibition of en- internal standard were also detailed. The standard zyme spectral characteristics (Omura and Sato curve was designed to show the linearity of the 1964). The second approach has been to attempt heme peroxidase assay with hemoprotein concen- the solubilization, resolution, purification and re- tration. Cytochrome C was used as an internal stan- constitution of the enzyme system. This has re- dard for each microplate. Cytochrome C was cho- quired measurement using specific substrates of sen because of its wide availability and low cost. known oxidase enzyme functions, such as: O-de- This method is intended to provide a simple means alkylation using 7-methoxy-4-methylcoumarin of quantifying oxidase activity in mixed popula- (Feyereisen and Vincent 1984), 7-ethoxycoumarin tions of resistant and susceptible mosquitoes. (Patil et al. 1990), or methoxyresorufin (Mayer et

'Entomology Branch F-22, Division of Parasitic Dis- MATERIALS AND METHODS eases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Public Health Service, Mosquitoes: Reference strains of Anopheles al- U.S. Department of Health and Human Services, Cham- bimanus Wied. used in this study are maintained at blee, GA 30341. the Centers for Disease Control and Prevention in 2 Kenya Medical Research Institute, Nairobi, Kenya. Atlanta, GA. Resistant mosquitoes were from a

233 234 JounNar- oF THE AMERTcANMosetlro CoNrRoL Assocnrrox Vor. 13,No. -? strain of An. albimanu.s from Guatemala that con_ CytochromeC tains oxidase resistance to permethrin, but not the high esterase resistance that also occurs in the 1.6 country (Brogdon and Barber 1990b). This colony is periodically selected to preserve the mechanism. 14 The association of the elevated oxidase measured in these sample populations with insecticide resis- E tance has been confirmed via a single-family isolate from the parent strain through bioassay (resistance O E ln level >100X based on dose-response), complete synergism by piperonyl butoxide, and induction ex- 3 og periments. Details of these experiments will be re- ported elsewhere. .i oo Sample preparation: Adult 3- to 4-day-old, non- bloodfed female mosquitoes were frozen and ho- mogenized in 100 pl assay buffer in plastic micro- centrifuge tubes using plastic pestles. Late 4th-in- 0.2 star larvae and pupae were transferred alive into assay buffer for homogenization. Homogenates 0.0 were diluted to I ml with additional buffer. For 0 5 10 15 20 25 30 35 bloodfed mosquitoes, the abdomen was removed before homogenization. Time (Minutes) Enzyme assays: The method for enzyme assays Fig. 1. Linearity of TMBZ substrate peroxidation for was an adaptation of that for detecting heme per- O.03 to 1.06 pg of cytochrome C. n : 8 replicates. oxidase activify on electrophoresis gels (Thomas et al. 1976) using the substrate 3,3'5,5'-tetramethylbenzidine (TMBZ). Both the free base and (PB), an oxidase inducer, that was provided over- dihydrochloride forms of TMBZ were obtained night in a l07o sucrose solution provided to nourish from Sigma Chemical Company, St. Louis, MO. the mosquito colony. Sucrose/PB was replaced with The TMBZ working solution was prepared fresh normal lOTo sucrose the next morning to prevent daily by dissolving 50 mg TMBZ free base or mortality from PB. Mosquitoes were collected 24 h dihydrochloride in 25 ml absolute methanol and after initial exposure to PB along with mosquitoes adding 75 ml O.25 M sodium acetate buffer (pH from the same sample population that had not been 5.0; pH adjusted with acetic acid). induced. Another sample population from the To 100 pl of mosquito homogenate, 200 pl of mixed resistant colony was selected at the LDro lev- '|MBZ solution were added in assay wells, fol- el with O.O25Vopermethrin-impregnated resistance lowed by 25 1t"lof 37o hydrogen peroxide (com- test papers (World Health Organization, Geneva, mercial grade). The plate was read in a microplate- Switzerland). A sample population of the F, gen- reading spectrophotometer at test wavelengths of eration from survivors was compared for oxidase 630-690 nm at time T : 0 and at intervals there- activity with the original population values for the after, depending on the purpose of the experiment. parent strain. Mosquitoes were conveniently assayed in sample sizes of 32, since 3 replicates of 32 mosquitoes fill RESULTS a single microplate. Standard curves (8 replicates) for heme peroxi- Standard cunes: The cytochrome C standard dase were prepared using dilutions of a commercial showed linear peroxidation kinetics for at least 30 preparation of cytochrome C (Sigma Chemical Co., min at concentrations below 0.5 pg €ig. l). Above St. Louis, MO). Mosquito protein was measured on that concentration, linearity was lost after 5 min. 50-pl aliquots of assay homogenates using the Therefore, initial enzyme rates measured at T : 5 Bradford assay (Bradford 1976; Brogdon 1984a, min were used for all microassays. 1984b). This was used as a general indicator of size The complete standard curve for 0.03-1.066 ng and to test for differences in size among different of cytochrome C is accurately described by an ex- populations. ponential equation ! : aeb' (Fig. 2a). However, for Population comparisons : Individual mosquitoes practical purposes, the data may be adequately de- from 4 populations were compared. A permethrin- scribed using 2 linear equations of the form y : susceptible control population was compared with mx * b (Figs. 2b--c). The first describes the curve a resistant sffain (LD.u > 100X the control popu- from 0 to l4O ng (l : 0.99), and the second de- lation) with elevated oxidase levels. A portion of a scribes the curve from 0.150 to over I pg (l : population sample of a mixed resistant colony 0.99). (ranging from susceptible individuals to highly re- Comparison of sample populations: The control sistant) was then induced with O.lVo phenobarbital and resistant sample populations were significantly Srprnagrn 1997 Itornrcr OXIDASE AssAY 235

't6 a 0.s

't4 D SusceptibleMosquitoesI m Mixed.Popul8l.ion I 0.4 Susceptiblemd I 7tz ResistantMosquit@s l Nl Single Fmily nesistant I !? o 1n Y"'19 i F o.s

.F o.z

24 0.1

0.5 0.6 0.7 0.8 0.9 1.0 1.'l 1.2 1.3 'l.4

Absorbance 629 1'''

b o.oo Fig. 3. Comparison of permethrin-susceptible and -resistant sample populations of Anopheles albimanus. For eachpopulation,n=32. 0.05

different (Mann-Whitney rank-sum test, P < 0.001) "€ o.o4 o in oxidase activity (Fig. 3). The single-family isolate from the resistant population showed a resis- level in greater than 3 0.03 tance bioassays of l00X and was completely synergized by piperonyl butoxide. The resistant strain of An. albimanus used for se- 'tP: o.o2 lection and induction experiments contains both re- o sistant and susceptible individuals, similar to a field population. The portion of the resistant population with absorbance of >0.3 (the resistance threshold) was equivalent to the portion of the population sur- viving beyond the upper range limit of a susceptible population in bioassays of the resistant strain (data 20 40 60 80 100 120 140 not shown). Cytochrome C (nanograms) The induced and control resistant mosquitoes from the same population showed significantly dif- C 1.0 ferent oxidase levels (t-test, p < 0.0O1), confirming induction by phenobarbital (Fig. 4). Exposure to permethrin selected for significantly higher oxidase 0.8 levels (Mann-Whitney rank-sum test, P < 0.001) than did the unselected portion of the sample (Fig. a 5). In these experiments, protein assays conducted

ct u.9 in triplicate on every mosquito showed that none of these populations differed significantly in the sizes of individual mosquitoes (data not shown). 04 Within-run and between-run variation' In all ex- ; periments involving individual mosquitoes, the maximum coefficient of variation (SD/i) was'7Vo a 0.2 (-

Fig. 2. Standard curves for cytochrome C. a. Cubic spline curve of range 0-660 ng. b. Linear regression of 0.0 o.2 0.4 0.6 0.E 1.0 1.20 range 0-140 ng (r, = O.994). c. Linear regression ofrange 0.150-1.066 pC (f = 0.991). In : Cytochrome C (micrograms) each case, n 8 replicates. 236 JounNal or rus AvrnrcaN Mosquno CoNrnol AssocnrroN Vol. 13. No. 3

Mixed -l Population f= MixedPopulation: Susceptibiemd Surceptiblemd ResistantMosquit@s ResistmtMosquitoes m Plre obTbitaL-iad\ced l% Pemerhin-seleted, Mixed Population z Mixed Population(Fl) q E8 '5O 1n

9B Oa z

a -o4 z4 z

0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Absorbance626 61 Absorbance626 n. Fig. 4. Comparison of permethrin-resistantand phen- Fig. 5. Comparison of permethrin-resistantand per- obarbitaf-inducedsample populations of Anopheles albi- methrin-selectedsample populations of Anopheles albimanus.For eachpopulation, n : 32. manus.For eachpopulation, n = 32. and the average was 2.5Vo. The day-to-day varia- This biochemical microassay must be correlated tion in assays was 157o. to bioassay data or known experimental variables from the same sample population for the results to be comparable to those of other sample popula- DISCUSSION tions. This is an indirect assay that would be sen- Use of heme peroxidation to alrralyzevariation in sitive to misinterpretation in the presence of an un- oxidase levels does not provide information on spe- known hemoprotein or interfering substance. Every cific oxidases. This is also true of Omura and Sato's effort must be made to minimize experimental vari- (1964) carbon monoxide method which is broadly ables or to quantify them, as in a protein microas- specific for hemoproteins (Agosin 1985). However, say run on the same sample to correct for mosquito there are a number of advantages in using a more size. general oxidase assay. First, microsomal prepara- 3,3' 5,5' -Tetramethylbenzidine is a nonmutagen- tions have low reproducibility (Agosin 1985) and ic, noncarcinogenic subsfrate (Holland et al. 1974, are not possible with single mosquitoes. Moreover, Garner 1975), commonly used in kinetic ELISA correlations of in vitro mono-oxygenases with their (Hancock and Tsang 1986). The TN.{BZ has the ad- in vivo functions are difficult to obtain due to the ditional advantage of a molar extinction coefficient multiplicity of closely related enzymes competing that is l0 'm that of other hydrogen donor sub- for the same substrate, and the broad, overlapping strates, making it possible to detect enzyme activity substrate specificities of enzymes in the family in small portions of single mosquito preps. Reading (Agosin 1985, Scott et al. 1994). In addition, it is the results at absorbances ranging from 630 to 690 rare to obtain linear reaction plots with the specific nm is simple due to the broad absorbance peak at substrates of various oxidases due to the high li- 655 nm (Bos et al. 1981). pophilicity of the substrates, multiple forms of the Although it is an indirect assay, the method de- enzyme, and the complex nature of the membrane scribed here has a number of practical applications (Hodgson 1985). In fact, enzyme rates in these re- where a difference in oxidase levels is suspected. actions depend not on the concentration of sub- The technique allows for rapid determination of re- strate, but on the absolute amount (Lewis et al. sistance frequency estimates, and the simplicity of 1967). Finally, association of a specific oxidase the method makes it potentially available to even with insecticide resistance or enzyme induction is the most financially pressured mosquito control or- extremely difficult, even with purified microsomal ganizations. enzymes (Plapp and Casida 1969). For these rea- sons, we feel that a generic assay for inferring dif- level based on hemoprotein lev- ferences in oxidase ACKNO\MLEDGMENTS els is a practical option that is broadly applicable to resistance detection or measurement of oxidase This work was supported, in part, by World enzyme induction. Health Organization TDR grant 910637. 23-l Srmsl.rern 1997 INDTRECTOxtoLst AsseY

REFERENCES CITED Garner, R. C. 19"15.Testing of some benzidine analogues for microsomal activation to bacterial mutagens. Cancer 1976. Insect cytochrome P-45O. Mol. Cell- Agosin, M. Lett. l:39-42. 12:33-44. Biochem. Hancock, K. and V. C. W. Tsang. 1986. Development and Agosin, M. 1985. Role of microsomal oxidations in in- optimization of the FAST-ELISA for detecting antibod- secticide degradation, pp. 647-73O. 1n.' G. A. Kerkut ies to Sclrislosoma mansoni. J. Immunol. Methods 92: and L. I. Gilbert (eds.). Comprehensive insect physiol- 167-176. ogy, biochemistry, and pharmacology, Volume 12. Per- Hodgson, E. 1985. Microsomal mono-oxygenases' pp. gamon Press, Oxford. 225-321. G. A. Kerkut and L. I. Gilbert (eds.). 1n.' Bos, E. S., A. A. Van der Doelen, N. Van Rooy and A. Comprehensive insect physiology, biochemistry, and H. W. M. Schuurs. 1981. 3,3',5,5'-Tetramethylbenzi- pharmacology, Volume ll. Pergamon Press, Oxford. dine as an Ames test negative chromogen for horse- R., L. F L. Rose and A. L. Wal- radish peroxidase in enzyme-immunoassay. J. Immu- Holland, V. F, Saunders, peroxidase action. 18. The oxi- noassay 2:187-2O4. pole. 1974. Studies in Tetrahedron Bradford, M. M. 1976. A rapid and sensitive method for dation of 4-chloraniline-further studies. the quantitation of microgram quantities of protein uti- 3O:3299-33O2. lizing the principle of protein-dye binding. Anal. Bio- Lewis,S. 8., C. E Wilkinsonand J. W. Ray. 1967.The chem. 72:248-254. relationship between microsomal epoxidation and lipid Brogdon, W. G. 1984a. Mosquito protein microassay-I. peroxidation in houseflies and pig liver and the inhibi- Protein determinations from small portions of single- tory effect of derivatives of 1,3-benzodioxole (methy- mosquito homogenates. Comp. Biochem. Physiol. 79B: lenedioxybenzene). Biochem. Pharmacol. 16:1 195- 457-4s9. t210. Brogdon, W. G. 1984b. Mosquito protein microassay- Mayeq R. T., J. W. Jermyn, M. D. Burke and R. A. II. Modification for potential field use. Comp. Biochem. Prough. 1977. Methoxyresorufin as a substrate for the Physiol. 798:461-464. fluorometric assay of insect microsomal o-dealkylases. Brogdon, W. G. 1989. Biochemical resistance detection: Pestic. Biochem. Physiol. 7 :349-354. an alternative to bioassay. Parasitol. Today 5:56-60. Omura, T. and R. Sato. 1964. The carbon dioxide binding Brogdon, W. G. and A. M. Barber. 199Oa. Microplate pigment of liver microsomes. I. Evidence for its hemo- assay of glutathione s-transferase activity for resistance protein nature. J. Biol. Chem. 239:2370-2378. detection in single-mosquito triturates. Comp. Biochem. Patil, T. N., R. A. Morton and R. S. Singh. 1990. Char- Physiol. 96B:339-342. acterization of 7-etoxycoumarin-O-deetylase from mal- Brogdon, W. G. and A. M. Barber. 1990b. Fenitrothion- athion resistant and susceptible strains of Drosophila deltamethrin cross-resistance conferred by esterases in melanogaster. Insect Biochem. 20:91-98. Guatemalan Anopheles albimanus. Pestic. Biochem. Plapp, E W., Jr. and J. E. Casida. 1969. Genetic control Physiol. 37:l3O-139. of housefly NADPH-dependent oxidases: relation to in- Brogdon, W. G. and C. M. Dickinson. 1983. A microas- secticide chemical metabolism and resistance. J. Econ. say system for measuring esterase activity and protein Entomof. 62:I174-l 179. concentration in small samples and in high-pressure liq- Scotr, J. A., E H. Collins and R. Feyereisen. 1994. Di- uid chromatography eluate fractions. Anal. Biochem. genes 13l:499-5O3. versity of cytochrome P450 in the mosquito, Dehnen, W., R. Tomingas and J. Roos. 1973. A modified Anopheles albimanus. Biochem. Biophys. Res. Com- method for the assay of benzo(a)pyrene hydroxylase. mtn. 2O5:1452-1459. Anal. Biochem. 53:373-383. Thomas, P 8., D. Ryan and W. Levin. 1976. An rm- Feyereisen, R. 1983. Polysubstrate monooxygenase (cy- proved staining procedure for the detection of the per- tochrome P-450) in larvae of susceptible and resistant oxidase activity of cytochrome P-450 on sodium do- strains in house flies. Pestic. Biochem. Physiol. 19:262- decyl sulfate polyacrylamide gels. Anal. Biochem. 75: 269. 168-176. Feyereisen, R. and D. R. Vincent. 1984. Characterization Yu, S. J., U. Kiigemagi and L. C. Terriere. 1971. Oxi- of antibodies to house fly NADPH-cytochrome P-450 dative metabolism of aldrin and isodrin by bean root reductase. Insect Biochem. 14:163-17 l. fraction. J. Agric. Food Chem. 19:5-9.