Phenol ╬▓-glucosyltransferase and ╬▓-glucosidase activities in the tobacco hornworm larva Manduca sexta L.Properties and tissue localizationкод для вставкиСкачать
Archives of Insect Biochemistry and Physiology 21 :207-224 (1992) Phenol p-Glucosyltransferase and p-Glucosidase Activities in the Tobacco Hornworm Larva Manduca sexta (1.): Properties and Tissue localization Saad A. Ahmad and Theodore L. Hopkins Department of Entomology, Kansas State Uniuersify, Manhattan, Kansas Phenol p-glucosyltransferase (PGT; EC 126.96.36.199) was studied in feeding fifth stadium larvae of the tobacco hornworm, Manduca sexta, using reversed-phase HPLC with absorbance detection to separate and quantify both the model substrate, p-nitrophenol (PNP), and the product, p-nitrophenyl f3-D-glucopyranoside (PNP-Clc). About 90% of total PGT activity in tissue homogenates was associated with the particulate fraction (15,00(%),with the remainder in the microsomal fraction. PGT activity was observed in all tissues, with highest activities in the labial gland and fat body. Appreciable activity occurred in rnidgut and hindgut tissue but none was found in hernolymph. PGT activity was also observed in eggs and larval fat body at different times during development. Activity was optimal at pH 7.5-9 and was highest with UDP-Glc as a glucose donor. However, appreciable PCT activity was observed with dTDP-Clc or GDP-Glc in place of UDP-Glc. The divalent cations Ca2+, Co2+, Mg", and Mn2' stimulated activity, whereas Zn2+ and Hg2+, as well as pretreatment with the detergent Triton X-100, were inhibitory. Endogenous P-glucosidase in the PGT-enrichedfractions, especially from the midgut, antagonized the pglucosylation process, and interference was minimized with higher pH and the addition of D-gluconic acid lactone to the incubation mixtures. The possible role of transglucosylation in the detoxication of phenolic xenobiotics and biotransformation of endogenous phenolic compounds in insects i s discussed. Comparisons of PCT with some glycosyltransferases involved in endogenous and xenobiotic conjugation in insects and other organisms are reviewed. Q 1992 Wiley-Liss, Inc. Key words: conjugation, glucosylation, p-glucosides, UDP-glucosyltransferase, insects Acknowledgments: The authors thank Dr. Karl Kramer for supplying the experimental insects and for helpful suggestions on the manuscript. This paper is contribution no. 92-543-Jfrom the Kansas Agricultural Experiment Station, Manhattan, Kansas. This research was supported in part by KAES Project 548. Received April 29, 1992; accepted July 28, 1992. Address reprint requests to T.L. Hopkins, Department of Entomology, Kansas State University, Manhattan, KS 66506. 0 1992 Wiley-Liss, Inc. 208 Ahmad and Hopkins INTRODUCTION The ability of insects to synthesize p-glucoside conjugates of natural and synthetic phenolic compounds is well established [1,2]. P-Glucosylation requires glucose donors in the form of specific high-energy glucosyl nucleotides and is catalyzed by the enzyme phenol P-glucosyltransferase. Conjugation with glucose is generally considered to be a mechanism of detoxification by insects, resulting in increased water solubility and more rapid excretion of xenobiotics and allelochemicals. 0-Glucosides are often less toxic than the parent aglucones, probably because of rapid elimination and the masking of functional groups . Evidence is increasing that, in addition to detoxication, P-glucoside formation in insects serves as a means to store and regulate levels of endogenous phenolic metabolites [PB]and plant allelochemicals [9,10; unpublished data] for physiological processes. Therefore, P-glucosylation appears to be an important biochemical reaction in insects, allowing them to cope with potentially harmful compounds in the diet, as well as accommodating metabolic needs during growth and development. However, relatively little is known about the properties and functional role(s) of the glucosyltransferases that catalyze glucosylation reactions in insects . In this study, we have examined the localization and properties of PGT" in tissues of feeding fifth instar tobacco hornworm larvae, Munducu sexfa (L.). A sensitive method utilizing HPLC has been developed to measure PGT activity by specific quantitation of product formation. MATERIALS AND METHODS Insects Larvae of M . sextu reared at 27 f 1°C with a photoperiod of 16L:8D [ll] were supplied by Dx. Karl J. Kramer, U.S. Grain Marketing Research Laboratory, Agricultural Research Service, U.S.D.A., Manhattan, KS. Chemicals P-Glucosidase (almond) and all biochemicals were obtained from Sigma Chemical Co., St. Louis, MO, except 7-methoxycoumarin, which was from Aldrich Chemical Co., Milwaukee, WI. HPLC grade acetonitrileand methanol were from Fisher Scientific Co., St. Louis, MO. Tissue Preparation and Subcellular Fractions Three-day-old fifth instars were immobilized in crushed ice, the anal horn was cut, and hemolymph was collected in ice-cold 1.5 ml plastic centrifuge vials containing a few crystals of diethyldithiocarbamic acid to inhibit phenoloxidases. The hemolymph was assayed immediately or stored at -20°C for 2-3 h. The bled larvae were dissected dorsally and tissues in situ were washed with distilled water and then with 0.1 M sodium phosphate, pH 7.5, or with *Abbreviations used: PGT = phenol P-glucosyltransferase; PIPES = piperazine-N,N-bis-(2-ethanesulfonic acid) monosodium; PNP = p-nitrophenol; PNP-Clc = pnitrophenyl-P-D-glucopyranoside. Phenol 6-Glucosyltransferases in M. sexta 209 0.125 M KC1 (when the effect of pH was studied). The labial glands and fat body were isolated. The mid- and hindguts (including Malpighian tubules) were separated, cleaned of fat body, dissected longitudinally, and washed with buffer or KC1 solution to remove the lumen contents, The skeletal muscle and epidermis were cleaned of other tissues and pieces of epidermis-free muscle were isolated. Epidermis with some muscle was collected by scraping the inner surface of abdominal cuticle. Each tissue was kept in buffer or KCl solution and homogenized in the proper medium in a glass tissue grinder for 1min. All steps were carried out at 0-5°C. The homogenates were centrifuged at 15,0008 for 30 rnin at 5"C, and the supernatants were discarded unless stated otherwise. The pellets were washed twice with medium without disruption and then were resuspended for use as enzyme sources. To study the subcellular distribution of PGT, the crude tissue hornogenate in 60 mM phosphate buffer (pH7) was centrifuged at 600g for 15rnin to obtain a pellet of nuclear and cell debris. The supernatant was centrifuged at 10,OOOg for 20 min to obtain the mitochondria1 pellet, followed by centrifugation at 110,OOOg for 90 min to obtain a microsomal pellet and cytosol. Each pellet was resuspended in buffer in a glass tissue grinder. To study the activity of PGT during development, 15,000.g pellets and supernatants were prepared from eggs (1-2 days before hatching) and fat body from fourth and fifth instars (at ecdysis and 1 and 4 days old), Enzyme Assays All PGT assays were performed with same-day enzyme preparations, and the typical incubation mixture contained 5 mM PNP, 5 mM UDP-Glc, 25 mM MgC12,5mM D-gluconic acid lactone, and enzyme preparation (equivalent to 0.06-0.3 mg protein) in 0.1 M phosphate buffer (pH 7.5) in a final volume of 0.15 ml. The reaction was started by the addition of the enzyme source to the incubation mixture at 30" C in a shaker water bath. The mixture was vortexed and returned to the water bath. After 30 min, the reaction was stopped by the addition of 75 PI of ice-cold 9% perchloric acid, vortexed, and held 15 min on crushed ice. The mixtures were centrifuged to remove denatured protein, and the supernatant was held at 6 5 ° C until analyzed for PNP-Glc by HPLC. In control incubations in which UDP-Glc was omitted, no PGT activity was observed. Antagonistic P-glucosidase activity, particularly in midgut tissues, was suppressed by using the inhibitor, D-gluconic acid lactone, and an alkaline buffer (see Results). The amount of PNP that disappeared was equal to the amount of PNP-Glc formed in the PGT complete incubations, indicating that PNP underwent P-glucosylation only. The identity of the products was confirmed by comparing retention times with standards and by hydrolysis of PNP-Glc formed by commercial P-glucosidase. To study the effect of glucosyl nucleotides other than UDP-Glc, 5 mM of ADP-Glc, CDP-Glc, GDP-Glc, dTDP-Glc, UDP-GlcNAc, or UDP-GlcUA replaced UDP-Glc in the incubations. When the effect of Triton X-100 was studied, it was incubated with the enzyme source in the incubation mixture for 15 min and the reaction was started by addition of UDP-Glc. To study the effect of pH on PGT activity, PNP, UDP-Glc, and MgC12 were prepared in 210 Ahmad and Hopkins appropriate buffers and added in incubation mixtures to yield 0.1 M final buffer Concentration. PGlucosidase activity was measured with PNP-Glc as the substrate. The incubation mixture contained, in a final volume of 0.15 ml, the enzyme preparation (0.05-0.1 mg protein) in 0.125 M KCl and variable substrate concentrations and buffers, as presented in Results. No PGT activity interfered with the P-glucosidase activity when UDP-Glc was omitted from the incubations. Enzyme activity was determined by the amounts of PNP released as quantified by HPLC. Cytochrome P-450-dependent monooxygenase activity was measured with 7-methoxycoumarin as the substrate. The incubation mixture contained 1mM NADPH, 50 p,M 7-methoxycoumarin, and midgut 15,0009 pellet or supernatant (equivalent to 0.1 mg protein) in 0.1 M phosphate buffer at pH 7.5. The reaction was started by addition of the substrate in methanol, the final concentration of which did not exceed 3% in the incubation mixture. After 1 h incubation at 30T, the reaction was stopped by addition of 4 volumes of cold methanol. The mixture was vortexed, held 15min at - 20" C, centrifuged, and the supernatant was analyzed for umbelliferone by HPLC, HPLC The separation and quantitation of PNP, PNP-Glc, 7-methoxycoumarin, and umbelliferone were achieved using reversed-phase HPLC with a C18 5 pm spherical particle 4.6 x 250 mm column with a flow rate of 1 mVmin at 30°C and a variable wavelength absorbance detector. The mobile phase contained either 20% methanol or 10%acetonitrile in 0.1 M sodium phosphate buffer, pH 3, for PNP and PNP-Glc separation or 30% methanol in the same buffer for separation of 7-methoxycoumarin and umbelliferone. The eluate was monitored at 303 nm for PNP and PNP-Glc, and at 321 nm for 7methoxycoumarin and umbelliferone. Quantitation of products was determined by comparing their peak areas with those of known concentrations of standard compounds. RESULTS SubcellularDistribution PGT activity was found to be mainly associated with the particulate fraction, with about 90% of total activity recovered in the 15,0008pellet from all tissues (Fig. 1).Subcellular fractionation of the fat body by differential centrifugation revealed that PGT activity was equally distributed between the nuclear (600g) and mitochondria1 (10,OOOg)pellets (Fig. 1).The remaining 10% of activity in the 15,0008supernatant was entirely microsomalbecause the cytosol obtained from 110,OOOg centrifugation was devoid of activity. To determine if the PGT in the 15,0009 pellet was a particulate enzyme and not due to early sedimentation of microsomes, we compared the subcellular distribution of PGT with a microsomal marker enzyme, 7-methoxycoumarin 0-demethylase (a cytochrome P-450-dependent monooxygenase), from the midgut. Midgut incubations showed that 64.3% of total 7-methoxycoumarin demethylase activity Phenol p-Glucosyltransferasesin M. sexta 100 A x 1- 1 5 , 0 0 0 a supernatant 0Microsokmes' 80 - eZa 600g pellet 659 10,0009 pellet EEd 15,OOOg pellet v x 4 ._ .-> 21 1 60 - 40 0 20 t 0 [ Fat body 1 Labia _1 land [ION 4k7-M€riOXYCOUMARIN -I 0- DEMETHYLATION Fraction Fig. 1 . Subcellular distribution of PGP activity and cytochrome P-450-dependent monooxygenase (7-methoxycoumarin 0-dernethylation) in tissues of fifth stadium larvae of Manduca sexta was recovered in the 15,OOOg supernatant, whereas the remaining activity was in the pellet (Fig. 1).In contrast, 87.4%of total PGT activity was found in the pellet and the remainder was in the supernatant. Because of enrichment of PGT in the particulate fraction, all further studies were carried out using the 15,OOOg pellet as the PGT source. Tissue Distribution All tissues except the hemolymph were active in the P-glucosylationof PNP (Fig. 2). However, a marked difference was observed in specific activity of PGT among tissues. The labial gland was the most active tissue, with a specific activity of 12.2 nmol PNP-Glc formed/min/mg protein, followed by fat body, 6.6; midgut, 4; hindgut, 3.1; epidermis with muscle, 0.4; and muscle, 0.2. Because of their higher activities, the labial gland and fat body were used in subsequent studies. Effect of pH PGT activity in labial gland was detectable at pH 5.5, increased sharply at pH 6, and reached a maximum at pH 7 . 5 9 (Fig. 3). Activity decreased steeply at higher pHs and was undetectable above pH 10. PGT from the fat body behaved similarly, except that it had a broader range of activity, with low glucosylation rates detected at pH 4.5 and 10.5 (Fig. 4). Buffers showed a difference in their influence on PGT activity from both the same and different tissues. PIPES and phosphate buffers gave maximum PGT activity from labial glands, whereas PIPES was slightly superior to 212 Ahrnad and Hopkins Tissue Fig. 2. PGT activity in tissues of fifth stadium larvae of Manduca sexta (means * SD, n = 3-6). phosphate in the case of fat body PGT. Tris-HClwas the poorest in the optimal pH range for both tissues. Highest activity in fat body was observed with glycine-NaOH buffer at pH 8.75-9. Such an effect of glycine buffer was not observed on PGT from labial glands; however, it was better than tris-HC1 at 2 4 a 6 10 12 PH Effect of pH o n PGT activity in preparations from labial gland of fifth stadium larvae (0) universal, (@I citrate, (A 1 PIPES, (A)phosphate, (0) tris-HCI, (m) glycine-NaOH, and (4)borate. Fig. 3. of Manducasexta. Buffers (0.1 M) were Phenol p-Clucosyltransferasesin M. sexfa 21 3 n .-cat Y x Pa 4 .-> r 'In .- -.-E -P 0 Q 0 .+ .- 'c E D E P 0 0 1/) 0- W GI fl n z 2 4 6 8 10 12 Fig. 4. Effect of pH on PCT activity in preparations from fat body of fifth stadium larvae of Manduca sexta. Buffers (0.1 M)were (a)acetate, (A)PIPES, (A)phosphate, (0)tris-HCI, and (W)glycine-NaOH. alkaline pH. Sodium phosphate buffer, pH 7.5, was used in further experiments because of its pKa2 of 7.2, which falls close to the optimal pH range of PGT. Interference by Endogenous P-Glucosidase During investigation of the tissue site of tyrosine P-glucoside biosynthesis in fifth stadium larvae of M. sexta, we observed that tyrosine glucoside in the hemolymph was hydrolyzed when incubated with crude homogenates of midgut at pH 7 (unpublished data). These homogenates also hydrolyzed esculin, a natural p-glucoside of 6,7-dihydroxycournarin, as well as PNP-Glc, PNP-a-glucoside, PNP phosphate, PNP sulfate, and PNP-N-acetyl glucosamine, but not PNP-glucuronide. Some inhibition of esculin hydrolysis was achieved when 10 mM gluconic acid lactone was included in the incubation. When a solution of esculin was perfused into the lumen of an isolated midgut with both ends ligated and bathed in Manducu saline, samples of the saline analyzed by HPLC 2 h later showed the presence of both esculin and its aglucone, esculetin. Those preliminary results drew attention to the likely presence of endogenous p-glucosidases, which may interfere with PGT assays by hydrolyzing the PNP-Glc formed. PNP-Glc hydrolysis was also observed with crude homogenates of labial gland and fat body, but to a lesser extent than with homogenates of the midgut. In addition, both the 15,0009 pellet and the supernatant of midgut homogenates hydrolyzed PNP-Glc. Washing the pellet twice by resuspension in buffer, centrifugation, and discarding the washes was unsuccessful in removing the hydrolytic activity. Two approaches 214 Ahmad and Hopkins 4 6 8 10 PH Fig. 5. The influence of pH on PNP-Clc hydrolysis by endogenous P-glucosidase in the 15,OOOg pellets from fat body and midgut of fifth stadium larvae of Manduca sexta. Buffers (0.1 M) were acetate for pH 5.0, phosphate for pH 7.5, tris for pH 8.5, and glycine for pH 8.75-9.5. Enzyme sources and substrate concentrations were midgut with (0) 3.33 mM PNP-Clc, ( A ) 0.25 mM PNP-Clc, [A)0.25 mM PNP-Glc 10 rnM D-gtuconic acid lactone; fat body (El)with 0.25 rnM PNP-CIC. + were studied to minimize P-glucosidase interference; (1)shifting the incubation buffer to a higher pH and (2) using the competitive inhibitor gluconic acid lactone. The influence of pH on endogenous P-glucosidase activity in the 15,OOOg pellet from midgut and fat body tissue is depicted in Figure 5. As the pH of incubation increased, the specific activity of p-glucosidase decreased. At an initial concentration of 3.33 mM PNP-Glc, P-glucosidase activity in the midgut preparation dropped from 3.54 nmol PNP-Glc hydrolyzed min/mg protein at pH 5 to 3.08 at pH 7.5 and then sharply to 0.33 at pH 8.5. However, glycine buffer at pH 9 was not effective in reducing hydrolytic activity (2.18 nmoVmin/mg). Lower P-glucosidase activities were obtained when midgut and fat body enzyme preparations were incubated with a 0.25 mM initial PNP-Glc concentration (equals 5% of initial PNP concentration in PGT assays and presumably equals the concentration of PNP-Glc formed when 0.1 mg labial gland pellet is incubated for 30 min in the complete PGT assays). P-Glucosidase specific activities decreased from 1nmol PNP-Glc hydrolyzed/midmg protein at pH 5 to 0.04 nmol PNP-Glc hydrolyzed/min/mg at pH 8.5. Glycine buffer at a higher pH range was also ineffective compared to the other buffers in suppressing p-glucosidase activity. The addition of 10 mM D-gluconic acid lactone to the midgut incubation mixture at pH 7.5 reduced P-glucosidase activity by about one half. D-Gluconicacid lactone inhibition of P-glucosidase from the midgut 15,0009 pellet and supernatant is shown in Figure 6. Percent inhibition was linear with Phenol p-Glucosyltransferasesin M. sexfa 215 100 15,OOOg pellet o 15,OOOg supernatant 801 .+ .n D-Gluconic Acid Lactone [ mM ] Fig. 6 , The inhibitory effect of D-gluconic acid lactone on PNP-Glc hydrolysis by endogenous P-glucosidase in midgut 15,OOOgfractions of (0) pellet and (0)supernatant from fifth stadium larvae of Manduca sexta. The initial PNP-Glc concentration was 3.33 mM. Buffer was 0.1 M sodium phosphate, pH 7.5. the concentration (mM) of the inhibitor. Fifty percent inhibition occurred at 7.58 and 9.92 mM for the supernatant and pellet, respectively. Addition of a high concentration of the inhibitor resulted in lowering the pH of the incubation mixture. Therefore, 5 mM D-gluconic acid lactone was added in the PGT assays at pH 7.5, effectively suppressing P-glucosidase activity to less than 50 pmol PNP-Glc hydrolyzed/min/rng protein. Effect of Divalent Cations The effect of various divalent cations on PGT activity is presented in Figure 7. Zn2+ was the most inhibitory, with PGT activities being 11.1,0.1, and 0% of the control value, whereas H$+ was the next most inhibitory, with activities 69, 7.1, and 0% of control at concentrations of 1, 5, and 25 mM, respectively. Co2+ and Mn enhanced activity at 1 and 5 mM, but the stimulatory effect of Co2+ decreased, whereas Mn2+ was inhibitory at 25 mM. Mg2+ and Ca2+ steadily increased PGT activity as the concentration was increased from 1to 25 mM with Ca2+ slightly more effective at all concentrations. No difference occurred in the pattern of stimulation between the sulfate and chloride salts of Mg. The Mg effect reached a plateau at concentrations between 15 and 45 mM, but PGT activity decreased slowly at higher concentrations (data not shown). Effect of Triton X-100 The non-ionic detergent, Triton X-100, was studied in regard to its effect on PGT activity (Fig. 8). It inhibited PGT at a final concentration as low as 0.07 216 Ahrnad and Hopkins 250 I 01 mM IEEl 5 mM L 0 150 W I 1004 n HZ' ' :Z Co2+ Mn2+ Cation Fig. 7. The influence of divalent cations on PNP glucosylation by PGT from labial glands of fifth stadium larvae of Manduca sexta. Ratio ( w/w ) Triton X-1 OO/Protein 0.0 0.0 0.4 0.2 1.2 0.8 0.4 0.6 1.6 0.8 1.0 2.0 1.2 2.4 1.4 Triton X-100 Conc. ( mg/ml ) Fig. 8. The effect of Triton X-100 on PNP glucosylation by PGT from labial glands of fifth stadium larvae of Manduca sexta. Phenol P-Glucosyltransferases in M. sexfa 217 7GDP-Glc &481 dTDP-Glc Hindgut Midgut Fat body Labial gland Tissue Fig. 9. Utilization of glucosyl nucleotides as glucose donors by PGT from tissues of fifth stadium larvae of Manduca sexta. mg/ml (or a ratio of detergent/protein of 0.113wiw). PGT activity was strongly inhibited at concentrations higher than 0.2 mg/ml. Effect of Glucose Donors Five glucosylated nucleotides were compared in relation to their preference as glucose donors in PGT mediated reactions (Fig. 9). UDP-Glc was the preferred glucose donor in all tissues. dTDP-Glc (a synthetic p-glucosyl nucleotide derivative of dTDP, unlike the other nucleotides that are glucosylated with a-linkages) was next in preference by PGT from all tissues, followed by GDP-Glc. Very low rates of PNP glucosylation in incubation mixtures of fat body (85and 13 pmollmidmg protein) and labial gland (40 and 51 pmollmidmg protein) were detected with ADP-Glc and CDP-Glc, respectively. When UDP-GlcUA was substituted for glucose donors in the labial gland incubations, the observed activity was 29 pmol PNP-glucuronide/min/mg protein. When the glucosyl nucleotides were individually included with UDP-Glc in the incubations, PGT activity was not affected compared with mixtures containing only UDP-Glc, except that UDP-GlcUA caused 13.6%inhibition. PGT Kinetics A plot of PGT specific activity from labial glands and fat body vs. specific activity/PNPconcentration (mM)is depicted in Figure 10. Apparent Kms were 0.14 and 0.1 mM PNP and Vmaxs were 15.2 and 7.9 nmol PNP-Glc formed/min/mg protein for the labial gland and fat body, respectively. PGT from both tissues obeyed simple Machaelis-Mentin kinetics, in that the K , did not change throughout the range of substrate concentration examined. 218 Ahmad and Hopkins 1 c1 c .W + e4 -' m1 -'.-Ec E 0 B I a z a 0 20 40 60 80 100 120 Fig. 10. Determination of kinetic parameters and specific activity (inset) of PGT towards PNP in the 15,OOOg pellets of (0) fat body and (a) labial gland of fifth stadium larvae of Manduca sexta. No inhibitory effect of the substrate on PGT was observed as the reaction approached zero-order kinetics at 1 mM PNP (Fig. 10, inset). To ensure a steady-state reaction, a final concentration of 5 mM PNP was employed in most experiments. Kinetic constants for UDP-Glc were not determined. PGT Activity During Larval Development Specific activities of PGT in the 15,OOOg pellet and supernatant of egg (1-2 days before hatching) homogenates were 2.5 and 0.3 nmol PNP-Glc formedmidmg protein, respectively. Activities in the 15,0009 pellet of fat body homogenate from fourth stadium feeding larvae and 0-, 1-, and 4-dayold fifth instars were 7.5,7.4,6.9, and 8.5 nmoYmidmg protein, respectively. DISCUSSION The subcellular distribution of PGT activity in different tissues of M . sexta showed that heavy cellular particles are the main source of the enzyme. About half of the enzyme activity was associated with the nuclear fraction, and most of the remaining activity in the supernatant sedimented with the mitochondria, Golgi apparatus, and lysosomes in fat body. About 10% of the total activity was present in the microsomes, whereas the soluble fraction (cytosol) was devoid of activity. The particulate fractions were also found to be the major or sole source of PGT activity in adult cockroaches, P e r ~ ~americana ~ ~ ~ e ~ a , and the locust, Schistocerca cancellafa . Our results differ from those of Mehendale and Dorough , who found that 1-naphthol glucosylation activity was contained in the 105,OOOg soluble fraction of an M . sexta larvae Phenol p-Clucosyltransferases in M, sexfa 21 9 whole body homogenate and that little activity was present in the 15,0009 pellet. However, in adult houseflies and P. americana, the microsomal fractions appear to have the highest PGT sources (14-17). Our studies of the distribution of PGT and 7-methoxycoumarin 0-demethylase, a microsomal marker enzyme, confirmed that PGT is indeed associated with the heavy cellular particles and is not a result of early sedimentation of microsomes. The 0-demethylation reaction was found to be a quantitative measure of microsoma1 contents in the centrifugal fractions of homogenates from the larval midgut of M . sexfa [ 181. Mannosyltransferases involved in the glycosylation of phospholipids and proteins have also been found in a high percentage in the particulate fractions from pupae of the stable fly, Stomoxys culcifrans . Although this conjugation reaction differs from p-glucosylation, the coexistence of both types of glycosyltransferasein cell structures and their stimulation by M g suggests a possible interrelationship between these enzymes. In mammals, a glucosyltransferase specificfor endogenous lipid acceptors has been localized in the plasma membrane and was stimulated by divalent cations [ZO]. Marked differences occurred among specificactivities of PGT from different tissues of M . sexta larvae. Activities from fat body and midgut supernatants were reported to be the same magnitude with 1-naphthol as a substrate [14,21].We have found that the fat body fraction with the major PGT activity was over 1.5 times as active as the midgut fraction in PNP glucosylation. The labial gland, however, had the highest PGT activity, whereas the epidermis and muscle were lowest. The specific activities of PGT in labial gland are the highest thus far reported for insects. It is not uncommon that some detoxication enzymes are present at varying levels of activity in the tissues of different insects, indicating the potential role of each tissue in the detoxication of harmful xenobiotics . The major site of P-450 monooxygenase was the fat body, midgut, or Malpighian tubules, depending upon the insect species. The common occurrence of PGT at different activity levels in M . sexta illustrates similarities among detoxication systems regarding tissue distribution and importance. The high activity of PGT compared to that of P-450 monooxygenase, 1.34 nmol/min/midgut , as well as its presence in early larval development stage, may reflect the role P-glucosylation plays in coping with phenolic xenobiotics. The role of PGT in the labial gland is uncertain. In saturniid moths, the silk gland secretes 0-glucosides of 3- and 5-hydroxyanthranilic acid or gentisic acid for silk tanning . In M. sexfa, the labial gland does not produce silk, except during the first larval stadium, but does produce a proteinaceous secretion . Because the midgut has high P-glucosidase activity as well as PGT, it could hydrolyze ingested plant glucosides and liberate the more toxic aglucones. We can speculate that the labial gland in this species actively participates in the detoxication of plant phenolics during the feeding stage. PGT from the labial gland accepted over 30 plant phenolics as substrates. However, only the fat body contained P-glucosyltransferases active towards endogenous phenolic precursors for cuticle tanning as well as towards plant phenolics (unpublished data). ' 220 Ahmad and Hopkins The effect of pH on PGT activity in the labial glands and fat body from M . sexta is in close agreement with results obtained for o-arninophenolglucosylation by P. americana  and generally agrees with other studies on insects in regard to a broad optimum pH range [14,15]. Phosphate and PIPES were generally the best buffers for optimal activity of PGT from salivary gland and fat body, although the glycine buffer also gave high activity for the latter. Tris-HCI gave lower PGT activity in the optimum pH range from both tissues, in contrast to studies of PGT activity from the midgut of M . sexta  and the hepatopancreas from the crustacean Homarus americanus . High activities of microsomal PGT with glycine buffers were also observed in the housefly, Musca domestica , and in P. americana . A similar pattern of effects of divalent cations on glucosyltransferases appears to be present in such distantly related organisms as plants, insects, and mammals. The stirnulatory effect of Mg2', Ca2+, and Mn2+ on PGT activity in this study is comparable with the activation of a specific glucosyltransferase from mouse liver plasma membranes  and PGT from papaya fruit . UDP-glucuronyltransferase from mammals, on the other hand, although stimulated by M$+, is inhibited by Ca2+ and MI-?-+ . Co2+ stimulated PGT activity in this study, whereas Mehendale and Dorough  found it inhibitory to glucosylation of 1-napthol by the insect species. Hg2+ and Zn2+ were inhibitory to glucosyltransferases from insects and plants and were mixed inhibitors (affectingboth K, and V,,,) to PGT from sugar beets . Detergents generally activate membrane-bound glycosylation enzymes, such as microsomal glucosyltransferase from houseflies [161; glucuronyltransferase from rat liver nuclear membrane , and rat intestinal and hepatic microsomes ; and UDP-G1c:sterol glucosyltransferasefrom white mustard seedlings . The activation is suggested to be due to membrane structure modification, exposure of otherwise nonfunctioning active sites, increased permeability to UDP-Glc, or solubilization of the enzyme, The non-ionic detergent, Triton X-100, however, inhibited PGT in this study. Detergents probably activate the enzyme from membrane preparations because glucuronyltransferase from intact, whole nuclei of rat liver was inhibited by digitonin but was activated in the nuclear membrane preparations . In this study PGT could have been in the membranes of intact cell organelles obtained by differential fractionation. Support for this suggestion comes from the work on PGT from white mustard seedlings . PGT is a membrane-bound enzyme occurring mainly in cellular structure sedimented at 3,000-15,OOOg. When the crude membrane preparations of the sediment were delipidated, Triton X-100 solubilized and activated the enzyme. Further studies on the exact localization of PGT in insects would enhance efforts for purification and characterization of the enzyme. The specificityof PGT toward the glucosylnucleotidesubstrate is apparently less restricted in M . sexfa than in other organisms studied so far. Nucleotides of the pyrimidines, uracil and thymine, and of the purine, guanine, served as glucose donors in this study and trace activity was observed with ADP-Glc and CDP-Glc. Although UDP-Glc was preferred, activity with dTDP-Glc was appreciable. The utilization of GDP-GLc as a glucose donor by PGT has not previously been reported [24,31]. For binding of the nucleotide to the enzyme, Phenol p-Glucosyltransferases in M. sexfa 221 the bases appear to need a carbonyl group on carbon 4 of the pyrimidine ring or on carbon 6 of the purine ring, in addition to a reduced nitrogen atom on carbon 3 of the pyrimidine and carbon 1of the purine. The same enzyme may transfer glucose from UDP-Glc, dTDP-Glc, and GDP-Glc because activity was not additive when glucose donating nucleotides were combined in the incubation mixtures. The specificity towards the sugar moiety was higher, however, because UDP-GlcUA and UDP-Glc NAc did not serve as donors and UDP-GlcUA inhibited PGT activity, probably due to competitive binding of the uridine moiety. Such inhibition was not observed in the analogous vertebrate glucuronyltransferase .It is interesting that PGT of insects was able to transfer glucose whether in the a- or p-form in the donor, and the product in both cases was a P-glucoside. Only the a-form of UDP-GlcUA served as donor in UDP-glucuronyltransferase reactions . Unlike crustacean PGT  and rat liver glucuronyltransferase , which deviated from simple Michaelis-Menten kinetics and had 2 apparent K, values, the M. sexta PGT preparation had one K, value for PNP. This could be due to the presence of only one enzyme that catalyzed the conjugation of PNP. The presence of p-glucosidases in the digestive system of insects is well documented . P-Glucosidases seem to have broad substrate specificityand are involved in the digestive hydrolysis of carbohydrates, as well as the hydrolysis of plant allelochemical glucosides. Aglucones often are more toxic than their glucosides, and their release catalyzed by P-glucosidase may be harmful to insects. Lindroth  concluded that lower p-glucosidase activity in the eastern tiger swallowtail subspecies, Papilio glaucus canadensis, relative to that in P. g. glaucus, contributed to the adaptation of the former subspecies to members of the Salicaceae that produce a variety of phenolic glucosides in substantial quantities. In M. sexta, p-glucosidase in the labial gland might be secreted in the saliva for hydrolysis of phenolic glucosides during feeding. We have also found that the labial gland of the black cutworm, Agrotis ipsilon, was active in PNP-Glc hydrolysis (unpublished data). The activity of pglucosidase from M. sexfa larval rnidgut was of a comparable magnitude to that from other lepidopteran species , but was much lower than that from larvae of the processionary moth Thaurnetopoca pityocampa . The 50% inhibition value of the P-glucosidase inhibitor D-gluconic acid lactone was more than 10 times higher for the M. sexta enzyme than for the fall armyworm enzyme . The antagonist enzymes, PGT and p-glucosidase, have not been studied simultaneously to determine more definitively the consequences of deglucosylation-glucosylation in conjunction with the net capacity of phytophagous insects to tolerate and adapt to host plants with defensive arsenals of phenolic glucosides. In this study, we have demonstrated the presence of both PGT and P-glucosidase in the same tissues and subcellular fractions. The coexistence of related antagonistic enzymes has been shown in other organisms. Rat liver microsomal preparations contained both P-glucuronidase and UDP-glucuronyltransferase . Glycosyltransferases and glycosidases active on the same type of secondary compounds have been demonstrated in the same plant tissues . UDP-G1c:sterol glucosyltransferase and steryl p-D-glucoside hydrolase were found to be of similar subcellular distribution in white 222 Ahmad and Hopkins mustard seedlings . Why these antagonistic enzymes are present in the same tissues or cellular membranes is not understood, but probably they regulate endogenous compound levels through glycosylation and hydrolysis. If so, then the enzymes themselves must be under tight regulation. Herbivorous insects are suitable models for exploring the regulation of phenolic conjugation as influenced by natural and modified diets. Regulation of endogenous glucosyl conjugates of tyrosine and catecholamines is important for cuticle tanning in insects [4,6], and the timing of glucosylation of tyrosine and hydrolysis of tyrosine glucoside is under hormonal control [36,37]. LITERATURE CITED 1. Smith JN: The comparative metabolism of xenobiotics. In: Advances in Comparative Physiology and Biochemistry. Lowenstein 0, ed. Academic Press, New York, vol 3, pp 173-232 (1968). 2. Ahmad S, Brattsten LB, Mullin CA, Yu SJ: Enzymes involved in the metabolism of plant allelochemics. In: Molecular Aspects of Insect-Plant Associations. Brattsten LB, Ahmad S, eds. Plenum Press, New York, pp 7S151 (1986). 3. Yu SJ: f3-Glucosidase in four phytophagous lepidoptera. Insect Biochem 19, 103 (1989). 4. Ahmed RF, Hopkins TL, Kramer KJ: Tyrosine and tyrosine glucoside titers in whole animals and tissue during development of the tobacco hornworm Manduca sexta (L.). Insect Biochem 23, 369 (1983). 5. Yago M, Sat0 H, Kawasaki H: The identification of N-acyldopamine glucosides in the left colleterial gland of the praying mantids, Mantis religiosa L., Statilia maculuta Thunberg and Tenodera angustipennis Saussure. Insect Biochem 14, 7 (1984). 6. Hopkins TL, Morgan TD, Kramer KJ: Catecholamines in haemolymph and cuticle during larval, pupal and adult development of Manduca senta (L.). Insect Biochem 14, 533 (1984). 7. Thompson MJ, Svoboda JA, Lusby WR, Rees HH, Oliver JE, Weirich GF, Wilzer KR: Biosynthesis of a C21 steroid conjugate in an insect. The conversion of [14C] cholesterol to 5-[14C] pregnen-3f3, 20P-diol glucoside in the tobacco hornworm, Manduca sexta. J Biol Chem260, 15410 (1985). 8. Real MD, Ferre J: Distribution of xanthurenic acid glucoside in species of the genus Drosophila. Insect Biochem 19, 111(1989). 9. Andersen JF, Plattner RD, Weisleder D: Metabolic transformations of cucurbitacins by Diabrotica virgifera virgifera Leconte and D. undecimpunctata hownrdi Barber. Insect Biochem 18, 71 (1988). 10. Hopkins TL, Ahmad S: Flavonoid wing pigments in grasshoppers. Experientia, 47, 1089 (1991). 11. Bell RA,Joachim FG: Techniques for rearing laboratory colonies of tobacco hornworm and pink bollworm. Ann Ent SOCAm 69, 365 (1976). 12. Dutton GJ: The mechanism of o-aminophenol glucoside formation in Periplanetu arnericunu. Comp Biochem Physiol7, 39 (1962). Phenol P-Clucosyltransferases in M. sexfa 223 13. Trivelloni JC: Estudio sobre la formacion de P-glucosidos en la langosta (Schistocerca cand a t a ) . Enzymologia 26, 329 (1964). 14. Mehendale HM, Dorough HW: In vitro glucosylation of 1-naphthol by insects. J Insect Physiol 18, 981 (1972). 15. Kumar SS, Dorough HW: Glucosylation by housefly microsomes and effect of monoamine oxidase inhibitors. Insect Biochem 5, 265 (1974). 16. Morello A, Repetto Y: UDP-glucosyltransferase activity of housefly microsomal fraction. Biochem J 177, 809 (1979). 17. Vaisanen MVT, Mackenzie PI, Hanninen OOF: UDP-glucosyltransferase and its kinetic fluorimetric assay. J Biochem 230, 141(1983). 18. Weirich GF, Adams JR: Microsomal marker enzymes of Manducu sexta (L.) midgut. Arch Insect Biochem Physiol I, 311 (1984). 19. Mayer RT, Chen AC, Deloach JR: Characterization of mannosyltransferases during the pupal instar of Stomoxys calcitrans (L.). Arch Insect Biochem Physiol 7, 1 (1983). 20. Fernandez-Briera A, Louisot P, Morelis R Characterization of a glucosyltransferase activity in liver plasma membrane: Modulation by cations and lipidic effectors. Int J Biochem 20, 951 (1988). 21. El-Shourbagy NA, Dorough HW: Glycoside conjugative activity in different insect and vertebrate species. J Econ Entomol 67, 344 (1974). 22. Brunet PCJ, Coles BC. Tanned silks. Proc R Soc Lond B, 187, 133 (1974). 23. Kramer KJ, Speirs RD, Lookhart G, Seib PA, Liang YT: Sequestration of ascorbic acid by the larval labial gland and haemolymph of the tobacco hornworm, Manduca sextu (L.) (Lepidoptera: Sphingidae). Insect Biochem 12, 93 (1981). 24. Elmamlouk TH, Gessner T Carbohydrate and sulfate conjugations of pnitrophenol by hepatopancreas of Hornarus americanus. Comp Biochem Physiol61 C, 363 (1978). 25. Keil U, Schreier P: Purification and partial characterization of UDP-glucose: Phenol p-Dglucosyltransferase from papaya fruit. Phytochemistry28, 2281 (1989). 26. Dutton GJ: The biosynthesis of glucuronides. In: Glucuronic Acid Free and Combined. Dutton GJ, ed. Academic Press, New York, pp 185-299 (1966). 27. Stolzel G, Pommer U, Hartung J, Graser H Employment of high-performance liquid chromatography for the determination of uridine-5'-diphosphoglucose:Phenol-P-D-glucosyltransferase activity in vitro by use of partially purified enzyme from plants of Beta vulgaris ssp. rupucea var. altissima Doll. J Chromatogr 280, 331 (1983). 28. ElmamloukTH, Mukhtar H, Bend JR: The nuclear envelope as a site of glucuronyltransferase in rat liver: Properties of and effect of inducers on enzyme activity. J Pharmacol Exp Ther 229, 27 (1981). 29. Koster AS, Noordhoek J: Similarity of rat intestinal and hepatic microsomal 7-hydroxycoumarin-UDP-glucuronyltransferase:In vitro activation by Triton-X100, UDP-N-acetylglucosamine and MgC12. Biochem Pharmacol 31, 2701 (1982). 224 Ahmad and Hopkins 30. Kalinowska M, Wojciechowski ZA: Modulation of activities of steryl glucoside hydrolase and UDPG: Sterol glucosyltransferase from Sinapzs alba by detergents and lipids. Phytochemistry 25, 45 (1986). 31. Matsuo M, Underhill EW: Purification and properties of a UDP glucose: Thiohydroximate glucosyltransferase from higher plants. Phytochemistry 10, 2279 (1971). 32. Pratviel-Sosa F, Clermont S, Percheron F, Chararas C: Studies on glucosidases and glucanases in Thuumefopoeapityocantpa larvae-11. Purification and some properties of a broad specificity P-D-glucosidase. Comp Biochem Physiol86B, 173 (1987). 33. Lindroth R: Hydrolysis of phenolic glycosides by midgut P-glucosidase in Pupilio gluucus subspecies. Insect Biochem 18, 789 (1988). 34. Gigon PL, Bickel MH: Interference of UDP-glucuronyltransferase and P-glucuronidase activity in rat liver microsomes at pH 7.5 with ynitrophenol and ynitrophenylglucuronide as substrates. Enzyme 24, 230 (1979). 35. Hose1 W: Glycosylation and glycosidases. In: The Biochemistry of Plants. Stumpf PK, Conn EE, eds. Academic Press, New York, vol7, pp 725-753 (1981). 36. Ahmed RF, Hopkins TL, Kramer KJ: Tyrosine glucoside hydrolase activity in tissues of Manducu sexta (L.): effect of 20-hydroxyecdysone. Insect Biochem 23, 641 (1983). 37. Ahmed RF, Hopkins TL, Kramer KJ: Role of juvenile hormone in regulating tyrosine glucoside synthesis for pupal tanning in Manducu sexta (L.). J Insect Physiol32, 341 (1985).