Differences in midgut serine proteinases from larvae of the bruchid beetles Callosobruchus maculatus and Zabrotes subfasciatus.код для вставкиСкачать
18 Silva et al. Archives of Insect Biochemistry and Physiology 47:18–28 (2001) Differences in Midgut Serine Proteinases From Larvae of the Bruchid Beetles Callosobruchus maculatus and Zabrotes subfasciatus Carlos P. Silva,1* Walter R. Terra,2 and Rodrigo M. Lima1 1 Laboratório de Química e Função de Proteínas e Peptídeos, Centro de Biociências e Biotecnologia, Universidade Estadual do Norte Fluminense, Campos dos Goytacazes, Brasil 2 Departamento de Bioquímica, Instituto de Química, Universidade de São Paulo, Caixa Postal 26077, CEP 05599-970, São Paulo, Brasil Proteinase activities in the larval midguts of the bruchids Callosobruchus maculatus and Zabrotes subfasciatus were investigated. Both midgut homogenates showed a slightly acidic to neutral pH optima for the hydrolysis of fluorogenic substrates. Proteolysis of e-aminocaproil-Leu-Cys(SBzl)-MCA was totally inhibited by the cysteine proteinase inhibitors E-64 and leupeptin, and was activated by 1.5 mM DTT in both insects, while hydrolysis of the substrate Z-ArgArg-MCA was inhibited by aprotinin and E-64, which suggests that it is being hydrolysed by serine and cysteine proteinases. Gel assays showed that the proteolytic activity in larval midgut of C. maculatus was due to five major cysteine proteinases. However, based on the pattern of E-64 and aprotinin inhibition, proteolytic activity in larval midgut of Z. subfasciatus was not due only to cysteine proteinases. Fractionation of the larval midgut homogenates of both bruchids through ion-exchange chromatography (DEAE-Sepharose) revealed two peaks of activity against ZArgArg-MCA for both bruchid species. The fractions from C. maculatus have characteristics of cysteine proteinases, while Z. subfasciatus has one non-retained peak of activity containing cysteine proteinases and another eluted in a gradient of 250–350 mM NaCl. The proteolytic activity of the retained peak is higher at pH 8.8 than at pH 6.0 and corresponds with a single peak that is active against N-p-tosyl-GlyGlyArg-MCA, and sensitive to 250 mM aprotinin (90% inhibition).The peak contains a serine proteinase which hydrolyzes a-amylase inhibitor 1 from the common bean (Phaseolus vulgaris). Arch. Insect Biochem. Physiol. 47:18–28, 2001. © 2001 Wiley-Liss, Inc. Key words: Bruchidae; proteinases; insect digestion; seed weevil Contract grant sponsor: CNPq; Contract grant sponsor: FAPERJ; Contract grant sponsor: PRONEX; Contract grant sponsor: FENORTE; Contract grant sponsor: FINEP; Contract grant sponsor: FAPESP; Contract grant sponsor: International Foundation for Science (IFS). *Correspondence to: Carlos P. Silva, Laboratório de Química e Função de Proteínas e Peptídeos, Centro de Biociências e Biotecnologia, Universidade Estadual do Norte Fluminense, CEP 28015-620 Campos dos Goytacazes, Brasil. E-mail: firstname.lastname@example.org Received 4 October 2000; Accepted 21 January 2001 © 2001 Wiley-Liss, Inc. Bruchid Midgut Serine Proteinases INTRODUCTION The digestive proteinases of larval bruchid beetles belong to a group of putative extra cellular cathepsins with digestive function, known to occur only in certain groups of insects, such as coleopterans of the series Cucujiformia, members of Heteroptera in the order Hemiptera and cyclorrhaphous Diptera (Terra and Ferreira, 1994). Non-serine digestive proteinases resembling cathepsins were at first considered to be restricted to the hemipterans (Gooding, 1969; Houseman and Downe, 1982, 1983), but they have now been described in other insects, mainly in phytophagous insects of the order Coleoptera (Terra and Ferreira, 1994). Digestive cysteine proteinases resembling cathepsins were first identified outside the hemipterans in bruchids (Xavier-Filho and Coelho, 1980; Gatehouse et al., 1985; Kitch and Murdock, 1986; Wieman and Nielsen, 1988; Campos et al., 1989), as well as aspartic proteinases (Lemos et al., 1990; Silva and Xavier-Filho, 1991). Recently, Ishimoto and Chrispeels (1996) suggested the existence of a minor serine proteinase in larval midgut of Zabrotes subfasciatus, which may be involved in the hydrolysis of the Phaseolus vulgaris α-amylase inhibitor 1 (α-AI1) to detoxify the protein. The detection of this proteinase was based mainly on the degradation of the α-amylase inhibitor by a crude midgut homogenate followed by electrophoresis, and by the effect of serine proteinase inhibitors. These authors did not characterize the proteinase with other proteinaceous or synthetic substrates and did not describe the distribution of this proteinase in the midgut. Thus, the detoxification role of this proteinase is not clear. There are no detailed data on the specificity of bruchid digestive proteinases. Curiously, only proteinaceous substrates were used in the most detailed studies referring to digestive proteinases in bruchid beetles. This prompted us to reinvestigate the specificity of intestinal proteinases from larvae of two important bruchid pests using synthetic substrates that allow comparisons with other insects. MATERIALS AND METHODS Materials The following substrates were used: carbobenzoxy-L-arginine-arginine-p-nitroanilide (Z- 19 ArgArg-pNA), carbobenzoxy-L-glycine-glycinearginine-4-methylcoumarin (Z-GlyGlyArg-MCA), carbobenzoxy-L-phenylalanine-L-arginine-4methylcoumarin (Z-PheArg-MCA), ε-aminocaproil-LeuCys-(SBzl)-MCA, N-p-tosyl-GlyProArg-MCA, N-succinyl-AlaAlaProPhe-p-nitroanilide (SAAPF-pNA), N-succinyl-AlaAlaProLeu-p-nitroanilide (SAAPL-pNA), Nα-Benzoyl-L-arginine-pnitroanilide (BApNA). The following inhibitors were used: aprotinin, leupeptin, trans-epoxy succinyl-L-leucylamido-(4-guanidino)-butane (E-64) and pepstatin. Dithiothreitol (DTT), sodium dodecyl sulphate (SDS), and trichloroacetic acid (TCA) were all purchased from Sigma-Aldrich. αAmylase inhibitor 1 (αAI-1) from the seeds of Phaseolus vulgaris was a gift from Dr. Maria F. Grossi de Sá (Cenargen-EMBRAPA- Brasília). Insects The colony of C. maculatus used in this work was established using animals that were supplied originally by Dr. J.H.R. Santos, Centro de Ciências Agrárias, Universidade Federal do Ceará, Fortaleza, Brazil. A Z. subfasciatus colony was established from insects supplied originally by Prof. F.M. Wiendl of the Centro de Energia Nuclear na Agricultura, Piracicaba, Brazil. Stock cultures of both species are grown in Campos since 1994. The insects are reared on Vigna unguiculata seeds (cultivar Epace 10) in the dark inside an incubator chamber maintained at 29 ± 1°C and relative humidity of 65 ± 5%. Seeds are frozen in order to prevent infestation from the field. Preparation of Samples From Insects Final instar larvae were cold immobilized, dissected, and the whole midgut was removed in cold 250 mM NaCl. Actively feeding larvae with food-filled gut tract were selected for the experiments. The whole gut was then cleaned from adhering unwanted tissues, midgut walls were pulled apart, and luminal contents were collected into a known volume of distilled water. Pooled midgut tissues, after being cleaned of luminal contents by rinsing in 250 mM NaCl, were homogenized in cold distilled water in a hand-held Potter-Elvehjem homogenizer on ice. Midgut tissue homogenates were centrifuged at 20,000g for 30 min at 4°C and the supernatant (soluble frac- 20 Silva et al. tion) collected and the sediment (membrane fraction) was dispersed in distilled water using a Potter-Elvehjem homogenizer. Hydrolase Assays and Protein Determination Assays for the hydrolysis of methylcoumarin derivatives were carried out using a Hitachi F4500 fluorometer to monitor the release of free methylcoumarin (excitation wavelength 380 nm and emission wavelength 440 nm) in a 2-ml thermostated cuvette (1-cm path length). Enzyme samples (80 µl) were mixed with warmed 1.9 ml of 100 mM citrate/phosphate buffer containing 1.5 mM DTT, 3 mM EDTA, pH 6.0. Reactions were initiated by adding 20 µl of 1 mM methylcoumarin substrate. Total proteinase activity of midgut homogenates was determined using the protein substrate azoalbumin. The azoalbumin assays were adapted from a previously described method (Lemos et al., 1990). Extracts (100 µl per assay) were incubated with 100 µl of a mixture of 1% substrate (w/v) in 100 mM citrate/phosphate, pH 6.0. Proteolysis was stopped by adding 100 µl of 25% (w/v) TCA, and the reaction was incubated for 30 min on ice. Precipitated substrate was removed by centrifugation for 5 min at 7,000g at room temperature. Aliquots of 300 µl of the supernatant were added to 300 µl of 2N NaOH solution and absorbance was read at 440 nm in a Spercord spectrophotometer (Zeiss). Hydrolysis of p-nitroanilide derivatives were assayed by the release of p-nitroaniline, which absorbs maximally at 410 nm (Erlanger et al., 1961). The assays used 50 µl of 4 mM substrates, 50 µl of enzyme sources, and 100 µl of 100 mM citrate/phosphate buffer, pH 6.0, with or without inhibitors or activator. The reaction was stopped by adding 100 µl of 30% (v/v) acetic acid, and the absorbance was read at 410 nm in a Spercord spectrophotometer (Zeiss). All assays were performed at 30°C. Buffers (100 mM) that were used to determine pH optima were: sodium acetate, citrate-sodium phosphate, Tris-HCl ranging from 3–9, with intervals of 0.2 pH units. Incubations were carried out for at least four different periods of time, unless otherwise stated, and initial rates of hydrolysis were calculated. All assays were performed under conditions in which enzyme activity was proportional to protein concentration and to the time of incubation. One enzyme unit is defined, except for proteinaceous substrates, as the amount that catalyzes the cleavage of 1 µmol of substrate/min. One proteinase unit, with azoalbumin as substrate, was the amount that caused a change in absorbance of 0.01 U/min. Protein contents of samples were determined by the method of Bradford (1976) using ovalbumin as a protein standard. In Gel Assays Proteinases from bruchid midgut preparations were detected and partially characterized by 10 or 12% SDS-polyacrylamide gel electrophoresis containing 0.1% (w/v) gelatin (Heussen and Dowdle, 1980). Samples were diluted twofold in electrophoresis sample buffer (2.1 ml distilled water, 0.5 ml 0.5 M Tris-HCl, pH 6.8, 0.4 ml glycerol, 0.8 ml 10% (w/v) SDS, 0.2 ml 1% (w/v) bromophenol blue) without 2-mercaptoethanol and subjected to electrophoresis (Laemmli, 1970) at 150 V and 4°C without heating the samples. Following electrophoresis, the gels were transferred to a 2.5 % (w/v) aqueous solution of Triton X-100 for 20 min at room temperature in order to allow renaturation of the enzymes. The gels were then incubated with a proteolysis buffer (100 mM citrate/phosphate 1.5 mM DTT, pH 6.0) for different periods of time. Proteolysis was stopped by transferring gels to a staining solution (0.1% w/v Coomassie Brilliant Blue in 40% v/v methanol/ 10% v/v acetic acid). After a brief decoloration in 40% v/v methanol/10% v/v acetic acid, clear bands on a blue background identified the location of the active proteinases on the gel. In order to establish the effect of pH on the proteolytic activities in the gel assays, strips of the gels, after the renaturation procedure, were incubated in different buffers containing 1.5 mM DTT (and 100 mM of sodium acetate, citrate-sodium phosphate, Tris-HCl, and pHs ranging from 3– 9, and intervals of 0.5 pH units). Ion-Exchange Chromatography Samples containing 250 midguts were homogenized in 2 ml of a solution containing 10 µM E-64 and 5 µg/ml pepstatin by using a PotterElvehjem homogenizer. The homogenate was centrifuged at 20,000g for 30 min at 4°C. The Bruchid Midgut Serine Proteinases supernatant was applied to a DEAE-Sepharose column (10 × 0.5 cm i.d.) equilibrated with 10 mM imidazole buffer, pH 6.0, containing 10 µM E-64 and 5 µg/ml pepstatin, using an Econo System (BioRad, Richmond, CA) apparatus. The column was washed with 25 ml of the same buffer and then eluted with 50 ml of a linear gradient (0-1M NaCl) in imidazole buffer, followed by 10 ml of isocratic elution in this NaCl containing buffer. The flow rate was 1.0 ml/min and fractions of 1.0 ml were collected. Recoveries of the activities from the column were 70–80%. In Vitro Proteolysis of the a-Amylase Inhibitor Proteolysis of the αAI-1 by the isolated midgut serine proteinase from Z. subfasciatus was determined by incubating 15 mg of the purified α-amylase inhibitor with the proteinase fraction in the presence of 10 µM E-64 and 5 µg/ml pepstatin at 30°C for 16 h. The reaction was stopped by addition of sample buffer and boiling the mixture for 5 min before being electrophoresed on SDS-PAGE (15% acrylamide) according to Laemmli (1970). Incubating αAI-1 with water in place of the proteinase fraction was used as a control. RESULTS Specificities of the Digestive Proteinases From Larvae of C. maculatus and Z. subfasciatus The pH optima for the hydrolysis of fluorogenic substrates by crude midgut homogenates from both bruchid species were found for all fluorogenic substrates and enzyme sources between 6–7 (Figs.1 and 2). Similar profiles were also observed for the chromogenic substrates (data not shown). Crude larval midgut homogenates of C. maculatus contain 5 times more activity than homogenates from larvae of Z. subfasciatus against Z-PheArg-MCA, while larvae of Z. subfasciatus contained 4 times more protease activity against ZArgArg-MCA than larvae of C. maculatus (Table 1). Figures for the hydrolysis of Z-Arg-MCA do not present differences in terms of absolute activity, but in terms of specific activity, C. maculatus presents the double of the activity observed for Z. subfasciatus (Table 1). Another important difference also shown in Table 1 is the hydrolysis of the substrate N-p-tosyl-GlyProArg-MCA, where larvae of Z. 21 subfasciatus contain larger absolute and specific activities (8 to 5 times, respectively) than larvae of C. maculatus. Enzymatic activities were the highest against ε-aminocaproil-LeuCys-(SBzl)-MCA, for both C. maculatus and Z. subfasciatus. The activity of C. maculatus protease was higher than Z. subfasciatus (Table 1). The activity against this substrate was not inhibited by aprotinin, whereas E-64, antipain, and leupeptin completely inhibited this activity (Tables 2 and 3). The activity against ε-aminocaproil-LeuCys-(SBzl)-MCA was enhanced by 1.5 mM DTT (Tables 2 and 3). Protease activities from the two bruchid species toward the other fluorogenic substrates, as well as the activities against azoalbumin, were highly inhibited by E-64, antipain, and leupeptin (Tables 2 and 3). Intestinal homogenates from larvae of Z. subfasciatus were two times more active toward p-nitroanilide derivatives, which are substrates for serine proteinases (N-SAAPL-pNA and NSAAPF-pNA), than C. maculatus homogenates. Conversely, higher activity against BApNA was observed in larvae of C. maculatus (Table 1). The spatial distribution of activities against the synthetic substrates was determined by assaying activities in the luminal contents and in the soluble and particulate fractions of the midgut epithelium. The results showed that 80 to 90% of the proteolytic enzymes that were tested on all the substrates were found in the luminal contents in the two bruchid species. In Gel Activities of Larval Midgut Homogenates From C. maculatus and Z. subfasciatus Profiles of gelatinolytic activity bands of intestinal homogenates from larvae of C. maculatus and Z. subfasciatus are different. It was possible to distinguish at least 5 bands from samples of C. maculatus, with RF between 0.20 and 0.60, whereas in Z. subfasciatus one major band corresponding to a RF 0.15–0.30 and several less distinct bands with RF between 0.35 and 0.45 were found (Fig. 3). The major bands in C. maculatus had a higher electrophoretic mobility than the major band of Z. subfasciatus (Fig. 3). Midgut homogenates from both bruchid species showed pH optima in the range 6–7, as assayed by mildly-denaturing electrophoresis, similarly to the determinations using synthetic 22 Silva et al. Fig. 1. Effect of pH on the hydrolysis of fluorogenic substrates by larval midgut homogenates of Callosobruchus maculatus. The used buffers were: citrate/phosphate (circles), phosphate (squares), and Tris/HCl (triangles). The results are representative of three determinations. substrates. C. maculatus exhibited very low acitivity below pH 5, while Z. subfasciatus had a diffused band that peaked at pH 5.0. The major bands of proteolytic activity from the two bruchids could be observed up to pH 9.0 (Fig. 3). Most of the proteinases of larval midgut homogenates from C. maculatus appears to be cysteine proteases, because incubating the gels inhibitor E-64 abolished the gelatin hydrolysis (Fig. 4) and incubating the gels with 1.5 mM DTT caused gelatinase activation (Fig. 5). These results are in agreement with the results obtained by incubating the enzymes with inhibitors and DTT and assaying with synthetic substrates Bruchid Midgut Serine Proteinases Fig. 2. Effect of pH on the hydrolysis of fluorogenic substrates by larval midgut homogenates of Zabrotes subfasciatus. The used buffers were: citrate/phosphate (circles), 23 phosphate (squares), and Tris/HCl (triangles). The results are representative of three determinations. TABLE 1. Digestive Proteinase Activities Present in Midgut of Larval Callosobruchus maculatus and Zabrotes subfasciatus Reared on Vigna unguiculata Seeds* Callosobruchus maculatus Substrates Z-Arg-MCA Z-ArgArg-MCA Z-PheArg-MCA Z-GlyGlyArg-MCA N-p-GlyProArg-MCA ε-LeuCys-(SBzl)-MCA SAAPL-pNA SAAPF-pNA BApNa nmol/min/gut 400 500 2,500 77 43 4,650 65 33 133 nmol/min/mg P 6,250 7,700 39,000 1,165 670 72,600 795 394 1,628 Zabrotes subfasciatus nmol/min/gut 420 2,000 570 130 500 2,500 145 75 100 nmol/min/mg P 3,200 14,600 4,400 2,027 3,800 19,200 1,771 910 1,010 *Figures are the means calculated from four assays performed in each of five different preparations obtained from 20 animals. Specific activities are expressed in nmol/min per mg of gut protein. SEM were found to be 5–10% of the means. 24 et al. TABLE Silva 2. Effect of Inhibitiors and One Activator on the Hydrolysis of Synthetic Substrates and on One Proteinaceous Substrate by Larval Midgut Homogenate From Callosobruchus maculatus* Substrates Z-ArgArg-AMC Z-PheArg-AMC ε-LeuCys-(SBzl)-MCA Azoalbumin DTT 100 776 1,084 695 E-64 0 0 0 5 Effectors (% relative activity) Aprotinin Leupeptin 50 46 100 60 0 0 0 0 Antipain 0 0 0 5 *The figures are percentages in relation to the measured activity in the controls for each substrate. In the case of DTT, the activity in the control was measured using an assay medium without the activatior, whereas for the inhibitors, the assays were made in the presence of 1.5 mM DTT. Triplicate measurements showed a variation of 2–5% for all cases. TABLE 3. Effect of Potential Inhibitors and One Activator on the Hydrolysis of the Synthetic Substrates and on One Proteinaceous Substrate by Larval Midgut Homogenate From Zabrotes subfasciatus* Substrates Z-ArgArg-AMC Z-PheArg-AMC ε-LeuCys-(SBzl)-MCA Azoalbumin DTT 100 423 2,000 142 E-64 0 0 0 23 Effectors (% relative activity) Aprotinin Leupeptin 30 45 100 78 0 0 0 24 Antipain 0 0 0 nd *The figures are percentages in relation to the measured activity in the contorls for each substrate. In the case of DTT, the activity in the control was measured using an assay medium without the activator, whereas for the inhibitor, the assays were made in the presence of 1.5 mM DTT. Triplicate measurements showed a variation of 3–5% for all cases. nd, not determined. (Table 2). Similarly, the activity observed in gel assays for larvae of Z. subfasciatus was also completely abolished by E-64 and there was a significant activation of the major band by 1.5 mM DTT (Figs. 4 and 5). Isolation of a Serine Proteolytic Enzyme From Larvae of Z. subfasciatus Fig. 3. Effect of pH on the hydrolysis of gelatin by larval digestive proteinases from Callosobruchus maculatus (A) and Zabrotes subfasciatus (B). Samples containing 0.4 gut equivalents were resolved on gelatin-containing polyacrylamide gels and allowed to hydrolyse gelatine at 30°C for 4 h (C. maculatus) and 15 h (Z. subfasciatus) at different pH values. Buffers used in the experiment were citrate/phosphate (3.0–7.0); phosphate (6.0–8.0), and Tris-HCl (8.0–9.0). The pH values from left to right were: 3.0, 4.0, 5.0, 6.0, 7.0, 7.5, 8.0, 8.5, and 9.0. A single peak of activity against the proteinase substrates Z-ArgArg-MCA or Z-GlyGlyArg-MCA was obtained by DEAE-Sepharose chromatography of intestinal homogenates of Z. subfasciatus in the chromatography performed in the presence of 20 mM E-64 and 5 µg/ml pepstatin (Fig. 6). The activity in the retained peak is higher at pH 8.8, which favors the action of serine proteinases, than at pH 6.0, which favors the action of cysteine proteinases (data not shown) and was inhibited by 250 mM aprotinin (Fig. 6), suggesting that it is a serine proteinase. Incubation of a purified α-amylase inhibitor 1 from seeds of P. vulgaris with the isolated serine proteinase fraction resulted in the cleavage of the inhibitor, as demonstrated by SDS-PAGE (Fig. 7). DISCUSSION Most of the proteolytic activities from crude midgut homogenates of the two bruchid species assayed with synthetic substrates or by using in Bruchid Midgut Serine Proteinases 25 Fig. 4. Effect of the inhibitor E-64 on the gelatinolytic activity of larval midgut proteinases from Callosobruchus maculatus and Zabrotes subfasciatus as revealed by in gel assay. After migration and a renaturation step, proteinase activities were assayed at 30°C for 4 h in 50 mM citratephosphate buffer containing 1.5 mM DTT at pH 6.0 in the absence or in the presence of 10 µM E-64. gel assays showed pH optima in the range 6–7 (Figs. 1, 2, and 3), which corresponds to the pHs found in the midguts of the following bruchids C. maculatus (Kitch and Murdock, 1986; Silva et al., 1999), Acanthoscelides obtectus (Osuala et al., 1994), and Bruchus pisorum (Lagadic, 1994). The use of in gel assays to determine pH optima showed that Z. subfasciatus possessed at least one proteinase that is stable during the mildly-denaturing electrophoresis and has a pH optimum lower than the pH found in the luminal contents. On the other hands, the major proteinases from both bruchids are maximally active at the higher pH range up to pH 9.0 (Fig. 3). Gel assays also revealed a greater diversity of proteinase forms as compared with column chromatography. Incubating the gels with 5 µg/ml pepstatin did not inhibit the gelatinase activity (data not shown). It is probable that the activity of aspartic proteinases cannot be observed on our zymograms, because these enzymes are inactivated at alkaline pHs (Barrett and Kirschke, 1981). Differences in activities using N-p-tosylGlyProArg-MCA, Z-ArgArg-MCA, N-SAAPL-pNA, Fig. 5. Effect of DTT on the gelatinolytic activities of larval midgut proteinases from Callosobruchus maculatus and Zabrotes subfasciatus using in gel assay. After migration and a renaturation step, proteinase activities were assayed at 30°C for 4 h in 50 mM citrate-phosphate buffer at pH 6.0 in the absence or in the presence of 1.5 mM DTT. and N-SAAPF-pNA and intestinal homogenates of larvae from C. maculatus and Z. subfasciatus suggest a larger amount of serine proteinase activities in Z. subfasciatus than in C. maculatus (Table 1). C. maculatus larvae contain more cysteine proteinases than larvae of Z. subfasciatus, and this was demonstrated by the hydrolysis of the substrate ε-aminocaproyl-LeuCys-(SBzl)-MCA (Table 1). The effects of the activator DTT and the inhibitor E-64 strengthen this observation (Figs. 4 and 5). The lack of activation by DTT on the activity against the synthetic substrates ZArgArg-MCA and N-CBZ-GlyProArg-MCA, in addition to the partial inhibition of these activities by aprotinin, suggest that these substrates are being hydrolyzed by serine proteinases (Tables 2 and 3). The complete inhibition of the enzymatic activities of midgut homogenates from larvae of C. maculatus and Z. subfasciatus with E-64 sug- 26 Silva et al. Fig. 6. Effect of the inhibitor aprotinin (100 mM) on the proteolytic activity (assayed with ZGlyGlyArg-MCA) of fractions obtained from the DEAE-Sepharose column. Circles, activity against Z-GlyGlyArg-MCA at pH 8.8 without inhibitor; diamonds, activity in the presence of 100 mM aprotinin. The column was equilibrated with 10 mM imidazole buffer pH 6.0 containing 10 µM E-64 and 5 µg/ml pepstatin, washed with 25 ml of the same buffer and then was eluted with 50 ml of a linear gradient (0-1M NaCl) in imidazole buffer, followed by 10 ml of isocratic elution in this NaCl containing buffer. The flux was 1.0 ml/min and fractions of 1.0 ml were collected. Recoveries of the activities applied to the column were 70–80%. Fig. 7. In vitro hydrolysis of the α-amylase inhibitor 1 from seeds of Phaseolus vulgaris by the intestinal serine proteinase fraction from Zabrotes subfasciatus isolated by ion-exchange chromatography. The amylase inhibitor was incubated with the proteolytic fraction for 16 h. The reaction was stopped by the addition of sample buffer, boiled, and run on SDS-PAGE. The gel was stained with coomassie blue dye. The arrow points to the band corresponding to nonhydrolyzed α-amylase inhibitor, while the asterisks show the hydrolytic products. Lane 1: Control without the proteolytic fraction. Lane 2: Proteolysis in the presence of the minor serine proteinase. gests the existence of cysteine proteinases in these species (Table 2, Fig. 4). Recently, it was reported that E-64 is also capable of inhibiting activities of some trypsinlike enzymes in addition to cysteine proteinases (Sreedharan et al., 1996; Novillo et al., 1997). The inhibition pattern of the cathepsins of the bruchids C. maculatus and Z. subfasciatus by E-64 is more in line with the presence of cysteine proteinases, because incubating the enzymes with the inhibitor led to a total loss of the enzymatic activity, while the minor serine proteinase detected in this paper and by Ishimoto and Chrispeels (1996) was not inhibited by E-64. The data reported in this paper using in gel assays and synthetic substrates are in agreement with our previous data showing that C. maculatus has more cysteine proteinases than Z. subfasciatus (Silva and Xavier-Filho, 1991). The presence of a serine proteinase in intestinal homogenates from Z. subfasciatus was confirmed with synthetic substrates and by column chromatography isolation of a fraction that is insensitive to E-64, inhibited by aprotinin, and shows a higher activity against Z-ArgArg-MCA and Z-GlyGlyArg-MCA at pH 8.8 than at pH 6.0. The isolated proteinase is a serine proteinase capable of degrading αAI-1 (Fig. 7) as was shown by Ishimoto and Chrispeels (1996). These Bruchid Midgut Serine Proteinases authors reported that intestinal homogenates from larvae of Z. subfasciatus contain a serine proteinase, which is capable of detoxifying the αamylase inhibitor. We determined the spatial distribution of the digestive proteinases in both bruchid species. Assays performed in samples from luminal contents and epithelium revealed that most of the digestive proteinases of both weevils are found in the luminal contents, and therefore they may have digestive or detoxifying roles of potentially toxic proteins. In summary, our results suggest that the larvae of C. maculatus rely on a larger variety of cysteine proteinases, while the larvae of Z. subfasciatus use, in addition to the aspartic and cysteine proteinases, serine proteinases. Larvae of Z. subfasciatus seem to have a greater variety in classes of intestinal proteinases and, possibly due to this fact, they are able to use many more hosts than C. maculatus. We speculated that the specialization to grow inside seeds led the bruchid beetles to lose the digestive serine endopeptidases of the chymotrypsin family (C1, as defined by Rawlings and Barrett, 1993), and that these insects rely only on the cysteine or aspartic cathepsin-like proteinases (Campos et al., 1989; Terra and Ferreira, 1994). These enzymes are appropriate to digest the seed storage proteins and are not affected by the major serine proteinase inhibitors of the seeds. The description of a minor serine proteinase in larvae of Z. subfasciatus seems to challenge this hypothesis, unless the observed serine proteinase is a cathepsin G-like proteinase that also belongs to the family of the chymotrypsins (Rawlings and Barrett, 1993). The enzyme that we isolated, which should correspond to the enzyme described by Ishimoto and Chrispeels (1996), is also capable of acting on the extended binding substrates used to assay chymotrypsin and elastase. Because these substrates are also susceptible to cathepsin G, we speculate that this minor enzyme could be a serine proteinase with a lysosomal origin, similar to the mammalian cathepsin G (Barrett, 1981). Unfortunately, the low activity level and its instability have been hindering the preparation of this enzyme in a homogeneous state and in a large enough amount to further characterization and to test this hypothesis. 27 ACKNOWLEDGMENTS We thank Dr. Richard I. Samuels for his comments on the manuscript and for the English revision. R. M. Lima is an undergraduate fellow from CNPq. C.P. Silva and W.R. Terra are staff members of their respective departments and are also research fellows of CNPq. LITERATURE CITED Barrett AJ. 1981. Cathepsin G. In: Kaplan NO, editor. Methods in enzymology. New York: Academic Press. p 561– 565. Barrett AJ., Kirschke H. 1981. Cathepsin B, cathepsin H, and cathepsin L. In: Kaplan NO, editor. Methods in enzymology. New York: Academic Press. p 535–561. Bradford MM. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analyt Biochem 72:248–254. Campos FAP, Xavier-Filho J, Silva CP, Ary MB. 1989. Resolution and partial characterization of proteinases and -amylases from midguts of larvae of the bruchid beetle Callosobruchus maculatus (F.). Comp Biochem Physiol 92B:51–57. Erlanger BF, Kokowsky N, Cohen W. 1961. The preparation and properties of two new chromogenic substrates of trypsin. Arch Biochem Biophys 95:271–278. Gatehouse AMR, Butler KJ, Fenton KA, Gatehouse JA. 1985. Presence and partial characterization of a major proteolytic enzyme in the larval gut of Callosobruchus maculatus. Entomol Exp Appl 39:279–286. Gooding RH. 1969. Studies on proteinases from some bloodsucking insects. Proc ent Soc Ont 100:139–145. Heussen C, Dowdle EB. 1980. Electrophoresis analysis of plasminogen activators in polyacrylamide gels containing sodium dodecyl sulphate and copolymerized substrates. Anal Biochem 102:196–202. Houseman JG, Downe AER. 1982. Characterization of an acidic proteinase from the posterior midgut of Rhodnius prolixus Stäl (Hemiptera: Reduviidae). Insect Biochem 12:651–655. Houseman JG, Downe AER. 1983. Cathepsin D-like activity in the posterior midgut of hemipteran insects. Comp Biochem Physiol 75B:509–512. Ishimoto M, Chrispeels MJ. 1996. Protective mechanism of the Mexican bean weevil against high levels of α-amylase inhibitor in the common bean. Plant Physiol 111:393–401. Kitch LW, Murdock LL. 1986. Partial characterization of a 28 Silva et al. major thiol proteinase from larvae of Callosobruchus maculatus F. Arch Insect Biochem Physiol 3:561–575. subfasciatus (Boh.) (Coleoptera: Bruchidae). Comp Biochem Physiol 99B:529–533. Laemmli UK. 1970. Cleavage of structural proteins during the assembly of the bacteriophage T4. Nature 227:680– 685. Silva CP, Terra WR, Xavier-Filho J, Grossi-de-Sá MF, Lopes AR, Pontes EG. 1999. Digestion in larvae of Callosobruchus maculatus and Zabrotes subfasciatus (Coleoptera: Bruchidae) with emphasis on α-amylases and oligosaccharidases. Insect Biochem Mol Biol 29:355– 366. Lagadic L. 1994. Some characteristics of digestive α-glycosidases from adults of Bruchus affinis Frölich, in relation with intestinal pH. Comp Biochem Physiol 108A:249– 253. Lemos FJA, Campos FAP, Silva CP, Xavier-Filho J. 1990. Proteinases and amylases of larval midgut of Zabrotes subfasciatus reared on cowpea (Vigna unguiculata) seeds. Entomol Exp Appl 56:219–227. Novillo C, Castañera P, Ortego F. 1997. Inhibition of digestive trypsin-like proteases from larvae of several lepidopteran species by the diagnostic cysteine protease inhibitor E-64. Insect Biochem Mol Biol 27:247–254. Osuala CI, Donner RL, Nielsen SS. 1994. Partial purification and characterization of an aminopeptidase from the bean weevil larvae Acanthoscelides obtectus Say (Coleoptera: Bruchidae). Comp Biochem Physiol 107B: 241–248. Rawlings ND, Barrett AJ. 1993 Evolutionary families of peptidases. Biochem J 290:205–218. Silva CP, Xavier-Filho J. 1991. Comparison between the levels of aspartic and cysteine proteinases of the larval midguts of Callosobruchus maculatus (F.) and Zabrotes Sreedharan SK, Verma C, Caves LSD, Brocklehurst SM, Gharbia SE, Shah HN, Brocklehurst K. 1996. Demonstration that 1-trans-epoxysuccinyl-L-leucylamido-(4guanidino)butane (E-64) is one of the most effective low Mr inhibitors of trypsin-catalysed hydrolysis. Characterization by kinetic analysis and by energy, minimization and molecular dynamics simulation of the E-64-β-trypsin complex. Biochem J 316:777–786. Terra W R, Ferreira C. 1994. Insect digestive enzymes: properties, compartmentalization and function. Comp Biochem Physiol 109B:1–62. Wieman KF, Nielsen SS. 1988. Isolation and partial characterization of a major gut proteinase from larval Acanthoscelides obtectus Say (Coleoptera: Bruchidae). Comp Biochem Physiol 89B:419–426. Xavier-Filho J, Coelho AN. 1980. Acid proteinases of Callosobruchus maculatus and proteinase inhibitors of Vigna unguiculata. Abstract from the Annual Meeting of the American Soc. Plant Physiol. and Phytochemical Soc. of North America. Plant Physiol 65:138.