Quinone and quinone methide as transient intermediates involved in the side chain hydroxylation of N-acyldopamine derivatives by soluble enzymes from Manduca sexta cuticle.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 16:123-138 (1 991) Quinone and Quinone Methide as Transient Intermediates Involved in the Side Chain Hydroxylation of N-Acyldopamine Derivatives by Soluble Enzymes from Manduca sexta Cuticle Steven J. Saul, Hemalata Dali, and Manickam Sugumaran Departvierit of Biology, University of Massachusetts at Boston, Harbor Campus, Massachusetts Proteins solubilized from the pharate cuticle of Manduca sexta were fractionated by ammonium sulfate precipitation and activated by the endogenous enzymes. The activated fraction readily converted exogenously supplied N-acetyldopamine (NADA) to N-acetylnorepinephrine (NANE). Either heat treatment (70°C for 10 min) or addition of phenylthiourea (2.5 pM) caused total inhibition of the side chain hydroxylation. If chemically prepared NADA quinone was supplied instead of NADA to the enzyme solution containing phenylthiourea, it was converted to NANE. Presence of a quinone trap such as N-acetylcysteine in the NADA-cuticular enzyme reaction not only prevented the accumulation of NADA quinone, but also abolished NANE production. In such reaction mixtures, the formation of a new compound characterized as NADA-quinone-N-acetylcysteine adduct could be readily witnessed. These studies indicate that NADA quinone is an intermediate during the side chain hydroxylation of NADA by Manduca cuticular enzyme(s). Since such a conversion calls for the isomerization of NADA quinone to NADA quinone methide and subsequent hydration of NADA quinone methide, attempts were also made to trap the latter compound by performing the enzymatic reaction in rnethanol. These attempts resulted in the isolation of p-methoxy NADA (NADA quinone methide methanol adduct) as an additional product. Similarly, when the N-P-alanyldopamine (NBAD)-Manduca enzyme reaction was carried out i n the presence of L-kynurenine, two diastereoisomers of NBAD quinone methide-kynurenine adduct ( = papiliochrome Ila and Ilb) could be isolated. The above results are i n agreement with our hypothesis that N-acylnorepinephrine formed in Manduca cuticle is biosynthesized by an indirect route involving intermediary formation of N-acyldopamine quinone and N-acyldopamine quinone methide as established in the case of Sarcophaga bullata and is not produced by the action of a p-hydroxylase. Acknowledgments: This work was supported by grants from NIH (R01-AI-14753) and University of Massachusetts, Boston. We thank Mr. Victor Semensi for his help. Our thanks are also due to Dr. M. Yago for a generous supply of synthetic papiliochrorne Ila and Ilb. Received June15,1990; accepted September 24,1990. Address reprint requests to Dr. Manickam Sugumaran, Department of Biology, University of Massachusetts at Boston, Harbor Campus, Boston, M A 02125. 0 1991 Wiley-Liss, Inc. 124 Saul et al. Key words: cuticular sclerotization, N-acylnorepinephrine,quinone isornerization, P-hydroxylation, phenoloxidase, quinone isomerase INTRODUCTION (3-Hydroxylation of NADA,*a well known insect cuticular sclerotizing precursor, by cuticular enzyme(s) from Manduca sexta was first reported by Peter [l]nearly a decade ago. Using optical rotatory dispersion studies, he also found that the NANE formed in the reaction is a racemic mixture. Unlike the mammalian dopamine-P-hydroxylase which performs the stereoselective hydroxylation of dopamine to L-norepinephrine , the cuticular enzymes generated racemic NANE [l],the significance of which was not clear, until we first reported quinone methide production from 4-alkylcatecholby cuticular enzymes from Sarcophaga bullata . Based on the reaction of different substrates with insoluble cuticular enzyme(s), we proposed that quinone methides are involved as transient intermediates during the activation of NADA and related compounds and put forward the quinone methide sclerotization hypothesis [3,4]. According to our proposal, enzymatic activation of 4-alkyl catechols generates 2-hydroxy-p-quinone methides, which are one of the reactive species involved in sclerotization [3,4]. Since quinone methides undergo rapid Michael-1,6addition reactions with nucleophiles [S], they accounted for the formation of catechol-cuticle adducts. Since the reaction is nonenzymatic and hence nonstereoselective, it generated racemic products as found in the case of NANE [l].Two years after the advocation of quinone methide sclerotization hypothesis, Peter and Vaupel reconfirmed the racemic nature of NANE formed in Manduca cuticle by an entirely different experiment and proposed three different routes for NANE formation, only one of which included the intermediary formation of NADA quinone methide. However, we maintained that quinone methide is the key intermediate in the conversion of NADA to NANE [7-lo]. Using trapping experiments, we also found that NADA quinone is another intermediate in the enzymatic conversion of NADA to NANE in Manduca and other insects and invoked the isomerization of NADA quinone to its quinone methide [11,12]. Although quinone isomerization to quinone methides can occur nonenzymatically, as in the case of methylene substituted benzoquinones [13-151 and carboxyethyl-o-benzoquinonederivatives [16,17], we did identify an enzyme catalyzing this reaction in insect cuticle [11,12,18] and reported the purification and characterization of o-quinone isomerase from the hemolymph as well as cuticle of S. bullata [19-221. Thus, the side chain hydroxylation of NADA to NANE observed in insect cuticles is caused by the combined enzymatic (phenoloxidase and quinone isomerase) and nonenzymatic reactions, rather than the action of a single P-hydroxylase (Fig. 1). In this paper we examine the mechanism of side chain hydroxylation of N-acyldopamine derivatives by the soluble enzymes from the pharate pupal *Abbreviations used: m.pt. = melting point; NAcCys = N-acetylcysteine; NADA = N-acetyldopamine; NANE = N-acetylnorepinephrine; NBAD = N-P-alanyldopamine; NBANE = N-P-alanylnorepinephrine. Phenoloxidase/Quinone lsornerase of Manduca Cuticle 125 OH 1 2 3 4 Fig. 1. Alternate routes for the P-hydroxylation of N-acyldopamine derivatives. N-acyldopamine (1) (R = CH3; NADA; R = CH2CH2NH2;NBAD) can be directly hydroxylated bya P-hydroxylase type enzyme or converted to N-acylnorepinephrine (2) through the intermediary formation of quinone (3) and quinone methide (4). A = phenoloxidase; 5 = quinone isomerase; C = nonenzymatic reaction; D = P-hydroxylase. cuticle of M. sextn. Our results confirm the contention that in Manduca cuticle the conversion of N-acyldopamine to N-acylnorepinephrine involves the intermediary formation of N-acyldopamine quinone and N-acyldopamine quinone methide as established in the case of S. b u h t a and not direct P-hydroxylation. MATERIALS AND METHODS Enzyme Preparation Eggs of M. sexta larvae were kindly donated by Dr. J.S. Buckner of Metabolism and Radiation laboratory, Agricultural Research Service, U. S. Department of Agriculture, Fargo, ND. Larvae were reared on a synthetic medium , and kept at 25°C during a 16 h light-8 h dark photoperiod. The cuticles were harvested from pharate pupae by dissection and washed thoroughly with ice cold water. They were then homogenized in a Waring blender (DynamicsCorp. America New Hartford, CT) for 30 s and washed with excess water. After repeating homogenization and washing once more, the recovered cuticle was suspended in 0.1% sodium borate, pH 8.0, containing 1 p,M phenylthiourea for 3 h. At the end of this period, the solubilized proteins were collected by passing the mixture through cheese cloth and the contents centrifuged. The clean supernatant was subjected to ammonium sulfate fractionation. The proteins precipitated between 20-40% ammonium sulfate saturation were collected by centrifugation and individually suspended in 5 ml of water. The contents were left at 5°C for 1week and used as the enzyme source. Chemicals P-Methoxy-NADA was synthesized as follows: NANE (270 mg) was stirred in 1N methanolic hydrochloric acid (25 ml) at room temperature for 2 h. After this period, the reaction mixture was neutralized with ammonium hydroxide and concentrated on a rotary evaporator. The solid material obtained was extract- 126 Saul et al. ed with ethyl acetate. Distillation of the organic solvent on a rotary evaporator gave a white hygroscopic solid of m. pt. 53-55°C (160 mg; yield, 56%). H-nuclear magnetic resonance (NMR) (DMSO-d6)6 = 1.85 (s, 3H, COCH3), 2.85-3.40 (m, 2H, CH2),3.11 (s, 3H, OCH3),3.80-4.25 (m, lH, CH), 6.40-6.90 (m, 3H, ArH), 7.70-8.05 (br s, lH, NH), 8.45-9.30 ppm (br s, 2H, OH); NH and OH signals were exchanged by D20. NADA-quinone-NAcCys adduct was synthesized by the procedure published from this laboratory 1241. NADA, 3,Pdihydroxyphenylglycol were procured from Sigma Chemical Co. ,St. Louis, MO. NANE was prepared by acetylation of norepinephrine . A similar procedure was used to synthesize N-acetylarterenone. 3,4-Dihydroxyphenethyl alcohol was prepared as outlined in an earlier publication .Papiliochrome IIa and IIb were gifts from Dr. Motoko Yago of Iwate Medical College, Morioka, Japan. High Performance Liquid Chromatography HPLC analysis of catechols was performed using a Beckman (Berkeley, CA) 332 liquid chromatography system equipped with two model 1IOB pumps, a model 420 controller, a model 160 absorbance detector (280 nm), and model 427 integrator. Chiral separation of P-methoxy NADA was achieved on an acetylated-P-cyclodextrin column as described in an earlier publication . RESULTS Figure 2 shows the HPLC analysis of different reaction mixtures containing NADA and soluble enzyme(s). As soon as NADA and the enzyme fraction were mixed, the only catecholic compound detected by HPLC analysis of the reaction mixture is parent NADA (Fig. 2, trace A). However, 5 min incubation with the Manduca enzyme fraction generated additional products (Fig. 2, trace B). The peak at 1.99 min was identified as NANE by comparison of its elution time with that of synthetic NANE. Both this material and NANE co-chromatographed as a single symmetrical peak and exhibited the same ultraviolet absorbance spectrum (X max = 278 nm), confirming that they are one and the same compound. The peak at 2.62 min was confirmed as NADA quinone by comparison of its visible spectrum (data not shown) and elution profile (Fig. 2, trace C). Heat inactivated enzymes failed to cause these conversions (Fig. 2, trace D). From these studies, it can be concluded that NADA is converted by the Manduca cuticular enzymes to NADA quinone and NANE. NANE can be formed either by P-hydroxylation of NADA, or by hydration of NADA quinone methide formed by the isomerization phenoloxidase generated NADA quinone. The following line of evidences indicates that NANE is generated from NADA quinone and NADA quinone methide and not directly from NADA. When the NADAenzyme reaction is performed in presence of phenylthiourea, a specific inhibitor of phenoloxidase, not only the formation of NADA quinone is totally inhibited, but also NANE production is abolished (Fig. 2, trace E), indicating that phenoloxidase is essential for the synthesis of both NADA quinone and NANE. In order to check whether NADA quinone is the precursor for NANE, chemically synthesized NADA quinone was supplied to the Manduca enzyme prep- PhenoloxidaselQuinone Isomerase of Manduca Cuticle 127 E 0 m N I- 4 w 0 z U m U 0 v) m d 0 2.0 4.0 0 2.0 4.0 ELUTION TIME (mln) Fig. 2. Oxidation of NADA by soluble enzymes from M . sexta. A reaction mixture (1 ml) containing 0.22 mM NADA in 50 mM sodium phosphate buffer, pH 6.0, was incubated at room temperature. A 50 pI aliquot of enzyme preparation (14 mg/ml) was added to the reaction mixture and at the indicated times, a 20 pI aliquot was subjected to HPLC analysis on a Beckman C t 8ultrasphere reversed phase column (5 km, 4.6 x 150 rnm) using isocratic elution with 50 rnM acetic acid containing 0.2 mM sodium octylsulfonate in 20% methanol at a flow rate of 1.0 ml/min. Generation of NADA quinone was achieved by incubating the reaction mixture (lacking enzyme) with excess silver oxide for 2 min. The contents were filtered and subjected to HPLC analysis. Inactivation of enzyme was achieved by heating at 70°C for 10 min. A = zero time reaction; B = 5 min reaction; C = silver oxide generated NADA quinone reaction; D = 5 min reaction with heat inactivated enzyme; E = same as B, but the reaction mixture contained 2.5 pM phenylthiourea. The peaks at 1.99,2.62, and 4.0 min were identified as NANE, NADA quinone, and NADA, respectively. aration and its fate analyzed by HPLC studies. As shown in Figure 3 (trace A), when chemically synthesized NADA quinone was immediately subjected to HPLC analysis, three compounds could be identified. The major peak at 2.62 min was identified as NADA quinone by spectral analysis and comparison with mushroom tyrosinase generated NADA quinone. The peak at 4.0 min was due to unoxidized NADA. The minor peak observed at 1.99 min was identified as NANE. Five minute incubations of the NADA quinone with Munduca enzyme preparation caused complete conversion of NADA quinone to NANE (Fig. 3, trace B), attesting that NADA quinone is the precursor for NANE. If this is true, trapping of NADA quinone formed in the NADA-Manduca enzyme reaction should cause total inhibition of NANE formation. Therefore, we performed the NADA-Munduca enzyme reaction in the presence of NAcCys, a well established quinone trap. As shown in Figure 4, if NAcCys is included in the NADA-enzyme reaction, formation of both NADA quinone and NANE 128 Saul et al, E 0 Q N 2 w 0 z 4 m c 0 fn m 4 1 1 C ELUTION TIME (rnln) Fig. 4. Effect of NAcCys o n NADA-enzyme reaction. The reaction conditions are as outlined for Figure 2, except for the inclusion of NAcCys (2mm) in the experiment. Incubation time was S rnin. A = Control without NAcCys; B = experiment with NAcCys; C = standard NAcCysNADA quinone adduct. The peaks at 2.0,2.62,4.0,and 9.3 min were identified as NANE, NADA quinone, NADA, and NAcCys-NADA quinone adduct, respectively. Phenoloxidase/Quinone lsomerase of Manduca Cuticle 129 1.0 YI 0 2 4 m E 0.5 0 v) rn U 0 220 290 360 WAVELENGTH (nm) Fig. 5. Ultraviolet absorbance spectrum of NADA-quinone-NAcCys adduct in 0.2 M acetic acid. A = Synthetic material; 6 = enzymatic product. were totally abolished (Fig. 4, trace B). But, formation of a new product eluting 9.30 min could be observed. This product exhibited the same retention time as that of an authentic NADA-quinone-NAcCys adduct (Fig. 4, trace C) and possessed the same spectra1 characteristics as that of the synthetic NADA quinone-NAcCys adduct (Fig. 5). Therefore, trapping of NADA quinone by NAcCys prevented the formation of NANE. From these studies, it was concluded that NADA quinone and not NADA itself is the precursor for NANE formation in Marzduca. Chemical considerations indicated that NADA quinone has to be tautomerized to its quinone methide analog before yielding NANE (Fig. 1). If indeed NADA quinone methide is the precursor for NANE, then one should be able to trap the transiently formed NADA quinone methide. Earlier we have demonstrated that NADA quinone methide formed in S. bullata can be trapped by methanol to yield p-methoxy NADA [20,22]. We have used the same approach to prove the formation of NADA quinone methide in Munducu as well. Figure 6 shows the HPLC analysis of the NADA-enzyme reaction mixture conducted in the presence of methanol (traces A, 9). From Figure 6 it is evident that there is a new product formed (4.76 min peak in traces A, B) which is dependent on the presence of methanol. This compound could be either a NADA quinone methanol adduct or a NADA quinone methide methanol adduct. In our previous work , we relied on the ultraviolet absorbance spectra to distinguish between these two adducts. However, to unambiguously establish the structure of the adduct, we chemically synthesized P-methoxy NADA by employing the procedure outlined in the Materials and Methods section. The synthetic compound exhibited the same retention time as that of the biological isolate (Fig, 6, trace C) and co-chromatographed as a single symmetrical peak with the latter, confirming the trapping of NADA quinone methide as its methanol adduct. The NMR spectrum of synthetic P-methoxy NADA shown in Figure 7 confirmed its structural assignment. The synthetic (3-methoxy NADA exhibited the same ultraviolet absorbance spectrum (Fig, 8) as that of the biological isolate, and could be resolved into the individual stereoisomers by HPLC chromatography on a chiral column (Fig. 8, inset). Isomerization of NADA quinone to its quinone methide tautomer could occur 130 Saul et al. !- U c 0 2.0 4.0 ELUTION TIME ( m l d Fig. 6. Trapping of NADA quinone methide with methanol. A reaction mixture containing 0.22 m M NADA, enzyme protein, 50 m M sodium phosphate buffer, pH 6.0, in 10% methanol was incubated at room temperature and an aliquot was subjected to HPLC analysis as outlined for Figure 2. A = 5 rnin reaction; B = 15 min reaction; C = authentic p-methoxy NADA. either enzymatically or nonenzymatically. Kramer and his associates have claimed that the reaction occurs spontaneously and nonenzymatically [26,27]. To determine whether or not NANE is formed by an enzymatic route, the Manduca enzymes were heat inactivated at 70°C for 10 min and substituted for the native enzyme in NADA. quirtone-enzyme reaction. Such an experiment failed to produce NANE over the nonenzymatic rates reported from this laboratory , confirming the actual presence of a quinone isomerase in the enzyme preparation. Accordingly, the NANE formation exhibited a pH optimum characteristic of enzymatic reactions. The coupled phenoloxidase/ Fig. 7. NMR spectrum of p-methoxy NADA in DMSO (d6). Phenoloxidase/Quinone lsomerase of Manduca Cuticle 285 220 131 350 WAVELENGTH f n m ) Fig. 8. Ultraviolet absorbance spectrum of p-methoxy NADA. A = In 0.2 M acetic acid; B = In sodium borate, pH 8.5. Inset: HPLC analysis of p-methoxy NADA on acetylated cyclobond column. Conditions are as outlined in Materials and Methods. quinone isomerase exhibited a pH optimum of 6.5 (Fig. 9). Murtducu phenoloxidases have been shown to exhibit a pH optimum of 6 [27,28]. Therefore, it is possible that the pH optimum of quinone isomerase is about 6.5. Apart from NADA, the combined phenoloxidase/quinone isomerase system of Manducu also attacked NBAD and 3,4-dihydroxyphenethyl alcohol and generated their side chain hydroxylated products. Figure 10, for instance, shows the action of Manduca enzymes on 3,4-dihydroxyphenethylalcohol. As is shown, the enzyme action produced 4-hydroxyethyl-o-benzoquinoneand 3,kdihydroxyphenyl glycol analogous to NADA oxidation reaction. Figure 11gives the HPLC analysis of two reaction mixtures containing NBAD, L-kynurenine, and mushroom tyrosinase with and without Manducu soluble enzymes. The peak at 1.44 min was identified as NBANE in comparison with the chromatographicbehavior of the authentic compound. The peak at 1.82 min was identified as NBAD quinone. While NBAD gave rise to the 1.97 min peak, kynurenine eluted at 2.23 min. In addition to the two unidentified peaks elut- 4.0 5.0 0.0 7.0 8.0 PH Fig. 9. pH optimum of phenoloxidase/quinone isomerase. NANE formed in a Manduca phenoloxidase/quinone isomerase was quantified at different pH values and plotted. 132 Saul et al. E 0 W N s W 0 2 U m a 2m U 1 0 2.0 4.0 ELUTION TIME (mln) 6 .O Fig. 10. Enzymatic oxidation of 3,4-dihydroxyphenethyl alcohol. A reaction mixture (1 ml) containing 0.35 m M 3,4-dihydroxyphenethyl alcohol in 50 mM sodium phosphate buffer, pH 6.0, was incubated at room temperature. A 50 pI aliquot of soluble enzyme solution was added to the reaction mixture and an aliquot (20 kI) was subjected to HPLC analysis as outlined for Figure 2. A = zero time reaction; B = 5 min reaction. The peaks at 1.88, 2.46, and 3.54 min were identified as 3,4-dihydroxyphenyl glycol, hydroxyethyl-o-benzoquinone,and 3,4-di hydroxyphenethyl alcohol, respectively. ing at about 3 min, two peaks eluting at 3.94 and 4.65 min were observed. The last two peaks did not appear in the chromatograms of the reaction mixture lacking either L-kynurenine or NBAD (data not shown), indicating that they are some kind of NBAD-kynurenine adducts. Papiliochrome IIa and IIb are two diastereoisomeric structures arising from the nonstereospecific condensation of L-kynurenine with NBAD quinone methide [29-321. An authentic sample of papiliochrome IIa and IIb supplied by Dr. Yago exhibited the same retention time as the above two peaks (Fig. 11, dotted chromatograms) and co-chromatographed with each other indicating that they are indeed papiliochrome IIa and IIb. Comparison of the two reactions also reveals that NBANE production as well as papiliochrome formation is a lot higher in the reaction mixture containing Manduca enzymes than in the control reaction, attesting to the presence of an enzyme system generating NBAD quinone methide in the Manduca enzyme preparation. DISCUSSION P-Hydroxylation of NADA by cuticular enzyme(s) from M. sexta was reported nearly a decade ago by Peter [l],but the mechanism of the hydroxylation has not been clarified. Based on Peter’s [l]finding that NANE formed in the enzymatic hydroxylation of NADA side chain in M. sexta cuticle is a racemic PhenoloxidaseIQuinone lsomerase of Manduca Cuticle A 133 ?I3 7 Il V- E N I s ' 9 c ? - 0 m N t * . y1 0 2 U m a E U Fig. 11. Enzymatic oxidation of NBAD in presence of L-kynurenine. A reaction mixture (1 mi) containing 2 mM NBAD, 2 mM L-kynurenine in 50 mM sodium phosphate buffer, p H 6.0, and mushroom tyrosinase (40 pg) was incubated at room temperature for 76 min and a20 pl aliquot was subjected to HPLC analysis on a Beckman Cq8ultrasphere reversed phase column (5 pm, 4.6 x 150 rnrn) using isocratic elution with 100 mM sodium phosphate buffer, pH 3.3 in 10% acetonitrile at a flow rate of 1 ml/min (trace A). Trace 6 is the chromatogram of reaction as described above, but also contained Manduca enzyme. The peaks at 1.44, 1.02, 1.97, and 2.23 min were identified as NBANE, NBAD quinone, NBAD, and L-kynurenine, respectively. The two peaks at 3 min are due to unidentified compounds. The peaks at 3.94 and 4.65 min are due to papiliochrorne Ilaand Ilb. The chromatograms with broken lines are due to authentic papiliochrome Ilaand Ilb. mixture and our own experiments on the substrate specificity and product analysis of sarcophagid cuticular phenoloxidase 131, we proposed that quinone methides are transient intermediates in the reaction and put forward the quinone methide sclerotizationhypothesis [3,4]. Initially, we failed to observe the accumulation of NADA quinone in reaction mixtures containing NADA and cuticular enzymes  and believed that NADA oxidation to NADA quinone methide involves a direct two electron oxidation process [3,4,7-91, although we did acknowledge the feasibility of quinone to quinone methide isomerization route for this conversion. The possibility of N-acyldopamine quinone isomerizing nonenzyma tically to N-acyldopamine quinone methide was considered as early as 1958by Witkop and his associates, who also proposed that the (3-hydroxylation of dopamine to norepinephrine involves a similar reaction [33-351. However, in a dramatic turn of events, it was demonstrated that the dopamine-P-hydroxylationinvolves the stereospecific incorporation of one atom of molecular oxygen into the substrate and not a nonstereoselective and nonenzyinatic hydration of dopamine quinone methide [2,36]. As a consequence of the well established enzymology of dopamineP-hydroxylasereaction, it was often assumed that N-acyldopamine-P-hydroxylation commonly observed in insect cuticle is also catalyzed by P-hydroxylase. 134 Saul et al. Thus, Kramer and his associates recently designated the enzyme causing the conversion of N-acyldopamine to N-acylnorepinephrine as P-hydroxylase [37-391. However, recent studies from this laboratory [10-12,17-221 conclusively prove the operation of the original pathway proposed by Witkop and his associates [33-351, viz., hydration of the N-acyldopamine quinone methide generated from the isomerization of N-acyldopamine quinones as the route for the side chain hydroxylation of N-acyldopamine derivatives. We first reported the presence of an enzyme catalyzing the isomerization of N-acyldopamine quinones to N-acyldopamine quinone methides in the insoluble pharate cuticle of M . sexta [ll],and subsequently obtained the enzyme in the soluble form from both the cuticle and hernolymph of S. bullata larvae [ 19-22]. Apart from trapping the enzymatically generated NADA quinone methide as its methanol adduct, we also demonstrated that the resultant pmethoxy NADA formed was a racemic mixture and confirmed the nonenzymatic, and hence nonstereoselective nature of the observed Michae1-lf6-additionreaction [20,22].In addition, we invoked the quinone methides in immune responses in insects [21,40] and established that the biosynthesis of the sclerotizing precursor, 1,2-dehydro-NADA (discovered by Andersen and Roepstorff ) involves direct tautomerization of NADA quinone methide [42-441. Independent of the discovery of quinone isomerase, we delineated a few facile nonenzymatic routes for quinone methide formation in the case of methylene- and carboxy ethyl-substituted benzoquinones [13-171. As a result, it was essential to compare the rate of enzymatic with nonenzymatic rates of synthesis of NANE from NADA. In a reaction mixture (1ml) containing 1mM NADA and 10 pg tyrosinase, NANE production was found to be about 1nmol/ min [17,221. In contrast, inclusion of 1unit of purified quinone isomerase from S. bullutu in the above reaction dramatically increased the rate of NANE production to about 400 nmoVmin [17,22]. However, the specific activity of quinone isomerase is about 4,000 unitsimg 1221. Therefore, on a per milligram basis, if the tyrosinase reaction is generating about 100 nmol of NANE/min, quinone isomerase is able to generate as much as 1.6 mmol of NANE/min. In spite of our claim that there is a quinone isomerase in Manducu cuticle [11,12], Kramer and his associates recently doubted the existence of quinone isomerase in the cuticle of Manduca [26,27]. They asserted that the side chain hydroxylation could be due to nonenzymatic origin  or phenoloxidase catalyzed  and also suggested that the side chain hydroxylase activity was due to P-hydroxylase [37-391. In order to prove the existence of quinone isomerase, clarify the role of P-hydroxylase, and to conclusively establish the intermediary nature of N-acyldopamine quinone and N-acyldopamine quinone methide, we undertook the present study. The results presented in this paper with soluble enzyme preparation from Manducu cuticle confirm our contention that NADA conversion to NANE observed in Manduca is caused by the intermediate formation of NADA quinone and NADA quinone methide [ll,121. Accordingly, the side chain hydroxylation involved enzymatic isomerization of phenoloxidase generated NADA quinone to its quinone methide tautomer and its subsequent nonenzymatic hydration (Figs. 1,12). That the NADA quinone is an obligatory intermediate in this conversion is confirmed by trapping experiments with NAcCys, as well as inhibition by phenylthiourea, while PhenoloxidasdQuinone lsomerase of Manduca Cuticle 135 OH HO 2 1 4' Me 0 OH HoJ&R ' 3$ b I , " " --3 HO HO 3 4 L-KYNURENINE HO 6 Fig. 12. Summary of the observed reactions. The side chain of N-acyldopamine derivatives (1: NADA, R = CH2NHCOCH3;NBAD, R = CH2NHCOCH2CH2NH,)i s not directly hydroxylated by a dopamine-p-hydroxylasetype enzyme (D) in Manduca cuticle but by the combined action of phenoloxidase (A) and quinone isomerase ( 8 ) through the intermediary formation of 2: quinone, and 3: quinone methide. Accordingly, the reaction of NADA with cuticular enzymes generates quinone methide (3),which can be trapped either by water to produce 4: NANE, or by exogenously added methanol to generate 5: racemic p-methoxy NADA. In the case of NBAD reaction, the corresponding quinone methide reacts with water to yield NBANE or reacts with the exogenously added L-kynurenine to give 6 : papiliochrome Ila and Ilb. the intermediacy of NADA quinone methide has been confirmed by trapping experiments with methanol (Fig. 12). Experiments with NBAD and kynurenine also confirmed the formation of quinone methide as papiliochrome IIa and IIb were recovered as additional products in the reaction mixture (Figs. 1 1 , l Z ) . Evidence for the existence of quinone isomerase (or a similar enzyme involved in side chain hydroxylation) in Manduca cuticle comes from the work of Kansas group only. During their studies on NBANE biosynthesis by Manduca cuticle, they observed the formation of about 22 Fmol of NBANE per gram of wet cuticle from 67 pmol of NBAD in 10 min . Therefore, the rate of NBANE synthesis is about 2.2 pmol/min/g tissue. If we assume that the wet cuticle had about 0.01% by weight phenoloxidase, this gives a specific activity of 22 pmol/min/mg of phenoloxidase(s).But in reality, the purified phenoloxidase(s) from Maizdtaca exhibit a rate of about 300 nmol of side chain hydroxylated product formed/min/mg of protein , which is of the same order as that of purified mushroom tyrosinase reaction (about 100 nmol/min/mg protein), but at least 70 times lower than the rate reported for whole cuticle that calls for enzymes other than phenoloxidase(s) to be responsible for side chain hydroxylation of N-acyldopaminederivatives.This enzyme could be either a f3-hydroxylase[37-391 or the quinone isomerase. The mechanistic studies outlined in the present paper favor the route involving quinone isomerization to quinone methide, and discount the p-hydroxylase action. Evidence for the presence of quinone isomerase in the soluble cuticular preparation comes from the following studies. If NADA quinone is chemically synthesized and provided to the enzyme preparation treated with phenylthiourea, 136 Saul et al. it is still converted to NANE (Fig. 3). This conversion, however, returns to basal level if the heat inactivated enzyme is substituted to the native enzyme. Our repeated attempts to separate the quinone isomerase from phenoloxidase ended in vain. This could be due to the formation of a complex between phenoloxidase and quinone isomerase. Such a complex formation is common among enzymes catalyzing the consecutive reactions in a metabolic pathway and the complexes are called metabolons . Metabolon formation is advantageous especially for those enzymes dealing with unstable intermediates. If phenoloxidase and quinone isomerase exist independently, the phenoloxidase generated quinone has to be released into solution before it can diffuse to the active site of quinone isomerase. On the other hand, if phenoloxidase/quinone isomerase active sites are coupled by complex formation, loss of reactive NADA quinone intermediate into the solution could be avoided and efficiency can be achieved by directly transferring the NADA quinone generated at the active site of phenoloxidase to the active site of quinone isomerase (channeling),which also ensures raising of the local concentration of the substrate for effective catalysis . We have already reported that in Surcophuga phenoloxidase/quinone isomerase comigrate on Sepharose6B column chromatography, probably reflecting complex formation. In this organism, we could separate these two enzyme activities independent of each other. However, we have so far been unsuccessful in resolving these two activities from Munducu cuticle. We are continuing our studies with the Munducu cuticle to obtain more information on the nature of quinone isomerase in this organism. LITERATURE CITED 1. Peter MG: Products of in vitro oxidation of N-acetyldopamine as possible components in the sclerotization of insect cuticle. Insect Biochem 20,221 (1980). 2. Taylor KB: Dopamine-p-hydroxylase. Stereochemical course of the reaction. J Biol Chem 249, 454 (1974). 3. Sugumaran M, Lipke H: Quinone methide formation from 4-alkylcatechols-A novel reaction catalyzed by cuticular phenoloxidase. FEBS Lett 255,65 (1983). 4. Lipke H, Sugumaran M, Henzel W: Mechanisms of sclerotization in dipterans. Adv Insect Physioll7, 1 (1983). 5. Wagner HV, Grompper R: Quinone methides. In: The Chemistry of the Quinonoid Compounds. Patai S, ed. Wiley, London, part 2, pp 1145-1178 (1974). 6. Peter MG, Vaupel W Lack of stereoselectivity in the enzymaticconversion of N-acetyldopamine into N-acetylnoradrenaline in insect cuticle. J Chem SOCChem Commun 848 (1985). 7. Sugumaran M: Quinone methide sclerotization: A revised mechanism for p-sclerotization of insect cuticle. Bioorg Chem 25,194 (1987). 8. Sugumaran M: Quinone methide sclerotization. In: Molecular Entomology. Law JA, ed. UCLA Symposia on Molecular and Cellular Biology, New Series, Alan R. Liss, Inc., New York, Vol 49, pp 357-367 (1987). 9. Sugumaran M: Quinone methidesand not dehydrodopamine derivative-s reactive intermediates of P-sclerotization in the puparia of flesh fly Sarcophuga bullata. Arch Insect Biochem Physiol8, 73 (1988). 10. Sugumaran M: Molecular mechanisms for cuticular sclerotization. Adv Insect Physiol 21, 179 (1988). 11. Saul SJ, Sugumaran M: A novel quinone: quinone methide isomerase generates quinone methides in insect cuticle. FEBS Lett 237,155 (1988). 12. Sugumaran M, Saul SJ, Semensi V: On the mechanism of formation of N-acetyldopamine quinone methide in insect cuticle. Arch Insect Biochem Physiol9, 269 (1988). 13. Sugumaran M, Semensi V, Dali H, Mitchell W: Navel transformations of enzymatically gen- PhenoloxidaselQuinone lsomerase of Manduca Cuticle 137 erated carboxymethyl-o-benzoquinone to 2,5,6-trihydroxybenzofuran and 3,4-dihydroxymandelic acid. Bioorg Chem 17,86 (1989). 14. Sugumaran M: Novel oxidation chemistry of catecholamine derivatives and related compounds. In: Biological Oxidation Systems. Reddy CC, Hamilton GA, Madyastha KM, eds. Academic Press, San Diego, Vol I, pp 347-363, (1990). 15. Sugumaran M, Semensi V, Dali H, Nellaiappan K: The oxidation of 3,4-dihydroxybenzyl alcohol-A sclerotizing precursor for cockroach ootheca. Arch Insect Biochem Physiol 16,31-44 (1990). 16. Sugumaran M, Dali H, Kundzicz H, Semensi V: Unusual intTamolecularcyclization and side chain desaturation of carboxyethyl-o-benzoquinonederivatives, Bioorg Chem 17,443 (1989). 17. Sugumaran M, Semensi V, Dali H, Saul SJ: Nonenzymatic transformations of enzymatically generated N-acetyldopamine quinone and isomeric dihydrocaffeiyl methyl amide quinone. FEBS Lett 255,345 (1989). 18. Sugumaran M, Semensi V, Saul SJ: On the oxidation of 3,4,-dihydroxyphenethyl alcohol and 3,4-dihydroxyphenyl glycol by cuticular enzyme(s) from Surcophugu bullatu. Arch Insect Biochem Physiol20,13 (1989). 19. Saul SJ, Sugumaran M: o-Quinoneiquinone methide isomerase: A novel enzyme preventing the destruction of self-matterby phenoloxidase-generatedquinones during immune response in insects. FEBS Lett 249, 155 (1989). 20. Sugumaran M, Saul SJ, Semensi V: Trapping of transiently formed quinone methide during enzymatic conversion of N-acetyldopamine to N-acetylnorepinephrine. FEBS Lett 252,135 (1989). 21. Saul SJ, Sugumaran M: Characterization of quinone tautomerase activity in the hemolymph of Sarcoyhugu bullntu larvae. Arch Insect Biochem Physiol12,157 (1989). methide isomerase from the 22. Saul SJ, Sugumaran M: 4-Alkyl-o-quinone/2-hydroxy-p-q~inone larval hemolymph of Surcophqa bullatu. I. Purification and characterization of the enzyme catalyzed reaction. J Biol Chem 265, 16992 (1990). 23. Bell RA, Joachim FG: Techniques for rearing laboratory colonies of Tobacco hornworms and pink bollworm. Ann Entomol SOCAm 69,365 (1976). 24. Sugumaran M, Dali H, Semensi V: Chemical and cuticular phenoloxidase mediated synthesis of cysteinyl-catechol adducts. Arch Insect Biochem Physiol11,127 (1989). 25. Dali H, Sugumaran M: An improved synthesis of 1,2-dehydro-N-acetyldopamine. Org Prep Proc Int 20,191 (1988). 26. Morgan TD, Yonekura M, Kramer KJ, Hopkins TL: Oxidative metabolism of catecholamines by tyrosinases from the integument of pupal Munduca sexta. Proc XVIII International Congress on Entomology, Vancouver, Canada, p 141 (1988). 27. Thomas BR, Yonekura M, Morgan TD, Czapla TH, Hopkins TL, Kramer KJ: A trypsin solubilized laccase from pharate pupal integument of the tobacco hornworm Munducu sexta. Insect Biochem 19,611 (1989). 28. Aso Y, Kramer KJ, Hopkins TL, Whetzel SZ: Properties of tyrosinase and dopa quinone imine conversion factor from pharate pupal cuticle of Munducu sextu L. Insect Biochem 14, 463 (1984). 29. Umebachi Y, Yoshida K: Some chemical and physical properties of papiliochrome I1 in the wings of Papilio xuthus. J Insect Physioll6, 1203 (1970). 30. Rembold H, Umebachi Y: The structure of papiliochrome 11, the yellow wing pigment of the papilionid butterflies. In: Progress in Tryphophan and Serotonin Research. SchlossbergerHG, Kochen W, Lingen B, Steinhart H, eds. Walter de Gruyter & Co., Berlin, pp 743-746 (1984). 31. Yago M: Enzymatic synthesis of papiliochrome 11, a yellow pigment in the wings of papilionid butterflies. Insect Biochem 19,673 (1989). 32. Sugumaran M, Saul SJ,Dali H: On the mechanism of side chain oxidation of N-p-alanyldopamine by cuticular enzymes from Surcophg~b u h t a . Arch Insect Biochem Physiol15,255(1990). 33. Senoh S, Witkop B, Creveling CR, Udenfriend S: Oxidation mechanisms of catecholamines and the biosynthesis of noradrenalin. Fourth International Congress on Biochemistry, Pergamon Press, London, Vol13, pp 176-188 (1958). 34. Senoh S, Witkop B: Nonenzymatic conversions of dopamine to norepinephrine and trihydroxyphenethylamines.J Am Chem Soc 81,6222 (1959). 35. Senoh S, Creveling CR, Udenfriend S, Witkop B: Chemical, enzymatic and metabolic studies on the mechanism of oxidation of dopamine. J Am Chem SOC81,6236 (1959). 138 Saul et al. 36. Kaufman S, Bridges WF, Eisenberg F, Friedman S: The source of oxygen in the phenylalanine hydroxylase and the dopamine-p-hydroxylase catalyzed reactions. Biochem Biophys Res Commun 9,497 (1962). 37. Kramer KJ, Hopkins TL: Tyrosinase metabolism for insect cuticle tanning. Arch Insect Biochem Physiol6, 279 (1987). 38. Czapla TH, Hopkins TL, Kramer KL, Morgan TD: Diphenols in hemolymph and cuticle during development and cuticle tanning of Periplunetu americuna (L.) and other cockroach species. Arch Insect Biochem Physiol7,13 (1988). 39. Kramer KJ, Bork V, Schaefer J, Morgan TD, Hopkins TL: Solid state I3C nuclear magnetic resonance and chemical analysis of insect noncuticular sclerotized support structures: Mandid ootheca and cocoon silks. Insect Biochem 19,69 (1989). 40. Sugumaran M: Prophenoloxidase activation and insect immunity. In: Defense Molecules. Marchalonis JJ, Reinisch CL, eds. UCLA Symposia on Molecular Biology. Wiley-Liss, NY, Vol121, p p 47-62 (1990). 41. Andersen SO, Roepstorff P: An unsaturated derivative of N-acetyldopamine and its role in sclerotization. Insect Biochem 12,269 (1982). 42. Saul SJ, Sugumaran M: Characterization of a new enzyme system that desaturates the side chain of N-acetyldopamine. FEBS Lett 252, 69 (1989). 43. Saul SJ, Sugumaran M: N-acetyldopamine quinone methide/l,2-dehydro-N-acetyldopamine tautomerase. A new enzyme involved in sclerotization of insect cuticle. FEBS Lett 255,340(1989). 44. Saul SJ, Sugumaran M: Biosynthesis of dehydro-N-acetyldopamine by a soluble enzyme preparation from the larval cuticle of Sarcophugu bullutu involves intermediary formation of N-acetyldopamine quinone and N-acetyldopamine quinone methide. Arch Insect Biochem Physiol 15,237 (1990). 45. Morgan TD, Hopkins TL, Kramer KJ, Roseland CR, Czapla TH, Tomer KB, Crow FW: N-p-alanylnorepinephrine: Biosynthesis in insect cuticle and possible role in sclerotization. Insect Biochem 17,255 (1987). 46. Srere PA: Complexes of sequential metabolic enzymes. Annu Rev Biochem 56,89 (1987).