Archives of Insect Biochemistry and Physiology 1:311-321 (1984) Microsomal Marker Enzymes of Manduca sexta (1) Midgut Gunter F. Weirich and Jean R. Adams Insect Physiology Luboratory (G.F.W.) and Insect Pathology Laboratory (J.R.A.),Agricultural Research Sewice, USDA, Beltsville, Maryland The subcellular distribution of four enzymes (glucose-6-phosphatase, phosphodiesterase I, NADPH-cytochrome c reductase, and p-nitroanisole 0demethylase) in the rnidgut of "wandering" fifth-instar larvae of the tobacco hornworm, Manduca sexta (L), was determined and the composition of mitochondria1 and microsomal pellets was examined by electron microscopy. Most of the glucose-6-phosphatase activity and one-third of the phosphodiesterase 1 activity were found in the high-speed supernatant. NADPH-cytochrome c reductase activity was marginal and 0-demethylase activity was undetectable in the supernatant. The highest specific activities for phosphodiesterase I , NADPH-cytochrome c reductase, and p-nitroanisole 0-demethylase were measured in microsomes, but the relative specific activity of phosphodiesterase I was only half that obtained with the latter two enzymes. In all subcellular preparations the relative specific activities of NADPH-cytochrome c reductase and p-nitroanisole 0-demethylase were closely correlated. It i s concluded that glucose-6-phosphatase and phosphodiesterase I are not microsomal marker enzymes i n the midgut, but the activities of NADPH-cytochrome c reductase and p-nitroanisole 0dernethylase are quantitative measures of microsomal content. Key words: NADPH-cytochrome c reductase, pnitroanisole O-demethylase, Manduca sexta, midgut, microsomal marker enzymes INTRODUCTION In enzyme distribution studies on insects, marker enzymes that have been established for mammalian tissues are commonly used for the characterization of subcellular fractions. The concept of marker enzymes implies that specific enzymes are predominantly localized in a single cellular component [1,2]. Systematic studies to show that mammalian marker enzymes satisfy this definition when used for the characterization of insect tissue fractions are lacking. Address reprint requests to Gunter F. Weirich, Insect Physiology Laboratory, Building 467, BARC-East, Beltsville, MD 20705. Received October 7,1983; accepted April 13,1984. 0 1984 Alan R. Liss, Inc. 312 Weirich and Adams The aim of the present investigation was to identlfy microsomal marker enzymes in an insect tissue. We report the subcellular distribution of four enzymes in the midgut of fifth-instar larvae of the tobacco hornworm, Manduca sexta. Two of these enzymes, glucose-6-phosphatase and phosphodiesterase I, have previously been identified as microsomal enzymes in rat liver [1,3]. The other two enzymes, NADPH-cytochrome c reductase and p-nitroanisole Odemethylase,* have been found in microsomes of vertebrates [4-71 as well as insects [5,8-121. Their exclusive microsomal localization, however, has not been established in insects, and in some vertebrate tissues, NADPH-cytochrome c reductase has been found in cellular components other than microsomes [13-151. We report that only two of the enzymes-are NADPH-cytochrome c reductase and p-nitroanisole 0-demethylase-suitable as microsomal markers in the midgut of M sexta. The subcellular distributions of glucose-6-phosphatase and phosphodiesterase I, on the other hand, are inconsistent with a predominantly microsomal localization. These findings suggest that enzyme distribution patterns of mammalian tissues should not be applied to insect tissues without verification. MATERIALS AND METHODS** Chemicals NADP+, sodium salt; NADPH, tetrasodium salt, type I; glucose 6-phosphate, monosodium salt; glucose-6-phosphate dehydrogenase, type XV (from baker’s yeast); cytochrome c, type VI; thymidine 5’-monophospho-p-nitrophenyl ester, sodium salt, grade I; and bovine serum albumin,? fraction V, were obtained from Sigma Chemical Co (St Louis, MO). p-Nitroanisole was purchased from Aldrich Chemical Co (Milwaukee, WI), and recrystallized from hexane. All other chemicals were of the highest purity commercially available, and solvents were redistilled in glass. Homogenization and Fractionation Tobacco hornworms were reared on artificial diet . Fifth-instar larvae were taken on the first day of ”wandering” [17,18]and the midguts dissected as described previously . All of the following operations were carried out at 5°C. The carefully cleaned and rinsed guts (inside and outside) were minced with scissors and homogenized in 6 ml of buffer A (10 mh4 TRISHC1, pH 7.4, 300 mM sucrose, 1 mM EDTA) per gram gut tissue in a Ten ‘The term p-nitroanisole 0-demethylase is used in this paper for the total 0-demethylation activity measured with p-nitroanisole as substrate and does not imply the involvement of only one cytochrome P-450monooxygenase. ** Mention of a company name or proprietary product does not constitute an endorsement by the U.S. Department of Agriculture. +Abbreviations:bovine serum albumin = BSA; y intercept of linear regression line = b; slope of linear regression line = m; correlation coefficient = r; time integral of the squared angular velocity, W = 981 x 60 (g min)/raV. Microsomal Marker Enzymes 313 Broeck homogenizer (average clearance 125 pm) cooled in ice. Homogenization was initially performed by hand until the tissue was broken up and suspended in the buffer and subsequently with a motor-driven pestle (two strokes at approximately 650 rpm). Homogenates were fractionated as shown in Figure 1. The duration of the initial centrifugations was reduced to a minimum by increasing the speed , and some centrifugations consisted only of acceleration and deceleration with no plateau phase in between. Each centrifugation is described by the averaged g force and the time integral of the squared angular velocity W (given in brackets; Fig. 1). Centrifugations were carried out in a Beckman model L Ultracentrifuge with a Rotor Type 30 (Beckman Instruments, Inc, Palo Alto, CA) at 5°C. The W values were calculated by integration, based on the linear acceleration and deceleration curves obtained and the plateau phases, where applicable. An intermediate fraction (containing residual mitochondria as well as microsomes) was removed from the postmitochondrial supernatant prior to the high-speed sedimentation of the bulk of the microsomes. Mitochondria1 and microsomal pellets were washed by gentle mixing with the indicated volumes of buffer A in glass-Teflon homogenizers (100-200 pm clearance). The supernatants from washings of mitochondria were not subjected to further fractionation, and the supernatant from the washing of the microsomes was not combined with the original postmicrosomal supernatant. HOMOGENATE I accelerated to 3,0009 12.4 x lO'rad'sec-'] rehomogenized in 3 vols buffer A I centrifuged as above CRUDE.NUCLEAR FRACTION Repeated one or two more times I Supernatant I accelerated to 20,OOOg 14.0 x 108rad'sec-'] Pellet resuspended in 3 vols buffer A Supernatant accelerated to 13,0009 [2.0 x 108rad'sec-'] centrifuged at 28.OOOg 19.0x 1OBrad2sec-'] I MITOCHONDRIA Supernatant 1 INTERMEDIATE FRACTION Supernatant I centrifuged at 8O.OOOg 13.6 x 10i0rad2sec-'] I POSTMICROSOMAL Pellet resuspended In 3 vols byffer A SUPERNATANT centrifuged as above A MICROSOMES Supernatant Fig. 1. Flow diagram for fractionating M sexta midgut homogenates. 314 Weirich and Adams For use in enzyme assays and protein determinations , all particulate fractions were suspended in buffer A by gentle mixing in glass-Teflon homogenizers. Enzyme Assays Glucose-6-phosphatase (D-glucose-6-phosphate phosphohydrolase; EC 220.127.116.11) activity was determined according to Baginski et a1  and phosphodiesterase I (oligonucleate 5'-nucleotidohydrolase; EC 18.104.22.168.) activity according to Touster et a1 . Initial tests showed that the phosphodiesterase I activities could be approximately doubled by the addition of 25 mM MgC12. Therefore, all subsequent assays were performed with 15 mM MgC12. Glucose-6-phosphatase and phosphodiesterase activities were determined immediately after preparation of the subcellular fractions. For the glucose-6phosphatase assays, 0.2-2.0 mg protein were incubated in a total volume of 0.8 ml; for the phosphodiesterase assays, 2.0-7.0 mg protein were used in 1.0 ml of incubation mixture. Both assays were conducted at 30°C. For most fractions, the reaction was linear for the first 60 min of incubation. The NADPH-cytochrome c reductase (NADPH:ferricytochrome oxidoreductase; EC 22.214.171.124) assay was adapted from Sottocasa et a1 . In addition to the subcellular preparation (0.1-1.0 mg protein), the assay mixtures contained 50 pM cytochrome c, 0.33 mM KCN, 48 mM potassium phosphate, pH 7.8, 0.33 mM TRIS, and 1.0 mM EDTA. The total volume was 3.0 ml. The reactions were started by the addition of NADPH (final concentration 0.1 mM). The reduction of cytochrome c at 30°C was monitored by recording the increase in absorbancy at 550 nm with the isosbestic point for reduced and oxidized cytochrome c at 541 nm as reference. All measurements were made with an Aminco DW-2a TJViVIS Spectrophotometer (American Instrument Co. Urbana, IL). Initial rates were determined and corrected for the absorbancy change recorded prior to the addition of NADPH less than 10% of the total). Enzyme activities were calculated using 18.5 mM- cm-' as absorption coefficient for reduced cytochrome c , which was verified for our experimental conditions. The rates were proportional to the amounts of mitochondrial or microsomal protein added to the incubation mixtures, and in mixed incubations the activities of mitochondria and microsomes were additive. p-Nitroanisole 0-demethylase (p-nitroanisole monooxygenase (0-demethylating); not yet registered) activity was determined at 30°C by a modification of the method of Hansen and Hodgson .In addition to the buffer components, the reaction mixtures contained an NADPH-generating system, consisting of 0.5 mM NADP+, 2.5 mM glucose 6-phosphate, and 0.25 units of glucose-6-phosphate dehydrogenase per ml, and the subcellular fraction (0.3-5.0 mg protein). Total volume was 3.0 ml. After 5 min of preincubation at 30T, the reactions were initiated by the addition of 20 pl of a 75 mM pnitroanisole solution in methanol (0.5 mM in the incubation mixture). The reactions were terminated by addition of 1ml 1.0 N HCl and the reaction products were transferred to 0.5 N NaOH via chloroform partitions. Absorbancies were measured at 400 nm (dual wavelength mode with 500 nm as reference), and p-nitrophenol standards, carried through the same partition procedure, were used to determine the product concentrations. i Microsomal Marker Enzymes 315 Preliminary assays were conducted in three different buffers: 33 mM potassium phosphate, 3 mM TRIS, pH 7.8,1 mM EDTA, and 300 mM sucrose (buffer B); 37 mM TRIS-HCI, pH 7.9, 0.3 mM EDTA, 5 mM MgC12, 3.3 mg BSA/ml, and either 100 mM sucrose (buffer C) or 300 mM sucrose (buffer D). Mitochondria1 as well as microsomal activities were lowest when assayed in isotonic phosphate-sucrose buffer (B), and highest when assayed in isotonic TRIS-sucrose buffer (D). Buffer D was used for all subsequent assays. The demethylation reaction was linear for 15 min, and the rate decreased slightly between 15 and 30 min of incubation. The amount of product was proportional to the amount of protein in both 15- and 30-min incubations of reaction mixtures containing up to 10 mg mitochondrial protein or 4 mg microsomal protein. In mixed incubations of mitochondria and microsomes, the yield was additive up to 5 mg mitochondrial protein plus 1.5 mg microsoma1 protein. In all assays of mitochondria or microsomes the protein content was within these limits. The NADPH-cytochrome c reductase proved to be a very stable enzyme. Its activity did not change for 48 h after isolation of the subcellular fractions, provided they were kept on ice. The 0-demethylase activity of microsomal suspensions was stable for 24 h, but the activity of mitochondrial suspensions decreased substantially. Therefore, the O-demethylase activity was routinely determined in fresh subcellular preparations, and the NADPH-cytochrome c reductase activity in preparations stored overnight on ice. Electron Microscopy Aliquots of mitochondrial and microsomal suspensions were centrifuged in small tubes to form very small pellets (200-300 pg protein). The pellets were fixed in 3% glutaraldehyde in 10 mh4 HEPES buffer (containing 300 mM sucrose and 1 mh4 EDTA), pH 7.40, for 16 h at 4°C. The pellets were rinsed in HEPES buffer three times and postfixed in 1%osmium tetroxide in HEPES buffer for 1h at 4°C. After two rinses in HEPES buffer, pellets were dehydrated in an ethanol series and then infiltrated and embedded in DER 324  (DER 324 is the replacement for DER 334, which is no longer available). Sections were obtained with an LKB Ultratome 111, stained with alcoholic uranyl acetate and lead citrate , and examined in a Philips 400 T electron microscope. RESULTS Glucose-6-Phosphataseand Phosphodiesterase I In contrast to the situation in rat liver [1,3], microsomes from M sextu midgut did not exhibit the highest glucose-6-phosphatase activity (Fig. 2). In fact, the microsomes had the lowest activity. Although some enzyme activity was associated with the particulate fractions, the highest activity (both specific and total) was found in the supernatant. The specific activity of phosphodiesterase I was highest in the microsomes; the intermediate fraction had somewhat lower activity. The crude nuclear fraction, mitochondria, and postmicrosomal supernatant had similar, low 316 Weirich and Adams Nucl 0 25 Mit I M 50 75 100 0 0 25 Mit I M 50 slo 75 Parcent 01 Total Prolsh 25 50 75 100 p-Nitroaniade O-demethy(.m NADFltcytochome c reductare Nucl Mil I M Nucl SUP Mil I M Nucl 100 0 25 50 SUD 75 100 Percent 01 Total Proteh Fig. 2. Distribution of glucose+-phosphatase, phosphodiesterase I, NADPH-cytochrome c reductase, and p-nitroanisole 0-demethylase i n midgut fractions of M sexta. Enzyme activities are shown on the ordinate as relative specific activities ((total activity in fraction)/(total activity in all fractions))/((total protein in fraction)/(total protein in all fractions)), a measure for the degree of enrichment in each fraction . The percentage of the total protein recovered in each fraction is shown on the abscissa, and the areas represent the total enzyme activities for each fraction. The fractions are (from left to right) crude nuclear (Nucl), mitochondria (Mit), intermediate (I), microsomes (M), and postmicrosomal supernatant (Sup). The sums of the 83.9, 36.15 f 5.44, 710.5 f 15.1, and 1.34 f 0.20 nmole/ activities in all fractions were 293.3 midmidgut (means SD of 2 experiments) for glucose-6-phosphatase, phosphodiesterase I, NADPH-cytochrome c reductase, and p-nitroanisole 0-demethylase, respectively. Data for glucose-6-phosphatase were corrected for a slow release of inorganic phosphate occurring in incubations without added Glc-6-P. specific activities. Thus, in the M sexta midgut, phosphodiesterase I exists in particle-bound as well as in soluble form. NADPH-Cytochrome c Reductase and p-Nitroanisole 0-Demethylase The specific activity of NADPH-cytochrome c reductase was very high in the microsomes, low in the crude nuclear fraction and in the mitochondria, intermediate in the intermediate fraction, and marginal in the supernatant (Fig. 2). p-Nitroanisole 0-demethylase was similarly distributed in the subcellular fractions, although it was not detectable in the supernatant. Microsomal Marker Enzymes 317 For both enzymes, the total activity in the nuclear fraction was slightly higher than that in the microsomes. Apparently, a large proportion of the endoplasmic reticulum cosedimented with the nuclei and was retained by this fraction. Sedimentation of microsomal enzyme activities at unusually low g forces has been reported for other insect tissues [27l, and a similar tendency for cosedimentation has been observed in rat liver fractionations . This problem may have been aggravated in our preparations by the muscle component of the midgut. Moreover, the homogenization of the midguts was very gentle and the isolation procedure was designed for optimum purity, not maximum yield of microsomes. While it cannot be ruled out that other fractions contain some NADPH-cytochrome c reductase andl or 0-demethylase of their own, the pattern of specific activities suggested that in the larval midgut of M sextu, NADPH-cytochrome c reductase and 0demethylase are enzymes primarily associated with the endoplasmic reticulum and that the activities found in other subcellular fractions are mostly due to carryover or entrapment of microsomes. This conclusion was supported by electron-microscopic examinations of mitochondrial and microsomal pellets. As shown in Figure 3, the microsomal pellets consisted mostly of smooth membrane vesicles and contained a few fragments of microvilli. The mitochondrial pellets contained predominantly mitochondria of various sizes and degrees of structural integrity, low concentrations of autophagic vacuoles, multilamellar bodies or lamellar lysosomal bodies, and a few multivesicular bodies, secretory granules of the Golgi, and smooth membrane vesicles. The biochemical data (6% microsomes) were in good agreement with the ultrastructural observations. To determine whether both of these enzymes are equally reliable as microsomal markers, we analyzed the results of three experiments. In these three experiments, the specific activities of NADPH-cytochrome c reductase in the microsomal fraction ranged from 139 to 177 nmollminlmg protein and those of 0-demethylase from 255 to 379 pmollminlmg. The variations in the two specific activities were not correlated, but for each experiment the distributions of NADPH-cytochrome c reductase and 0-demethylase among the subcellular fractions were identical. The specific activities declined by identical proportions with each washing of the mitochondrial pellets, and the enzyme activities in the supernatants were equal to the amounts removed from the mitochondria. Relative specific activities for mitochondrial pellets of varying purity ( l x , 2 x , 3x washed) were highly correlated (Fig. 4).Linear regression analysis yielded a correlation coefficient (r) of 0.994, a slope (m) of 1.01, and a y intercept (b) of -1.41. The results of a similar analysis of all other fractions and of whole homogenates were r = 0.983, m = 1.03, and b = 1.06. Thus, both enzyme activities seem to be equally reliable indicators of microsomal content. DISCUSSION Two of the four microsomal marker enzymes commonly used for mammalian tissues, glucose-6-phosphatase and phosphodiesterase I, were not predominantly associated with any one subcellular fraction of M sexta mid- 318 Weirich and Adams Fig. 3. Electron micrographs of sections of mitochondria1 (A) and microsomal (B) pellets of M sexta midgut prepared as described in Materials and Methods. NADPH-cytochrome c reductase activity indicated a 6% microsome content in the mitochondria1 preparation. Microsomal Marker Enzymes 319 .Ot - 0' 10 20 ~ O d T O l T 8 C r e d r c t a s e Fig. 4. Correlation between relative specific activities of p-nitroanisole 0-demethylase and NADPH-cytochrome c reductase in mitochondria1 preparations from M sexta midgut. The activities are shown as percentages of the relative specific activity found in rnicrosornes. gut. The bulk of the glucose-6-phosphatase activity was found in the postmicrosomal supernatant. Storey and Bailey  reported the solubilization of glucose-6-phosphatase of cockroach fat body microsomes in 100 mM TRISHC1 buffer, pH 8.0, containing 15 mM mercaptoethanol. It is uncertain whether solubilization was responsible for the glucose-6-phosphatase distribution we observed in the midgut fractions. However, if this enzyme is easily extracted from insect microsomes, it is not a suitable marker enzyme. The highest relative specific activity of phosphodiesterase I was found in the microsomes, but it was only half that observed for NADPH-cytochrome c reductase or p-nitroanisole 0-demethylase. Furthermore, the substantial soluble component precludes the use of phosphodiesterase I as a microsomal marker enzyme. The other two enzymes, NADPH-cytochrome c reductase and p-nitroanisole 0-demethylase, were most abundant in microsomes (Fig. 2). In fact, their relative specific activities, 6.2 and 6.5, respectively, were higher than those previously reported for either mammalian [3,4]or insect tissues . The NADPH-cytochrome c reductase activities were several hundred times as high as the p-nitroanisole 0-demethylase activities. Thus, NADPH-cytochrome c reductase can be measured more easily and more economically than 0-demethylase. The greater stability also makes the reductase the enzyme of choice. The p-nitroanisole 0-demethylase activities reported here are similar to those reported by Tate et a1  for the same tissue and species, but the NADPH-cytochrome c reductase activities we report are substantially higher. The high relative specific activities of NADPH-cytochrome c reductase and pnitroanisole 0-demethylase in microsomes, the close correlation of their activities in subcellular fractions, and the good agreement between biochemical and electron-microscopic observations support the conclusion that these two enzymes are reliable microsomal marker enzymes in the midgut of M sextu and do not occur at substantial levels in other subcellular components. 320 Weirich and Adam ACKNOWLEDGMENTS We thaAk Kathryn J. Green and Daniel J. Scholfield for their competent technical assistance, Dr Jan P. Kochansky for purifying p-nitroanisole, and Drs Jeffrey R. Aldrich, Mark F. Feldlaufer, and Raziel S. Hakim for valuable suggestions during the preparation of the manuscript. LITERATURE CITED 1. De Duve C, Pressman BC, Gianetto R, Wattiaux R, Appelmans F: Tissue fractionation studies. 6. Intracellular distribution patterns of enzymes in rat-liver tissue. Biochem J 60, 604 (1955). 2. Beaufay H, Amar-Costesec A: Cell fractionation techniques. In: Methods in Membrane Biology. Korn ED, ed. Plenum Press, New York, Vol6, pp 1-100 (1976). 3. Touster 0, Aronson NN Jr, Dulaney JT, Hendrickson H: Isolation of rat liver plasma membranes. Use of nucleotide pyrophosphatase and phosphodiesterase I as marker enzymes. J Cell Biol47, 604 (1970). 4. Phillips AH, Langdon RG: Hepatic triphosphopyridine nucleotide-cytochrome c reductase: Isolation, characterization, and kinetic studies. J Biol Chem 237, 2652 (1962). 5. Kulkarni AP, Smith E, Hodgson E: Occurrence and characterization of microsomal cytochrome P450 in several vertebrate and insect species. Comp Biochem Physiol 54B, 509 (1976). 6. Lu AYH: Liver microsomal drug-metabolizing enzyme system: Functional components and their properties. Fed Proc 35, 2460 (1976). 7. Sugiyama T, Miki N, Yamano T: NADH- and NADPH-dependent reconstituted p-nitroanisole 0-demethylation system containing cytochrome P-450 with high affinity for cytochrome bg. J Biochem 87, 1457 (1980). 8. Hansen LG, Hodgson E: Biochemical characteristics of insect microsomes. Biochem Pharmacol20, 1569 (1971). 9. Wikinson CF, Brattsten LB: Microsomal drug metabolizing enzymes in insects. Drug Metab Rev 1, 153 (1972). 10. Agosin M, Perry AS: Microsomal mixed-function oxidases. In: The Physiology of Insecta. Rockstein M, ed. Academic Press, New York, 2nd Ed, Vol5, pp 537-596 (1974). 11. Forgash AJ, Ahmad S: Hydroxylation and demethylation by gut microsomes of gypsy moth larvae. Int J Biochem 5, 11 (1974). 12. Crankshaw DL, Hetnarski HK, Wilkinson CF: Microsomal NADPH-cytochrome c reductase from the midgut of the southern armyworm (Spodopteru eriduniu). Insect Biochem 9, 43 (1979). 13. Kulkoski JA, Weber JL, Ghazarian JG: NADPH-cytochrome c reductase in outer membrane of kidney mitochondria. Arch Biochem Biophys 192, 539 (1979). 14. Howell KE, It0 A, Palade GE: Endoplasmic reticulum marker enzymes in Golgi fractionswhat does this mean? J Cell Biol 79, 581 (1978). 15. Ito A, Palade GE: Presence of NADPH-cytochrome P-450 reductase in rat liver Golgi membranes. J Cell Biol 79, 590 (1978). 16. Hoffman JD, Lawson FR, Yamamoto R: Tobacco hornworms. In: Insect Colonization and Mass Production. Smith CN, ed. Academic Press, New York, pp 479-486 (1966). 17. Truman JW, Riddiford LM: Physiology of insect rhythms. 111. The temporal organization of the endocrine events underlying pupation of the tobacco hornworm. J Exp Biol60, 371 (1974). 18, Nijhout HF, Williams CM: Control of moulting and metamorphosis in the tobacco hornworm, Munducu sextu (L.): Growth of the last-instar larva and the decision to pupate. J Exp Biol61,481(1974L 19. Mayer RT, Svoboda JA, Weirich GF: Ecdysone 20hydroxylase in midgut mitochondria of Munducu sexfu (L.).Hoppe-Seyler’s Z Physiol Chem 359, 1247 (1978). 20. Keeley LL: Characterization of insect fat body mitochondria isolated by a rapid procedure. Comp Biochem Physiol46B, 147 (1973). I . , Microsomal Marker Enzymes 321 21. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 193,265 (1951). 22. Baginski ES, Foa PP, Zak B: Glucose-6-phosphatase. In: Methods of Enzymatic Analysis. Bergmeyer HU, ed. Academic Press, New York, 2nd Engl Ed, Vol2, pp 876-880 (1974). 23. Sottocasa GL, Kuylenstierna B, Emster L, Bergstrand A: An electron-transport system associated with the outer membrane of liver mitochondria. J Cell Biol32, 415 (1967). 24. Ernster L: DT diaphorase. In:Methods in Enzymology. Estabrook RW, Pullman ME, eds. Academic Press, New York, Vol10, pp 309-317 (1967). 25. Wmbom WB: Dow epoxy resin with triallyl cyanurate and similarly modified Araldite and Maraglas mixtures, as embedding media for electron microscopy. Stain Techno1 40, 227 (1965). 26. Reynolds ES: The use of lead citrate at high pH as an electron opaque stain in electron microscopy. J Cell Bioll7, 208 (1963). 27. Wilkinson CF: The use of insect subcellular components for studying the metabolism of xenobiotics. In: Xenobiotic Metabolism: In Vitro Methods. Paulson GD, Frear DS, Marks El', eds. ACS Symposium Series 97, American Chemical Society, Washington, pp 249-284 (1979). 28. Lewis JA, Tata JR: A rapidly sedimenting fraction of rat liver endoplasmic reticulum. J Cell Sci 23, 447 (1973). 29. Storey KB, Bailey E: The intracellular distribution of enzymes of carbohydrate degradation in the fat body of the adult male cockroach. Insect Biochem 8,73 (1978). 30. Tate LG, Nakat SS, Hodgson E: Comparison of detoxification activity in midgut and fat body during fifth instar development of the tobacco hornworm, Manduca sexta. Comp Biochem Physiol72C, 75 (1982).