Journal of Chemical Ecology, Vol. 15, No. 7, 1989
CHEMICAL ECOLOGY OF THE LUNA MOTH Effects of Host Plant on Detoxification Enzyme Activity
RICHARD
L. LINDROTH
Department of Entomology University of Wisconsin 23 7 Russell Laboratories Madison, Wisconsin 53706 (Received May 13, 1988; accepted October 10, 1988)
Abstract--The effects of food plant on larval performance and midgut detoxification enzymes were investigated in larvae of the tuna moth, Aetias luna. Neonate larvae were fed leaves of black cherry, cottonwood, quaking aspen, white willow, red oak, white oak, tulip tree, paper birch, black walnut, buttemut, or shagbark hickory. First instar survival, larval duration, and pupal weights were monitored as indices of food quality. Midgut enzyme preparations from fifth instars were assayed for/3-glucosidase, quinone reductase, polysubstrate monooxygenase, esterase, and glutathione transferase activities. Larval survival on seven of the 11 plant species, including several recorded host plants, was extremely poor. Larvae performed well, and quite similarly, on birch, walnut, butternut, and hickory. Activities of all enzyme systems except /3-glucosidase were significantly influenced by larval host plant. Of the systems assayed, quinone reductase and glutathione transferase activities were especially high. Comparisons of these values with published values for other Lepidoptera support the hypothesis that these enzyme systems are involved in conferring tolerance to juglone and related quinones Occurring in members of the plant family Juglandaceae. Results suggest that host plant utilization by tuna is more specialized at the individual or population level than at the species level and that biochemical detoxification systems may play a role in such specialization. Key Words--Actias luna, Lepidoptera, Satumiidae, detoxification enzymes, enzyme induction, glutathione transferase, Juglandaceae, juglone, nutritional ecology, plant-insect interactions, quinone reductase.
2019 0098-0331/89/0700-2019506.00/0 9 1989 Plenum Publishing Corporation
2020
L~NDROTH INTRODUCTION
The luna moth, Actias luna L. (Saturniidae), occurs from southern Canada to Texas and Florida (Baker, 1972). As a species, it is moderately polyphagous, feeding on trees from at least eight plant families (Baker, 1972; Tietz, 1972). Little is known, however, of the relative suitability of various tree species as larval food plants. The research reported here was conducted in part to assess the influence of known or potential host plants on performance (survival and growth) of luna larvae. A second objective of this research was to investigate host plant alteration of allelochemical-metabolizing enzymes in luna larvae. Detoxification of plant allelochemicals by specialized enzyme systems is widely believed to be one of the most important forms of insect adaptation to plant diets. Yet few studies have documented the enzymatic detoxification capacities of tree-feeding insects, and fewer still have addressed potential host plant effects on multiple enzyme systems. These studies provided the opportunity to test several predictions about evolutionary and ecological adaptations of enzymatic detoxification systems to particular food plants. Prior to the onset of this study, I had observed that luna larvae perform well on black walnut (Juglans nigra) and shagbark hickory (Carya ovata). The most distinguishing secondary metabolites of these and other members of the Juglandaceae are juglone and related 1,4-naphthoquinones, compounds that are generally deterrent or toxic to insects (Gilbert et al., 1967; Hedin et al., 1980; Norris, 1986). The enzyme systems most likely responsible for detoxification of juglone in Juglandaceae-adapted insects include quinone reductase, which catalyzes the reduction of quinones to hydroquinones (Yu, 1987a), and glutathione transferase, which conjugates allelochemicals containing ~,/3-unsaturated carbonyls (e.g., juglone) with reduced glutathione (Wadleigh and Yu, 1987). These enzyme systems can be induced by dietary exposure to various allelochemicals. Thus I predicted that quinone reductase and glutathione transferase activities would (1) be especially high in luna larvae in comparison to published values for other polyphagous Lepidoptera, and (2) be induced in larvae fed members of the Juglandaceae, in comparison to larvae fed members of other plant families.
METHODS AND MATERIALS
Insects. Actias luna eggs were obtained from a gravid female moth captured in Laurel County, Kentucky, in May 1987. Larvae were reared in plastic boxes at ambient temperature on leaves of black walnut or shagbark hickory. Newly eclosed adult moths were paired to provide larvae for this study. To
LUNA MOTH DETOXIFICATION ACTIVITY
2021
introduce more genetic variability into the laboratory culture, an attempt was made to pair laboratory-reared females with local males by placing tethered females in nearby woodlots; these attempts proved unsuccessful. Feeding Trials. Neonate larvae were placed onto fresh leaves of 11 species of trees, including six genera on which luna are known to feed, and two genera for which no previous feeding records exist (Table 1). Each replicate consisted of 10-14 larvae, with leaves, in a ventilated plastic Petri dish (15 x 2.5 cm); leaf petioles were inserted into florist's Water Piks to maintain leaf turgor. Leaves were replaced at two- to three-day intervals. When larvae reached the fourth stadium, they were transferred to larger shoebox cages. Larvae were reared in a Percival environmental chamber at 24~ on a 12:12 light-dark cycle. Measurements were made of first instar survival, duration from hatching to spin-up of the cocoon, and pupal weight (six to seven days following spinup) as indices of food plant quality. Enzyme Preparation. Because larval midguts are the major site for detoxification of plant allelochemicals, they were used as the enzyme source. Midguts (5-9 per enzyme replicate) were dissected from fifth instar larvae (3 to 7 days old) and gut contents removed. Midguts were then washed (0.2 M phosphate buffer, pH 7.8) and homogenized by 10 strokes in a Ten Broeck tissue grinder. The homogenate was centrifuged at 10,000g (10 min) and the supernatant removed and centrifuged at 100,000g (60 min), to separate soluble (cytosolic) and microsomal (membrane-bound) enzymes. The enzyme preparations
TABLE 1. TREE SPECIES ASSAYED FOR LARVAL PERFORMANCE
Luna previously recorded to feed on the: a Species assayed
Plant family
Black cherry (Prunus serotina) Cottonwood ( Populus deltoides ) Quaking aspen ( Populus tremuloides ) White willow ( Salix alba ) Red oak (Quercus rubra) White oak (Quercus alba ) Tulip tree ( Liriodendron tulipifera ) Paper birch (Betula papyrifera ) Black walnut ( Juglans nigra ) Butternut ( Juglans cinerea ) Shagbark hickory ( Carya ovata)
Rosaceae Salicaceae Salicaceae Saticaceae Fagaceae Fagaceae Magnoliaceae Betulaceae Juglandaceae Juglandaceae Juglandaceae
Species
X X X X
Genus
Family
X
X X X
X X X X X X
X X X X
X X X X
~Host plant records compiled from Holland (1968), Baker (1972), and Tietz (1972). X = documented feeding; ? = probable feeding, not documented in preceding references.
2022
LINDROTH
were then processed according to Brattsten (1987a) for flash-freezing in liquid nitrogen and storage at - 7 0 ~ All procedures were conducted at 0-4~ Enzyme Assays. Assays were conducted for five different enzyme systems. ;3-Glucosidase activity was measured because these enzymes are involved in the metabolism of glycosidic allelochemicals (often activating rather than detoxifying them), and because little is known about host plant alteration of/3-glucosidases (Lindroth, 1988). Quinone reductase (QR) and glutathione transferase (GT) activities were determined because they are likely to be especially important in the adaptation of luna larvae to members of the Juglandaceae. Polysubstrate monooxygenase (PSMO) and esterase activities were measured because these enzyme systems are generally important detoxification mechanisms in insects. Enzyme assays were optimized with respect to substrate and enzyme concentrations, incubation time, and buffer pH. Each enzyme assay for each enzyme solution was conducted in duplicate or triplicate. Specific activities were calculated relative to protein concentrations of the enzyme preparations, as determined by the Folin-phenol procedure of Schacterle and Pollack (1973). /3-Glucosidase activity was measured according to Lindroth (1988). Each 1 ml incubation solution contained 50-175/xg protein, 50 t~mol salicin, and 0.1 M potassium phosphate buffer (pH 6.0). Solutions were incubated for 30 min at 35 ~ Glucose liberated from salicin by ~3-glucosidase activity was quantified enzymatically (Sigma Diagnostic Kit 315). Quinone reductase (QR) activity of soluble and microsomal fractions was measured using the juglone-dependent NADPH oxidation method of Yu (1987a). First, 40-70/xg protein, 160 #1 NADPH (2.5 mM in phosphate buffer), and 0.1 M sodium phosphate buffer (pH 8.0) were mixed together for a total volume of 2 ml. From that solution, 995/~1 were placed into a sample cuvette and mixed with 5 /~1 juglone (10 mM in methyl cellosolve). The remaining solution was placed into the reference cuvette. NADPH oxidation was determined as the decrease in absorbance at 340 nm over several minutes; a value of 6.22/mM/cm was used for the extinction coefficient of NADPH (Segel, 1976). Use of a double-beam spectrophotometer canceled out effects of endogenous NADPH oxidation because both cuvettes contained enzyme and NADPH. To confirm that the observed activity was due to a reductase rather than a microsomal oxidase, QR activities were compared between an untreated microsomal enzyme solution and a solution that had been bubbled with carbon monoxide (2 bubbles/min for I min). No difference was found in NADPH oxidation between the two samples. Cytochrome c reductase and O-demethylase activities of microsomal fractions were measured as indices of PSMO activity. Cytochrome c reductase activity was quantified as the production of reduced cytochrome c with time, according to Brattsten et al. (1980). Assay solutions (1 ml) consisted of 20-50 /~g protein, 200/zl cytochrome c (4 mg/ml in phosphate buffer), 100/xl NADPH
L U N A MOTH DETOXIFICATION ACTIVITY
2023
(1 mM in phosphate buffer), and 0.1 M potassium phosphate buffer (pH 8.0). Production of reduced cytochrome c was measured as the increase in absorbance (550 nm) over 1 min versus a blank containing everything but the enzyme. No significant endogenous reduction of cytochrome c was observed with this procedure. A value of 27.6/mM/cm was used for the extinction coefficient of reduced cytochrome c (Margoliash and Frohwirt, 1959). The assay for O-demethylase activity was adapted from Hansen and Hodgson (1971) and measured the PSMO-catalyzed formation ofp-nitrophenol from p-nitroanisole. Each assay solution contained 0.5-1.0 mg protein; an NADPH-generating system consisting of 0.36 mM NADP, 3.6 mM glucose-6-phosphate, and 2 units/ ml glucose-6-phosphate dehydrogenase (final assay concentrations); 0.15 M potassium phosphate buffer (pH 7.8, with 1 mM EDTA) to a total volume of 780 /xl; and 20 /xl p-nitroanisole (32 mM in methyl cellosolve) to initiate the reaction. Enzyme solutions were incubated at 32~ for 30 min; reactions were terminated with 200 #1 1 M HC1. Acidified solutions were extracted with 1 ml dichloromethane, and these in turn were extracted with 900/xl 0.5 M NaOH. Absorbance was read at 400 nm and converted to concentration ofp-nitrophenol using a standard curve. Soluble and microsomal esterase activities were quantified as described by Brattsten (1987a). Assay solutions contained 0.2-1.0 ~g protein and 0.05 M sodium phosphate buffer (pH 8.0) to a total volume of 495 /zl. Assays were initiated with addition of 5/zl 1-naphthyl acetate (50 mM in ethanol). Reactions were run at 32~ for 10 min and terminated with addition of a 1 ml solution of fast blue B and sodium dodecyl sulfate (0.2 and 7.5 mg/ml water, respectively). After 10 min, absorbance was read at 600 nm and converted to concentration of 1-naphthol using a standard curve. Soluble and microsomal glutathione transferase activities were measured as the conjugation of 1-chloro-2,4-dinitrobenzene (CDNB) with reduced glutathione, as documented previously (Lindroth, 1989). Reaction mixtures consisted of 4-15 #g protein, 50 #1 glutathione (0.05 M in phosphate buffer), and 0.1 M potassium phosphate buffer to a total volume of 975 #1. Reactions were initiated with addition of 25 #1 CDNB (40 mM in methyl cellosolve) and the increase in absorbance (340 nm) monitored for 1 min against a blank containing everything but enzyme. Again, use of a double-beam spectrophotomer automatically accounted for changes in absorbance due to nonenzymatic conjugation of CDNB. Concentrations of CDNB-glutathione were calculated using an extinction coefficient of 9.6/mM/cm (Cohen, 1986). Statistics. Statistical analyses were performed with the SYSTAT statistical software package. One-way analyses of variance and multiple comparisons among means were conducted using a multiple general linear model (Wilkinson, 19870. For results from the survival trials, percentages were transformed (arcsin -,/y) prior to analysis. For all other experiments in which Bartlett's test
2024
LINDROTH
of homogeneity indicated unequal distribution of variance, data were transformed (ln y) prior to analysis. RESULTS
Feeding Trials. Performance of luna larvae was exceptionally poor on seven of the 11 plant species tested (Table 2). For six of these species, no larvae survived through the first stadium. Survival of larvae fed willow was 33 % for one replicate, but 0% for all others. Larvae fed quaking aspen or red oak consumed small amounts of leaf material prior to death, whereas larvae fed black cherry, cottonwood, or tulip tree showed few, if any, signs of feeding. In contrast, larvae fed birch or any of the three members of the Juglandaceae exhibited very high first instar survival. Among the larvae that survived to pupation, those fed butternut had the shortest larval duration, but also the lowest average pupal weight (Table 2). Larvae fed birch required the longest time prior to pupation and had intermediate pupal weights. Luna reared on black walnut had the highest pupal weights (43 % higher than those reared on the congeneric buttemut) and intermediate larval development times. Enzyme Assays. Survival of larvae was sufficiently high in only four of the 11 treatments to provide enough fifth instars for enzyme assays. Larval food plant did not affect activity of midgut/3-glucosidases (Table 3), but it did influence activity of all the detoxification enzyme systems. TABLE 2. PERFORMANCE OF Actias luna LARVAE ON VARIOUS FOOD PLANTS a
Species Black cherry Cottonwood Quaking aspen White willow Red oak White oak Tulip tree Paper birch Black walnut Butternut Shagbark hickory
First instar survival (%) 0.0 0.0 0.0 5.6 0.0 0.0 0.0 98.1 96.3 92.1 97.6
(5) (5) (5) _+ 5.6(5)a (5) (5) (5) _+ 1.3(9)b _+ 1.5(9)b _+ 3.5(5)b _+ 1.5(7)b
Larval duration (days)
Pupal weight (g)
31.3 (1)
2.20 (1)
31.9 28.8 26.3 29.7
2.25 2.85 2.00 2.49
_+ 0.8(7)c _+ 0.5(8)b _+ 0.6(4)a +_ 0.4(7)b
_ 0.12(7)a _+ 0.15(8)b _+ 0.19(4)a --t-0.17(7)a,b
"Values represent X +_ 1 SE; sample sizes shown in parentheses. Within a column, means followed by different letters are significantly different ( P < 0.05).
98.6a _+11.6 126.2a _+10.7 108.4a -+15.4 118.9a -+9.3
/3-Glucosidase
47.0a +2.0 61.3c -+2.7 53. lb -+1.2 55.7b -+0.9
Soluble 288.3b _+17.3 195.9a _+7.9 173.9a _+9.6 255.0b __8.7
Microsomal 169.9b _+7.8 124. la -+4.2 139. la _+15.3 147.7a,b -+ 10.0
Cytochrome c reductase 249.6c +30.0 179.9b -+24.2 94.4a +16.6 145.2a,b :212.4
O-Demethylase 1072a +48 1940d +111 1337b -+28 1619c -+89
Soluble
1192.6c -+86.7 665.6a -+30.8 691.3a,b +_+_26.1 854.6b -+37.3
Microsomal
Esterase
1016a -+86 2246b +82 2867b _+378 2491b _+86
Soluble
160.4a -+16.4 258.3b,c _+18.2 226. lb -+13.0 282.4c _+21.4
Microsomal
Gluthathione transferase
aSample sizes shown in parentheses. All specific activities except O-demethylase shown as nmol/min/mg protein. O-Demethylase activities shown as pmol/min/mg protein. Within a column, means followed by different letters are significantly different (P _< 0.05).
Shagbark hickory (5)
Butternut (4)
Black walnut (5)
Paper birch (5)
Host plant
Quinone reductase
TABLE 3. SPECIFIC ACTIVITIES (X q- 1 SE) OF MIDGUT ENZYMES IN Actias luna LARVAE FED VARIOUS HOST PLANTSa
b.3 k/I
2026
LINDROTH
Both soluble and microsomal quinone reductase activities were affected by larval host plant (Table 3). Soluble QR activity was lowest in larvae fed birch and highest in those fed walnut, a 1.3-fold difference. Microsomal QR activity, however, was lowest in larvae fed butternut and highest in those fed birch, a 1.7-fold difference. Specific activities (per unit protein) were three to six times higher in microsomal fractions than in soluble fractions. However, given that total soluble protein averaged six to eight times that of microsomal protein, total soluble QR activity was higher than total microsomal QR activity in larvae from all treatments. Cytochrome c reductase and O-demethylase assays showed that highest PSMO activities occurred in larvae reared on birch (Table 3). Lowest cytochrome c reductase levels were in larvae fed walnut, whereas lowest O-demethylase activities were in larvae fed butternut. Host plant effects on enzyme activity accounted for a 1.4- and 2.6-fold variation in cytochrome c reductase and O-demethylase activities, respectively. Activities of soluble esterases were significantly different among all treatments and were 1.8-fold higher in larvae fed walnut than in larvae fed birch (Table 3). Microsomal esterases exhibited an opposite trend in activity, with lowest values in larvae fed walnut, and highest in those fed birch. Specific activities of soluble esterases were higher than those of microsomal esterases in all treatments except birch. Considering the much larger quantity of soluble protein than microsomal protein in the enzyme preparations, a preponderance of midgut esterase activity is effected by soluble enzymes. Glutathione transferase (GT) activities were strongly correlated with host plant genus (Table 3). There were no significant differences in soluble GT activities among larvae fed walnut, butternut, or hickory. Activities exhibited by these larvae were 2.2- to 2.8-fold higher than those of larvae fed birch. Similarly, variations in activities of microsomal GTs among larvae fed members of the Juglandaceae were small in comparison to the difference between those larvae and larvae reared on birch. With respect to major among-treatment trends in enzyme activities, few patterns are evident among larvae fed species of the Juglandaceae. Clearly though, the enzyme profile of larvae reared on birch differed substantially from those of larvae reared on walnut, butternut, or hickory. Birch-fed larvae showed the lowest/3-glucosidase, soluble QR, and soluble and microsomal esterase and GT activities, but the highest cytochrome c reductase, O-demethylase, and microsomal QR activities.
DISCUSSION Larval performance on the trees tested in this study differed significantly from that expected on the basis of published food plant records (Holland, 1968; Baker, 1972; Tietz 1972). Survivorship values of 0 % on white oak and red oak,
LUNA MOTH DETOXIFICATION ACTIVITY
2027
and of 5.6% on willow, were particularly surprising. Larvae performed quite well, and fairly similarly, on birch and the three species of Juglandaceae. These divergent results are best explained by the low genetic variability in the laboratory population. Although Actias luna feeds on a variety of plant families throughout its geographic range, local populations or individuals may specialize on particular plant families. Scriber and Feeny (1979) suggested that a similar situation exists for other species of saturniids. This study lends additional support to the notion of Fox and Morrow (1981) that insects exhibiting generalized diets at the species level are likely to have specialized diets at the population level. Little is known about host plant alteration of midgut enzymes in deciduous tree-feeding insects. These data show that, with one exception, larval food plant strongly affects the activity of midgut enzymes. The single exception was /3-glucosidase activity, which was not significantly affected by the plant species consumed. This lack of response contrasts with results from a similar study (Lindroth, 1988) in which strong host plant effects on/3-glucosidase activity were found in the eastern tiger swallowtail (Papilio glaucus glaucus). Although larval food plant significantly affected quinone reductase activities, the effects were not strong (1.3- and 1.7-fold maximum differences for soluble and microsomal QR activities, respectively). Similarly, Yu (1987a) found that individual allelochemicals induced QR activities in fall armyworm 1.3- to 2.5-fold. Quinone reductase activities in luna larvae were markedly higher than those found by Yu (1987a) in other lepidopteran species. Soluble QR activities in luna were up to 2.3-fold higher than those in the armyworm, and microsomal activities in luna were 5.6-6.2 times higher than those in four other species. Plant alteration of PSMO-related activities was low to moderate in the luna larvae. Cytochrome c reductase activities and induced responses were similar to those reported for Papilio glaucus glaucus fed various host plants (Lindroth, 1989). O-Demethylase activities were also in the range reported for P. g. glaucus (Lindroth, 1989), for fall armyworm fed various crop plants (Yu, 1983), and for fall armyworm, corn earworm, tobacco budworm, and velvetbean caterpillar reared on artificial diets (Yu, 1987b). Published studies of host plant alteration of insect esterase activities have typically shown small to moderate effects (Yu, 1986, 1987b; Brattsten, 1987b). In the present study maximum differences of 1.8-fold were found for both soluble and microsomal esterases. The range of esterase activities in luna larvae was similar to that of P. g. glaucus fed various host plants (Lindroth, 1989) and to the single value of 1004 nmol/min/mg protein reported for gypsy moth (Lymantria dispar) reared on an artificial diet (Kapin and Abroad, 1980). Soluble glutathione transferases exhibited the strongest apparent induction response of the enzymes assayed in this study, although the possibility exists that GT activity was simply inhibited in birch-fed larvae. In addition, the overall soluble GT activity of larvae fed members of the Juglandaceae was quite
2028
LINDROTH
high in comparison to that of other insect species. Midgut glutathione transferase activities (measured by the CDNB assay) for P. g. glaucus, the southern armyworm, fall armyworm, and corn earworm ranged from 600 to 970 nmol/ min/mg protein (Gunderson et al., 1986; Brattsten, 1987a; Yu, 1987b; Lindroth, 1989). Although this was a preliminary study, the results suggest that evolutionary and ecological adaptations of luna to particular food plants involve alterations in biochemical detoxification systems. Moreover, the enzyme systems of particular significance here have rarely demonstrated such importance in other plant-insect associations. Unusually high constitutive levels of quinone reductase and glutathione transferase activities in luna are probably an evolutionary adaptation to the presence of juglone and related quinones in their preferred food plants (Juglandaceae). The fact that soluble QR and soluble and microsomal GT activities appeared to be induced in larvae fed members of the Juglandaceae, relative to those fed birch, shows that biochemical alterations are involved in the adaptations of individual larvae to specific foods. These types of biochemical adjustments may well be involved in local feeding specialization of regionally generalist insects. Finally, this study illustrates that an understanding of the dominant allelochemicals in a plant-insect system can give clues as to the type of biochemical adaptations exhibited by the insects. Acknowledgments--I thank Miel Barman for technical assistance, and Mark Evans for advice on rearing the larvae. This research was supported by NSF grant BSR 8503464, by USDA (CRGO) grants 85-CRCR-1598 and 87-CRCR-1-258 l, and by the College of Agricultural and Life Sciences, University of Wisconsin (Hatch 3211).
REFERENCES
BAKER,W.L. 1972. Eastern Forest Insects. U.S. Department of Agriculture, Washington, D.C. BRATTSTEN, L.B. 1987a. Metabolic insecticide defenses in the boll weevil compared to those in a resistance-prone species. Pestic. Biochem. Physiol. 27:1-12. BRATTSTEN, L.B. 1987b. Inducibility of metabolic insecticide defenses in boll weevils and tobacco budworm caterpillars. Pestic. Biochem. Physiol. 27:13-23. BRATTSTEN, L.B., PRICE, S.L., and GUNDERSON,C.A. 1980. Microsomal oxidases in midgut and fatbody tissues of a broadly herbivorous insect larva, Spodoptera eridania Cramer (Noctuidae). Comp. Biochem. Physiol. 66C:231-237. COHEN,E. 1986. Glutathione-S-transferase activity and its induction in several strains of Tribolium castaneum. Entomol. Exp. Appl. 41:39-44. Fox, L.R., and MORROW, P.A. 1981. Specialization: Species property or local phenomenon.'? Science 211 :887-893. GILBERT, B.L., BAKER, J.E., and NORRIS, D.M. 1967. Juglone (5-hydroxy-l,4-naphthoquinone) from Carya ovata, a deterrent to feeding by Scolytus multistriatus. J. Insect Physiol. 13:14531459. GUNDERSON, C.A., BRATTSTEN, L.B., and FLEMING,J.T. 1986. Microsomal oxidase and glutathione transferase as factors influencing the effects of pulegone in southern and fall armyworm larvae. Pestic. Biochem. Physiol. 26:238-249.
LUNA MOTH DETOXIFICATION ACTIVITY
2029
HANSEN, L.G., and HODGSON, E. 1971. Biochemical characteristics of insect microsomes. N- and O-demethylation. Biochem. Pharmacol. 20:1569-1578. HEDIN, P.A., COLLUM,D.H., LANGHANS,V.E., and GRAVES,C.H. 1980. Distribution ofjuglone and related compounds in pecan and their effect on Fusicladium effusum. J. Agric. Food Chem. 28:340-342. HOLLAND, W.J. 1968. The Moth Book. Dover Publications, New York. KAPIN, M.A., and AHMAD, S. 1980. Esterases in larval tissues of gypsy moth, Lymantria dispar (L.): Optimum assay conditions, quantification and characterization. Insect Biochem. 10:331337. LINDROTH, R.L. 1988. Hydrolysis of phenolic glycosides by midgut /3-glucosidases in Papitio glaucus subspecies. Insect Biochem. In press. LINDROTH, R.L. 1989. Host plant alteration of detoxication enzyme activity in Papilio glaucus glaucus. Entomol. Exp. Appl. 50:29-35. MARGOLIASH, E., and FROHWIRT, N. 1959. Spectrum of horse-heart cytochrome c. Biochem. J. 71:570-572. NORRIS, D.M. 1986. Anti-feeding compounds, pp. 97-146, in G. Haug and H. Hoffman (eds.). Chemistry of Plant Protection. 1. Sterol Biosynthesis, Inhibitors and Anti-Feeding Compounds. Springer-Verlag, New York. SCHACTERLE, G.R., and POLLACK,R.L. 1973. A simplified method for quantitative assay of small amounts of protein in biologic material. Anal. Biochem. 51:654-655. SCRIBER, J.M., and FEENY, P. 1979. Growth of herbivorous caterpillars in relation to feeding specialization and to the growth form of their food plants. Ecology 60:829-850. SEGEL, I.H. 1976. Biochemical Calculations, 2nd ed. John Wiley & Sons, New York. TERRIERE, L.C. 1984. Induction of detoxication enzymes in insects. Annu. Rev. Entomol. 29:7188. TIETZ, H.M. 1972. An Index to the Described Life Histories, Early Stages and Hosts of the Macrolepidoptera of the Continental United States and Canada. A.C. Allyn, Sarasota, Florida. WADLEmH, R.W., and Yu, S.J. 1987. Glutathione transferase activity of fall armyworm larvae toward e~,~-unsaturated carbonyl allelochemicals and its induction by allelochemicals. Insect Biochem. 17:759-764. WILKINSON, L. 1987. SYSTAT: The System for Statistics. SYSTAT, Inc., Evanston, Illinois. Yu, S.J. 1983. Induction of detoxifying enzymes by allelochemicals and host plants in fall armyworm. Pestic. Biochem. Physiol. 9:330-336. Yu, S.J. 1986. Consequences of induction of foreign compound-metabolizing enzymes in insects, pp. 153-174, in L.B. Brattsten and S. Ahmad (eds.). Molecular Aspects of Insect-Plant Associations. Plenum Press, New York. Yu, S.J. 1987a. Quinone reductase of phytophagous insects and its induction by allelochemicals. Comp. Biochem. Physiol. 87B:621-624. Yu, S.J. 1987b. Biochemical defense capacity in the spined soldier bug (Podisus maculiventris) and its lepidopterous prey. Pestic. Biochem. Physiol. 28:216-223.
CHEMICAL ECOLOGY OF THE LUNA MOTH Effects of Host Plant on Detoxification Enzyme Activity
RICHARD
L. LINDROTH
Department of Entomology University of Wisconsin 23 7 Russell Laboratories Madison, Wisconsin 53706 (Received May 13, 1988; accepted October 10, 1988)
Abstract--The effects of food plant on larval performance and midgut detoxification enzymes were investigated in larvae of the tuna moth, Aetias luna. Neonate larvae were fed leaves of black cherry, cottonwood, quaking aspen, white willow, red oak, white oak, tulip tree, paper birch, black walnut, buttemut, or shagbark hickory. First instar survival, larval duration, and pupal weights were monitored as indices of food quality. Midgut enzyme preparations from fifth instars were assayed for/3-glucosidase, quinone reductase, polysubstrate monooxygenase, esterase, and glutathione transferase activities. Larval survival on seven of the 11 plant species, including several recorded host plants, was extremely poor. Larvae performed well, and quite similarly, on birch, walnut, butternut, and hickory. Activities of all enzyme systems except /3-glucosidase were significantly influenced by larval host plant. Of the systems assayed, quinone reductase and glutathione transferase activities were especially high. Comparisons of these values with published values for other Lepidoptera support the hypothesis that these enzyme systems are involved in conferring tolerance to juglone and related quinones Occurring in members of the plant family Juglandaceae. Results suggest that host plant utilization by tuna is more specialized at the individual or population level than at the species level and that biochemical detoxification systems may play a role in such specialization. Key Words--Actias luna, Lepidoptera, Satumiidae, detoxification enzymes, enzyme induction, glutathione transferase, Juglandaceae, juglone, nutritional ecology, plant-insect interactions, quinone reductase.
2019 0098-0331/89/0700-2019506.00/0 9 1989 Plenum Publishing Corporation
2020
L~NDROTH INTRODUCTION
The luna moth, Actias luna L. (Saturniidae), occurs from southern Canada to Texas and Florida (Baker, 1972). As a species, it is moderately polyphagous, feeding on trees from at least eight plant families (Baker, 1972; Tietz, 1972). Little is known, however, of the relative suitability of various tree species as larval food plants. The research reported here was conducted in part to assess the influence of known or potential host plants on performance (survival and growth) of luna larvae. A second objective of this research was to investigate host plant alteration of allelochemical-metabolizing enzymes in luna larvae. Detoxification of plant allelochemicals by specialized enzyme systems is widely believed to be one of the most important forms of insect adaptation to plant diets. Yet few studies have documented the enzymatic detoxification capacities of tree-feeding insects, and fewer still have addressed potential host plant effects on multiple enzyme systems. These studies provided the opportunity to test several predictions about evolutionary and ecological adaptations of enzymatic detoxification systems to particular food plants. Prior to the onset of this study, I had observed that luna larvae perform well on black walnut (Juglans nigra) and shagbark hickory (Carya ovata). The most distinguishing secondary metabolites of these and other members of the Juglandaceae are juglone and related 1,4-naphthoquinones, compounds that are generally deterrent or toxic to insects (Gilbert et al., 1967; Hedin et al., 1980; Norris, 1986). The enzyme systems most likely responsible for detoxification of juglone in Juglandaceae-adapted insects include quinone reductase, which catalyzes the reduction of quinones to hydroquinones (Yu, 1987a), and glutathione transferase, which conjugates allelochemicals containing ~,/3-unsaturated carbonyls (e.g., juglone) with reduced glutathione (Wadleigh and Yu, 1987). These enzyme systems can be induced by dietary exposure to various allelochemicals. Thus I predicted that quinone reductase and glutathione transferase activities would (1) be especially high in luna larvae in comparison to published values for other polyphagous Lepidoptera, and (2) be induced in larvae fed members of the Juglandaceae, in comparison to larvae fed members of other plant families.
METHODS AND MATERIALS
Insects. Actias luna eggs were obtained from a gravid female moth captured in Laurel County, Kentucky, in May 1987. Larvae were reared in plastic boxes at ambient temperature on leaves of black walnut or shagbark hickory. Newly eclosed adult moths were paired to provide larvae for this study. To
LUNA MOTH DETOXIFICATION ACTIVITY
2021
introduce more genetic variability into the laboratory culture, an attempt was made to pair laboratory-reared females with local males by placing tethered females in nearby woodlots; these attempts proved unsuccessful. Feeding Trials. Neonate larvae were placed onto fresh leaves of 11 species of trees, including six genera on which luna are known to feed, and two genera for which no previous feeding records exist (Table 1). Each replicate consisted of 10-14 larvae, with leaves, in a ventilated plastic Petri dish (15 x 2.5 cm); leaf petioles were inserted into florist's Water Piks to maintain leaf turgor. Leaves were replaced at two- to three-day intervals. When larvae reached the fourth stadium, they were transferred to larger shoebox cages. Larvae were reared in a Percival environmental chamber at 24~ on a 12:12 light-dark cycle. Measurements were made of first instar survival, duration from hatching to spin-up of the cocoon, and pupal weight (six to seven days following spinup) as indices of food plant quality. Enzyme Preparation. Because larval midguts are the major site for detoxification of plant allelochemicals, they were used as the enzyme source. Midguts (5-9 per enzyme replicate) were dissected from fifth instar larvae (3 to 7 days old) and gut contents removed. Midguts were then washed (0.2 M phosphate buffer, pH 7.8) and homogenized by 10 strokes in a Ten Broeck tissue grinder. The homogenate was centrifuged at 10,000g (10 min) and the supernatant removed and centrifuged at 100,000g (60 min), to separate soluble (cytosolic) and microsomal (membrane-bound) enzymes. The enzyme preparations
TABLE 1. TREE SPECIES ASSAYED FOR LARVAL PERFORMANCE
Luna previously recorded to feed on the: a Species assayed
Plant family
Black cherry (Prunus serotina) Cottonwood ( Populus deltoides ) Quaking aspen ( Populus tremuloides ) White willow ( Salix alba ) Red oak (Quercus rubra) White oak (Quercus alba ) Tulip tree ( Liriodendron tulipifera ) Paper birch (Betula papyrifera ) Black walnut ( Juglans nigra ) Butternut ( Juglans cinerea ) Shagbark hickory ( Carya ovata)
Rosaceae Salicaceae Salicaceae Saticaceae Fagaceae Fagaceae Magnoliaceae Betulaceae Juglandaceae Juglandaceae Juglandaceae
Species
X X X X
Genus
Family
X
X X X
X X X X X X
X X X X
X X X X
~Host plant records compiled from Holland (1968), Baker (1972), and Tietz (1972). X = documented feeding; ? = probable feeding, not documented in preceding references.
2022
LINDROTH
were then processed according to Brattsten (1987a) for flash-freezing in liquid nitrogen and storage at - 7 0 ~ All procedures were conducted at 0-4~ Enzyme Assays. Assays were conducted for five different enzyme systems. ;3-Glucosidase activity was measured because these enzymes are involved in the metabolism of glycosidic allelochemicals (often activating rather than detoxifying them), and because little is known about host plant alteration of/3-glucosidases (Lindroth, 1988). Quinone reductase (QR) and glutathione transferase (GT) activities were determined because they are likely to be especially important in the adaptation of luna larvae to members of the Juglandaceae. Polysubstrate monooxygenase (PSMO) and esterase activities were measured because these enzyme systems are generally important detoxification mechanisms in insects. Enzyme assays were optimized with respect to substrate and enzyme concentrations, incubation time, and buffer pH. Each enzyme assay for each enzyme solution was conducted in duplicate or triplicate. Specific activities were calculated relative to protein concentrations of the enzyme preparations, as determined by the Folin-phenol procedure of Schacterle and Pollack (1973). /3-Glucosidase activity was measured according to Lindroth (1988). Each 1 ml incubation solution contained 50-175/xg protein, 50 t~mol salicin, and 0.1 M potassium phosphate buffer (pH 6.0). Solutions were incubated for 30 min at 35 ~ Glucose liberated from salicin by ~3-glucosidase activity was quantified enzymatically (Sigma Diagnostic Kit 315). Quinone reductase (QR) activity of soluble and microsomal fractions was measured using the juglone-dependent NADPH oxidation method of Yu (1987a). First, 40-70/xg protein, 160 #1 NADPH (2.5 mM in phosphate buffer), and 0.1 M sodium phosphate buffer (pH 8.0) were mixed together for a total volume of 2 ml. From that solution, 995/~1 were placed into a sample cuvette and mixed with 5 /~1 juglone (10 mM in methyl cellosolve). The remaining solution was placed into the reference cuvette. NADPH oxidation was determined as the decrease in absorbance at 340 nm over several minutes; a value of 6.22/mM/cm was used for the extinction coefficient of NADPH (Segel, 1976). Use of a double-beam spectrophotometer canceled out effects of endogenous NADPH oxidation because both cuvettes contained enzyme and NADPH. To confirm that the observed activity was due to a reductase rather than a microsomal oxidase, QR activities were compared between an untreated microsomal enzyme solution and a solution that had been bubbled with carbon monoxide (2 bubbles/min for I min). No difference was found in NADPH oxidation between the two samples. Cytochrome c reductase and O-demethylase activities of microsomal fractions were measured as indices of PSMO activity. Cytochrome c reductase activity was quantified as the production of reduced cytochrome c with time, according to Brattsten et al. (1980). Assay solutions (1 ml) consisted of 20-50 /~g protein, 200/zl cytochrome c (4 mg/ml in phosphate buffer), 100/xl NADPH
L U N A MOTH DETOXIFICATION ACTIVITY
2023
(1 mM in phosphate buffer), and 0.1 M potassium phosphate buffer (pH 8.0). Production of reduced cytochrome c was measured as the increase in absorbance (550 nm) over 1 min versus a blank containing everything but the enzyme. No significant endogenous reduction of cytochrome c was observed with this procedure. A value of 27.6/mM/cm was used for the extinction coefficient of reduced cytochrome c (Margoliash and Frohwirt, 1959). The assay for O-demethylase activity was adapted from Hansen and Hodgson (1971) and measured the PSMO-catalyzed formation ofp-nitrophenol from p-nitroanisole. Each assay solution contained 0.5-1.0 mg protein; an NADPH-generating system consisting of 0.36 mM NADP, 3.6 mM glucose-6-phosphate, and 2 units/ ml glucose-6-phosphate dehydrogenase (final assay concentrations); 0.15 M potassium phosphate buffer (pH 7.8, with 1 mM EDTA) to a total volume of 780 /xl; and 20 /xl p-nitroanisole (32 mM in methyl cellosolve) to initiate the reaction. Enzyme solutions were incubated at 32~ for 30 min; reactions were terminated with 200 #1 1 M HC1. Acidified solutions were extracted with 1 ml dichloromethane, and these in turn were extracted with 900/xl 0.5 M NaOH. Absorbance was read at 400 nm and converted to concentration ofp-nitrophenol using a standard curve. Soluble and microsomal esterase activities were quantified as described by Brattsten (1987a). Assay solutions contained 0.2-1.0 ~g protein and 0.05 M sodium phosphate buffer (pH 8.0) to a total volume of 495 /zl. Assays were initiated with addition of 5/zl 1-naphthyl acetate (50 mM in ethanol). Reactions were run at 32~ for 10 min and terminated with addition of a 1 ml solution of fast blue B and sodium dodecyl sulfate (0.2 and 7.5 mg/ml water, respectively). After 10 min, absorbance was read at 600 nm and converted to concentration of 1-naphthol using a standard curve. Soluble and microsomal glutathione transferase activities were measured as the conjugation of 1-chloro-2,4-dinitrobenzene (CDNB) with reduced glutathione, as documented previously (Lindroth, 1989). Reaction mixtures consisted of 4-15 #g protein, 50 #1 glutathione (0.05 M in phosphate buffer), and 0.1 M potassium phosphate buffer to a total volume of 975 #1. Reactions were initiated with addition of 25 #1 CDNB (40 mM in methyl cellosolve) and the increase in absorbance (340 nm) monitored for 1 min against a blank containing everything but enzyme. Again, use of a double-beam spectrophotomer automatically accounted for changes in absorbance due to nonenzymatic conjugation of CDNB. Concentrations of CDNB-glutathione were calculated using an extinction coefficient of 9.6/mM/cm (Cohen, 1986). Statistics. Statistical analyses were performed with the SYSTAT statistical software package. One-way analyses of variance and multiple comparisons among means were conducted using a multiple general linear model (Wilkinson, 19870. For results from the survival trials, percentages were transformed (arcsin -,/y) prior to analysis. For all other experiments in which Bartlett's test
2024
LINDROTH
of homogeneity indicated unequal distribution of variance, data were transformed (ln y) prior to analysis. RESULTS
Feeding Trials. Performance of luna larvae was exceptionally poor on seven of the 11 plant species tested (Table 2). For six of these species, no larvae survived through the first stadium. Survival of larvae fed willow was 33 % for one replicate, but 0% for all others. Larvae fed quaking aspen or red oak consumed small amounts of leaf material prior to death, whereas larvae fed black cherry, cottonwood, or tulip tree showed few, if any, signs of feeding. In contrast, larvae fed birch or any of the three members of the Juglandaceae exhibited very high first instar survival. Among the larvae that survived to pupation, those fed butternut had the shortest larval duration, but also the lowest average pupal weight (Table 2). Larvae fed birch required the longest time prior to pupation and had intermediate pupal weights. Luna reared on black walnut had the highest pupal weights (43 % higher than those reared on the congeneric buttemut) and intermediate larval development times. Enzyme Assays. Survival of larvae was sufficiently high in only four of the 11 treatments to provide enough fifth instars for enzyme assays. Larval food plant did not affect activity of midgut/3-glucosidases (Table 3), but it did influence activity of all the detoxification enzyme systems. TABLE 2. PERFORMANCE OF Actias luna LARVAE ON VARIOUS FOOD PLANTS a
Species Black cherry Cottonwood Quaking aspen White willow Red oak White oak Tulip tree Paper birch Black walnut Butternut Shagbark hickory
First instar survival (%) 0.0 0.0 0.0 5.6 0.0 0.0 0.0 98.1 96.3 92.1 97.6
(5) (5) (5) _+ 5.6(5)a (5) (5) (5) _+ 1.3(9)b _+ 1.5(9)b _+ 3.5(5)b _+ 1.5(7)b
Larval duration (days)
Pupal weight (g)
31.3 (1)
2.20 (1)
31.9 28.8 26.3 29.7
2.25 2.85 2.00 2.49
_+ 0.8(7)c _+ 0.5(8)b _+ 0.6(4)a +_ 0.4(7)b
_ 0.12(7)a _+ 0.15(8)b _+ 0.19(4)a --t-0.17(7)a,b
"Values represent X +_ 1 SE; sample sizes shown in parentheses. Within a column, means followed by different letters are significantly different ( P < 0.05).
98.6a _+11.6 126.2a _+10.7 108.4a -+15.4 118.9a -+9.3
/3-Glucosidase
47.0a +2.0 61.3c -+2.7 53. lb -+1.2 55.7b -+0.9
Soluble 288.3b _+17.3 195.9a _+7.9 173.9a _+9.6 255.0b __8.7
Microsomal 169.9b _+7.8 124. la -+4.2 139. la _+15.3 147.7a,b -+ 10.0
Cytochrome c reductase 249.6c +30.0 179.9b -+24.2 94.4a +16.6 145.2a,b :212.4
O-Demethylase 1072a +48 1940d +111 1337b -+28 1619c -+89
Soluble
1192.6c -+86.7 665.6a -+30.8 691.3a,b +_+_26.1 854.6b -+37.3
Microsomal
Esterase
1016a -+86 2246b +82 2867b _+378 2491b _+86
Soluble
160.4a -+16.4 258.3b,c _+18.2 226. lb -+13.0 282.4c _+21.4
Microsomal
Gluthathione transferase
aSample sizes shown in parentheses. All specific activities except O-demethylase shown as nmol/min/mg protein. O-Demethylase activities shown as pmol/min/mg protein. Within a column, means followed by different letters are significantly different (P _< 0.05).
Shagbark hickory (5)
Butternut (4)
Black walnut (5)
Paper birch (5)
Host plant
Quinone reductase
TABLE 3. SPECIFIC ACTIVITIES (X q- 1 SE) OF MIDGUT ENZYMES IN Actias luna LARVAE FED VARIOUS HOST PLANTSa
b.3 k/I
2026
LINDROTH
Both soluble and microsomal quinone reductase activities were affected by larval host plant (Table 3). Soluble QR activity was lowest in larvae fed birch and highest in those fed walnut, a 1.3-fold difference. Microsomal QR activity, however, was lowest in larvae fed butternut and highest in those fed birch, a 1.7-fold difference. Specific activities (per unit protein) were three to six times higher in microsomal fractions than in soluble fractions. However, given that total soluble protein averaged six to eight times that of microsomal protein, total soluble QR activity was higher than total microsomal QR activity in larvae from all treatments. Cytochrome c reductase and O-demethylase assays showed that highest PSMO activities occurred in larvae reared on birch (Table 3). Lowest cytochrome c reductase levels were in larvae fed walnut, whereas lowest O-demethylase activities were in larvae fed butternut. Host plant effects on enzyme activity accounted for a 1.4- and 2.6-fold variation in cytochrome c reductase and O-demethylase activities, respectively. Activities of soluble esterases were significantly different among all treatments and were 1.8-fold higher in larvae fed walnut than in larvae fed birch (Table 3). Microsomal esterases exhibited an opposite trend in activity, with lowest values in larvae fed walnut, and highest in those fed birch. Specific activities of soluble esterases were higher than those of microsomal esterases in all treatments except birch. Considering the much larger quantity of soluble protein than microsomal protein in the enzyme preparations, a preponderance of midgut esterase activity is effected by soluble enzymes. Glutathione transferase (GT) activities were strongly correlated with host plant genus (Table 3). There were no significant differences in soluble GT activities among larvae fed walnut, butternut, or hickory. Activities exhibited by these larvae were 2.2- to 2.8-fold higher than those of larvae fed birch. Similarly, variations in activities of microsomal GTs among larvae fed members of the Juglandaceae were small in comparison to the difference between those larvae and larvae reared on birch. With respect to major among-treatment trends in enzyme activities, few patterns are evident among larvae fed species of the Juglandaceae. Clearly though, the enzyme profile of larvae reared on birch differed substantially from those of larvae reared on walnut, butternut, or hickory. Birch-fed larvae showed the lowest/3-glucosidase, soluble QR, and soluble and microsomal esterase and GT activities, but the highest cytochrome c reductase, O-demethylase, and microsomal QR activities.
DISCUSSION Larval performance on the trees tested in this study differed significantly from that expected on the basis of published food plant records (Holland, 1968; Baker, 1972; Tietz 1972). Survivorship values of 0 % on white oak and red oak,
LUNA MOTH DETOXIFICATION ACTIVITY
2027
and of 5.6% on willow, were particularly surprising. Larvae performed quite well, and fairly similarly, on birch and the three species of Juglandaceae. These divergent results are best explained by the low genetic variability in the laboratory population. Although Actias luna feeds on a variety of plant families throughout its geographic range, local populations or individuals may specialize on particular plant families. Scriber and Feeny (1979) suggested that a similar situation exists for other species of saturniids. This study lends additional support to the notion of Fox and Morrow (1981) that insects exhibiting generalized diets at the species level are likely to have specialized diets at the population level. Little is known about host plant alteration of midgut enzymes in deciduous tree-feeding insects. These data show that, with one exception, larval food plant strongly affects the activity of midgut enzymes. The single exception was /3-glucosidase activity, which was not significantly affected by the plant species consumed. This lack of response contrasts with results from a similar study (Lindroth, 1988) in which strong host plant effects on/3-glucosidase activity were found in the eastern tiger swallowtail (Papilio glaucus glaucus). Although larval food plant significantly affected quinone reductase activities, the effects were not strong (1.3- and 1.7-fold maximum differences for soluble and microsomal QR activities, respectively). Similarly, Yu (1987a) found that individual allelochemicals induced QR activities in fall armyworm 1.3- to 2.5-fold. Quinone reductase activities in luna larvae were markedly higher than those found by Yu (1987a) in other lepidopteran species. Soluble QR activities in luna were up to 2.3-fold higher than those in the armyworm, and microsomal activities in luna were 5.6-6.2 times higher than those in four other species. Plant alteration of PSMO-related activities was low to moderate in the luna larvae. Cytochrome c reductase activities and induced responses were similar to those reported for Papilio glaucus glaucus fed various host plants (Lindroth, 1989). O-Demethylase activities were also in the range reported for P. g. glaucus (Lindroth, 1989), for fall armyworm fed various crop plants (Yu, 1983), and for fall armyworm, corn earworm, tobacco budworm, and velvetbean caterpillar reared on artificial diets (Yu, 1987b). Published studies of host plant alteration of insect esterase activities have typically shown small to moderate effects (Yu, 1986, 1987b; Brattsten, 1987b). In the present study maximum differences of 1.8-fold were found for both soluble and microsomal esterases. The range of esterase activities in luna larvae was similar to that of P. g. glaucus fed various host plants (Lindroth, 1989) and to the single value of 1004 nmol/min/mg protein reported for gypsy moth (Lymantria dispar) reared on an artificial diet (Kapin and Abroad, 1980). Soluble glutathione transferases exhibited the strongest apparent induction response of the enzymes assayed in this study, although the possibility exists that GT activity was simply inhibited in birch-fed larvae. In addition, the overall soluble GT activity of larvae fed members of the Juglandaceae was quite
2028
LINDROTH
high in comparison to that of other insect species. Midgut glutathione transferase activities (measured by the CDNB assay) for P. g. glaucus, the southern armyworm, fall armyworm, and corn earworm ranged from 600 to 970 nmol/ min/mg protein (Gunderson et al., 1986; Brattsten, 1987a; Yu, 1987b; Lindroth, 1989). Although this was a preliminary study, the results suggest that evolutionary and ecological adaptations of luna to particular food plants involve alterations in biochemical detoxification systems. Moreover, the enzyme systems of particular significance here have rarely demonstrated such importance in other plant-insect associations. Unusually high constitutive levels of quinone reductase and glutathione transferase activities in luna are probably an evolutionary adaptation to the presence of juglone and related quinones in their preferred food plants (Juglandaceae). The fact that soluble QR and soluble and microsomal GT activities appeared to be induced in larvae fed members of the Juglandaceae, relative to those fed birch, shows that biochemical alterations are involved in the adaptations of individual larvae to specific foods. These types of biochemical adjustments may well be involved in local feeding specialization of regionally generalist insects. Finally, this study illustrates that an understanding of the dominant allelochemicals in a plant-insect system can give clues as to the type of biochemical adaptations exhibited by the insects. Acknowledgments--I thank Miel Barman for technical assistance, and Mark Evans for advice on rearing the larvae. This research was supported by NSF grant BSR 8503464, by USDA (CRGO) grants 85-CRCR-1598 and 87-CRCR-1-258 l, and by the College of Agricultural and Life Sciences, University of Wisconsin (Hatch 3211).
REFERENCES
BAKER,W.L. 1972. Eastern Forest Insects. U.S. Department of Agriculture, Washington, D.C. BRATTSTEN, L.B. 1987a. Metabolic insecticide defenses in the boll weevil compared to those in a resistance-prone species. Pestic. Biochem. Physiol. 27:1-12. BRATTSTEN, L.B. 1987b. Inducibility of metabolic insecticide defenses in boll weevils and tobacco budworm caterpillars. Pestic. Biochem. Physiol. 27:13-23. BRATTSTEN, L.B., PRICE, S.L., and GUNDERSON,C.A. 1980. Microsomal oxidases in midgut and fatbody tissues of a broadly herbivorous insect larva, Spodoptera eridania Cramer (Noctuidae). Comp. Biochem. Physiol. 66C:231-237. COHEN,E. 1986. Glutathione-S-transferase activity and its induction in several strains of Tribolium castaneum. Entomol. Exp. Appl. 41:39-44. Fox, L.R., and MORROW, P.A. 1981. Specialization: Species property or local phenomenon.'? Science 211 :887-893. GILBERT, B.L., BAKER, J.E., and NORRIS, D.M. 1967. Juglone (5-hydroxy-l,4-naphthoquinone) from Carya ovata, a deterrent to feeding by Scolytus multistriatus. J. Insect Physiol. 13:14531459. GUNDERSON, C.A., BRATTSTEN, L.B., and FLEMING,J.T. 1986. Microsomal oxidase and glutathione transferase as factors influencing the effects of pulegone in southern and fall armyworm larvae. Pestic. Biochem. Physiol. 26:238-249.
LUNA MOTH DETOXIFICATION ACTIVITY
2029
HANSEN, L.G., and HODGSON, E. 1971. Biochemical characteristics of insect microsomes. N- and O-demethylation. Biochem. Pharmacol. 20:1569-1578. HEDIN, P.A., COLLUM,D.H., LANGHANS,V.E., and GRAVES,C.H. 1980. Distribution ofjuglone and related compounds in pecan and their effect on Fusicladium effusum. J. Agric. Food Chem. 28:340-342. HOLLAND, W.J. 1968. The Moth Book. Dover Publications, New York. KAPIN, M.A., and AHMAD, S. 1980. Esterases in larval tissues of gypsy moth, Lymantria dispar (L.): Optimum assay conditions, quantification and characterization. Insect Biochem. 10:331337. LINDROTH, R.L. 1988. Hydrolysis of phenolic glycosides by midgut /3-glucosidases in Papitio glaucus subspecies. Insect Biochem. In press. LINDROTH, R.L. 1989. Host plant alteration of detoxication enzyme activity in Papilio glaucus glaucus. Entomol. Exp. Appl. 50:29-35. MARGOLIASH, E., and FROHWIRT, N. 1959. Spectrum of horse-heart cytochrome c. Biochem. J. 71:570-572. NORRIS, D.M. 1986. Anti-feeding compounds, pp. 97-146, in G. Haug and H. Hoffman (eds.). Chemistry of Plant Protection. 1. Sterol Biosynthesis, Inhibitors and Anti-Feeding Compounds. Springer-Verlag, New York. SCHACTERLE, G.R., and POLLACK,R.L. 1973. A simplified method for quantitative assay of small amounts of protein in biologic material. Anal. Biochem. 51:654-655. SCRIBER, J.M., and FEENY, P. 1979. Growth of herbivorous caterpillars in relation to feeding specialization and to the growth form of their food plants. Ecology 60:829-850. SEGEL, I.H. 1976. Biochemical Calculations, 2nd ed. John Wiley & Sons, New York. TERRIERE, L.C. 1984. Induction of detoxication enzymes in insects. Annu. Rev. Entomol. 29:7188. TIETZ, H.M. 1972. An Index to the Described Life Histories, Early Stages and Hosts of the Macrolepidoptera of the Continental United States and Canada. A.C. Allyn, Sarasota, Florida. WADLEmH, R.W., and Yu, S.J. 1987. Glutathione transferase activity of fall armyworm larvae toward e~,~-unsaturated carbonyl allelochemicals and its induction by allelochemicals. Insect Biochem. 17:759-764. WILKINSON, L. 1987. SYSTAT: The System for Statistics. SYSTAT, Inc., Evanston, Illinois. Yu, S.J. 1983. Induction of detoxifying enzymes by allelochemicals and host plants in fall armyworm. Pestic. Biochem. Physiol. 9:330-336. Yu, S.J. 1986. Consequences of induction of foreign compound-metabolizing enzymes in insects, pp. 153-174, in L.B. Brattsten and S. Ahmad (eds.). Molecular Aspects of Insect-Plant Associations. Plenum Press, New York. Yu, S.J. 1987a. Quinone reductase of phytophagous insects and its induction by allelochemicals. Comp. Biochem. Physiol. 87B:621-624. Yu, S.J. 1987b. Biochemical defense capacity in the spined soldier bug (Podisus maculiventris) and its lepidopterous prey. Pestic. Biochem. Physiol. 28:216-223.
