Purification and Characterization of Multiple-Charged Forms of Permethrin Carboxylesterase(s) from the Hemolymph of Resistant Colorado Potato Beetle

PESTICIDE BIOCHEMISTRY AND PHYSIOLOGY 60, 31–47 (1998) ARTICLE NO. PB982330

Purification and Characterization of Multiple-Charged Forms of Permethrin Carboxylesterase(s) from the Hemolymph of Resistant Colorado Potato Beetle1 Si Hyeock Lee2 and J. Marshall Clark3 Department of Entomology, University of Massachusetts, Amherst, Massachusetts 01003 Received July 23, 1997; accepted March 16, 1998 Permethrin carboxylesterases (e.g., pI 4.2–4.8 CbEs) have been purified from the hemolymph of a permethrin-resistant strain of Colorado potato beetle, Leptinotarsa decemlineata, through several chromatographic procedures in order to determine those proteins most responsible for the DEF-synergized aspect of resistance. The pI 4.8 CbE is a 46- to 48-kDa monomeric protein whereas the pI 4.5 CbE appears to be a 57- to 59-kDa dimeric protein. Both the pI 4.8 CbE and the pI 4.5 CbE are, in fact, composed of groups of multiple-charged forms. All pI 4.5–4.8 CbEs are glycoproteins but the charge heterogeneity is not associated with N-glycan moieties. A consistent tendency of increasing affinities for the pI 4.2–4.8 CbEs as the hydrophobicity of naphthyl substrate increased implicates that the catalytic sites of all the pI CbEs have hydrophobic properties. The hydrophobic catalytic site, however, is separate entity from the domain determining overall surface hydrophobicity of CbE as determined by hydrophobic interaction chromatography of the purified pI 4.5–4.8 CbEs. Overall, pI 4.8 CbE possessed the highest affinity and catalytic efficiency for all three naphthyl substrates tested and for permethrin. The hydrolysis rates for permethrin, however, were extremely low compared to those of the naphthyl substrates. Nevertheless, the over production and the high affinity toward hydrophobic substrates of the permethrin CbE in the hemolymph in the permethrin-resistant strain of Colorado potato beetle is likely to result in efficient hydrolysis of permethrin. In general, the PI 4.8 and 4.5 CbEs share a number of similarities in their biochemical properties and functional role in permethrin resistance despite their distinct molecular properties. The availability of highly purified and characterized forms of permethrin CbE will allow the development of efficient immunochemical and DNAbased diagnostics for the monitoring of permethrin resistance in field populations of beetles. q1998 Academic Press

INTRODUCTION

determined to be principally due to increased esterase activity (i.e., permethrin carboxylesterase) and site insensitivity associated with ner-

The Colorado potato beetle [CPB4; Leptinotarsa decemlineata (Say)] was one of the first insect pests to develop resistance to permethrin and other pyrethroids (1). Due to high levels of resistance, pyrethroids are no longer effective in controlling resistant CPB populations (2). The rapid loss of permethrin as an effective insecticide for the control of CPB was due to multiple resistance factors (3–6). The permethrin resistance in a near isogenic permethrin-resistant (PE-R) strain of CPB was

4 Abbreviations used: CbE(s), carboxylesterase(s); ConA, concanavalin A; CPB, Colorado potato beetle; DEAE–ion exchange, diethylaminoethyl-anionic exchange; 2-DGE, two-dimensional gel electrophoresis; DTE, dithioerythritol; HIC, hydrophobic interaction chromatography; IEF, isoelectric focusing; aNB, a-naphthyl butyrate; aNC, a-naphthyl caproate; PAGE, polyacrylamide gel electrophoresis; permethrin CbE(s), carboxylesterase(s) associated with permethrin resistance, principally referring to the pI 4.2–4.8 CbEs; pI 4.2–4.8 CbEs, a group of permethrin CbE(s) that have pI values of 4.2–4.8 as determined by chromatofocusing or isoelectric focusing; PE-R, permethrin-resistant strain of CPB nearly isogenic to the SS strain; PHMB, para-hydroxylmercuribenzoate; PMSF, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate; SE-HPLC, size exclusion– high-performance liquid chromotography; SS, susceptible laboratory strain of CPB.

1

This work was supported by USDA Regional Research Project NE-180, Biochemical and Molecular Mechanisms of Insecticide Resistance. 2 Current address: Department of Entomology, NYSAES, Cornell University, Geneva, NY 14456. 3 To whom correspondence should be addressed. 31

0048-3575/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.

32

LEE AND CLARK

vous system (5, 7, 8). Permethrin resistance was substantially synergized by DEF both in the multiply resistant field strain and in the PE-R strain, which had been extensively back-crossed the susceptible strain. An increased level of carboxylesterase activity was associated with the PER strain resulting in a 1.5-fold greater hydrolytic rate of a-naphthyl butyrate. Additionally, the PE-R strain produced significantly more hydrolytic metabolites of 14C-labeled permethrin when incubated under in vivo and in vitro conditions compared to the nearly isogenic susceptible strain. Most of the permethrin carboxylesterase (permethrin CbE) was distributed in hemolymph and was composed mainly of pI 4.2–4.8 CbEs (9). The pI 4.2–4.8 CbEs were more abundant in the PE-R strain and suggested that resistance was due to the sequestration of permethrin by the over produced CbEs followed by a low rate but efficient hydrolytic detoxification of permethrin (9). The present studies were undertaken to further characterize the biochemical and molecular properties of the highly purified forms of permethrin CbE from the hemolymph of CPB, which are most responsible for the DEF-synergized aspect of resistance. Once isolated, specific proteins can be used to generate molecular probes (e.g., antibodies, DNA) for resistance monitoring of field populations of CPB. MATERIALS AND METHODS

Insect Strains The susceptible (SS) and a near isogenic PER strains of CPB have been maintained for the past 10 years in the laboratory as described previously (9). Chemicals [14C]trans-Permethrin (methylene labeled, sp act 57 mCi/mmol) was a gift from FMC (Princeton, NJ). Unlabeled trans-permethrin (93% pure) was obtained from ICI (Goldsboro, NC). [3H]Diisopropylfluorophosphate (DFP, sp act 6.0 mCi/mmol) was purchased from New England Nuclear (Boston, MA). Chromatographic media were purchased from Sigma

Chemical Co. (St. Louis, Mo), Bio-Rad (Hercules, CA), or Pierce (Rockford, IL). Additional biochemicals were purchased from Sigma Chemical Co., except when stated otherwise. Enzyme Purification In following chromatographic purification procedures, the protein profile was monitored continuously at 280 nm using a UV monitor and CbE activities were evaluated for each fraction using a-naphthyl caproate (aNC) or a-naphthyl butyrate (aNB). Sample concentrations during purification procedures were conducted either with stirred cell concentrator or with Centricon 30 concentrator (Amicon, Beverly, MA). Hemolymph was collected from 100–150 individual fourth instars of the PE-R strain as described previously (9). Ammonium sulfate fractionation of hemolymph was accomplished by collecting the protein that precipitated at 40– 80% saturation level of ammonium sulfate and resuspended into 2 ml of 0.1 M sodium phosphate buffer (pH 7.0). The ammonium sulfate-precipitated sample was applied to a methyl–hydrophobic interaction chromatography (HIC) column (Bio-Rad) previously equilibrated with binding buffer (0.1 M sodium phosphate buffer containing 1.7 M ammonium sulfate, pH 7.0). The sample was eluted by a linear gradient of ammonium sulfate (i.e., 1.7 to 0 M). Fractions with relatively high levels of aNC hydrolysis were subsequently assayed for [14C]trans-permethrin hydrolysis activity. Fractions containing high levels of [14C]trans-permethrin hydrolysis were concentrated and dialyzed against ion-exchange buffer (10 mM imidazole, 10 mM NaCl, 1 mM DTE, pH 7.2) at 48C overnight. For the HIC of the purified permethrin CbEs, the rechromatofocused pI 4.8 and 4.5 CbEs samples (see chromatofocusing section) were dialyzed against 0.1 M sodium phosphate buffer (pH 7.0) overnight at 48C. The dialyzed samples were amended with ammonium sulfate (90% saturation) and concentrated. Samples were applied onto the methyl-HIC Econo-Pac Cartridge (Bio-Rad) previously equilibrated with

PERMETHRIN CARBOXYLESTERASES IN COLORADO POTATO BEETLE

the binding buffer and eluted with 0.1 M sodium phosphate buffer (pH 7.0) in the presence of a decreasing linear gradient of ammonium sulfate (i.e., 1.7 to 0 M). To determine the inhibition specificities to r-hydroxymercuribenzoate (PHMB) and phenylmethylsulfonyl fluoride (PMSF), individual fractions possessing high CbE activities were preincubated with PHMB or PMSF (125 mM final assay conc.) for 10 min and then incubated with aNB or aNC (125 mM final assay concentration for 10 min at 378C. The dialyzed HIC sample was loaded onto a diethylaminoethyl-anionic-exchange (DEAE) Bio-Gel A column (Bio-Rad) previously equilibrated with ion-exchange buffer and eluted by a linear NaCl gradient (i.e., 10–500 mM in 10 mM imidazole buffer containing 1 mM DTE, pH 7.2). Fractions with high CbE activities were concentrated and applied onto a Sephacryl S200-HR column (Pharmacia) and eluted with elution buffer (10 mM imidazole, 50 mM NaCl, 1 mM DTE, 4% glycerol, pH 7.2). The fractions with high CbE activities were concentrated and loaded onto a PBE94 chromatofocusing column (Pharmacia) previously equilibrated with 25 mM imidazole buffer (pH 7.4) containing 5% glycerol. The elution, evaluation of hydrolysis activity, and other procedures were the same as described previously (9). The CbEs associated with the pH 4.2–4.8 fractions were designated as the pI 4.2–4.8 CbEs and comprise the permethrin CbE(s) as discussed previously (9). To obtain better separation of pI 4.2–4.8 CbEs, the pH 4.0–4.9 fractions were combined and rechromatofocused using a narrower pH gradient (i.e., pH 4–5). To identify the PHMB-sensitive or PMSF-sensitive fractions, individual fractions were preincubated with PHMB or PMSF and CbE activity was determined as above. For further purification using preparative native polyacrylamide gel electrophoresis (PAGE), the pH 4.0–4.9 chromatofocused fractions were concentrated and dialyzed against electrophoresis sample buffer (60 mM Tris–HCl, 5% glycerol, pH 6.8) overnight at 48C. The dialyzed sample was concentrated and loaded onto a 1.5-mm-thick slab gel (7.5% separating gel

33

and 4.0% stacking gel). Electrophoresis was carried out at 200 V for 45–50 min. The gel was stained with aNB/aNC-dianisidine as described previously (9), and CbE bands were cut out and electroeluted according to the manufacturer’s instructions (Bio-Rad electroeluter Model 422). Two-Dimensional Gel Electrophoresis (2-DGE) The overall procedure for 2-DGE was based on the manufacturer’s instruction manual for the Bio-Rad Mini-Protein II 2-D cell system. Isoelectric focusing (IEF) was performed at 750 V for 3 h. The second dimensional gel (10% SDS–PAGE) was electrophoresed at 200 V for 50 min and then silver stained (10). Determination of Molecular Mass (Mr) Sephadex G-100 gel filtration chromatography (Pharmacia) and Bio-Sil SEC 250 sizeexclusion HPLC (SE-HPLC, Bio-Rad) were employed for the determination of native Mr of purified permethrin CbE. In both cases, fractions were assayed for CbE activity by aNB or aNC hydrolysis in the presence or absence of specific inhibitors, PHMB, and PMSF as described above. Mrs were estimated using a calibration curve (i.e., Mr vs elution volume) generated from a set of molecular weight standard proteins. The Mr of denatured CbE was determined by SDS–PAGE (10% separating gel). The denatured CbE sample and standard molecular weight marker proteins (SDS-6H, Sigma Chemical Co.) were electrophoresed at 200 V under constant voltage conditions for 45–50 min. The gel was stained with Coomassie brilliant blue and the Mrs were estimated from a standard curve of molecular weight marker proteins (i.e., Mr vs relative mobility). Labeling of Permethrin CbE with [3H]DFP [3H]DFP (6.0–30 mCi, sp act 6.0 mCi/mmol) was mixed with purified permethrin CbE (i.e., rechromatofocused pI 4.2–4.8 CbEs, 37–200 mg protein) and incubated at room temperature for 20 h. The unbound [3H]DFP was removed from the labeling reaction mixture using a Bio-Spin

34

LEE AND CLARK

6 desalting column (Bio-Rad) previously equilibrated with PAGE sample buffer (60 mM Tris– HCl, pH 6.8). The [3H]DFP-labeled CbE sample and unlabeled CbE sample were electrophoresed side by side on a 12% SDS–polyacrylamide gel at 200 V for 50 min. The gel lane with the [3H]DFP-labeled CbE sample was excised from the gel, cut into ca. 1.5-mm-wide pieces, and placed individually into 200 ml of 1 N NaOH overnight to base-hydrolyze protein. After neutralization, the radioactivity was determined by scintillation counting. The gel lane containing unlabeled sample was stained with Coomassie brilliant blue to determine the location of CbE band. Concanavalin A (ConA)–Sepharose Affinity Chromatography The pH 4.2–4.8 fractions from chromatofocusing were concentrated and mixed with ConA binding buffer (50 mM Tris–HCl, 0.1 M NaC1, 1 mM CaCl2, 1 mM MnCl2, 5% glycerol, pH 7.1). Following buffer exchange, the sample was applied onto the ConA–sepharose column (Sigma Chemical Co.) and the column was washed with binding buffer. The proteins bound to ConA–sepharose were eluted with elution buffer (binding buffer containing 0.5 M mannopyranoside). N-Deglycosylation of Permethrin CbE The purified permethrin CbE sample (i.e., chromatofocused pI 4.2–4.8 CbE sample, ca. 70 mg protein) was mixed with digestion buffer (50 mM potassium phosphate, 20 mM EDTA, pH 7.5) and then concentrated to ca. 20 ml. The CbE sample was incubated with 20–30 U of N-glycosidase F/endoglycosidase F (Boeringer Mannheim Biochemicals, Indianapolis, Indiana) at 378C for 20 h. For a control sample, CbE was incubated at 378C for 16 h without the Nglycosidase F. After deglycosylation, the CbE samples were analyzed by native IEF followed by aNB/aNC-dianisidine staining as described previously (9).

Determination of Optimum pH and Temperature Conditions for Hydrolytic Activity of Permethrin CbE In pH experiments, several types of reaction buffer were used to ensure maximum buffer capacity for each pH condition. CbE activities were assayed with aNC (125 mM final concentration) prepared in different pH reaction buffers (0.1 M glycine–HCl, pH 3.0; 0.1 M citric acid– NaOH, pH 4.0 and 5.0; 0.1 M sodium phosphate, pH 6.0 and 7.0; 0.1 M Tris–HCl, pH 8.0 and 9.0; 0.1 M glycine–NaOH, pH 10.0) at 378C for 10 min. Since high levels of nonenzymatic hydrolysis were observed at high pH (e.g., pH 8.0–10.0), control reactions without CbE under each pH condition were carried out at the same time to subtract the nonenzymatic hydrolysis. In temperature experiments, CbE activities were assayed with aNC (125 mM final concentration) at 10, 20, 30, 40, 50, and 608C for 10 min. Kinetic Analyses of Naphthyl Substrates Three representative pH fractions from chromatofocusing (i.e., pH 4.8, 4.5, and 4.2 fractions) were used as the pI 4.8, 4.5, and 4.2 CbEs for the kinetic analyses, respectively. Three naphthyl substrates were selected to investigate relationships between increasing substrate hydrophobicity (i.e., aNC.aNB.aNA) and the affinity (i.e., 1/Km) and catalytic (i.e., Vmax) constants of the three charged forms of CbE. The CbE activities were assayed at a range of substrate concentrations (25, 37.5, 62.5, 125, and 250 mM final concentration in 50 mM sodium phosphate containing 0.05% Triton X-100, pH 7.2) at 328C for 3 min. Other procedures were the same as previously described (9). The data were analyzed by construction of LineweaverBurk double reciprocal plots and the Km and Vmax kinetic constants were determined graphically. RESULTS

Purification of Permethrin CbE (i.e., pI 4.2–4.8 CbEs) The CbEs associated with hemolymph were typically precipitated in the 40–80% saturation

35

PERMETHRIN CARBOXYLESTERASES IN COLORADO POTATO BEETLE

range of ammonium sulfate as judged by aNC hydrolysis (Table 1). The broad precipitation range of this activity implied the presence of a heterogeneous group of CbEs. Three CbE activity peaks (i.e., peak I, II, and III in the order of elution) were found by methylHIC (Fig. 1A). The peak I, II, and III fractions accounted 27.8, 32.3, and 39.9% of total aNC hydrolysis activity, respectively. Since the hydrophobicity of the fraction is proportional to the order of elution in HIC, the CbEs associated with peak III (i.e., peak III CbE) is the most hydrophobic. The peak III CbE was also determined to possess the highest level of [14C] transpermethrin hydrolysis activity when compared to peaks I and II (i.e., 21.8 versus 6.7 and 1.3 pmol permethrin hydrolyzed/h/ml, respectively). Thus, peak III CbE was processed for further purification. The purification factor of the methyl HIC step for peak III CbE was 1.6fold (i.e., 2.9/1.8, Table 1). When the peak III CbEs were separated by DEAE–ion-exchange chromatography, the major peak of CbE activity was eluted over a 250–280 mM NaCl concentration range (i.e., fraction 15–22, Fig. 1B), which indicates an overall negative-charged property of the permethrin CbE at physiological pH 7.2. The purification factor of the DEAE–ion-exchange step was 5.6-fold (i.e., 16.2/2.9, Table 1). A single peak of aNC hydrolysis activity was observed following gel filtration and its Mr was estimated as ca. 60 kDa (Fig. 1C). A 2.2-fold

purification factor was obtained from the gel filtration chromatography step (i.e., 35.5/16.2, Table 1). The majority of CbE activity was eluted over a pH 4.0–5.0 range following chromatofocusing (Fig. 1D). Within this pH range, a number of CbE activity peaks were noted (i.e., pI 4.2–4.8 CbEs). The presence of multiple peaks of CbE activity correlated well with the sawtooth-like protein elution profile (i.e., fraction 40–56, Fig. 1D). These findings corroborate the presence of multiple-charged forms of permethrin CbE as previously demonstrated in the analytical IEF of hemolymph (9). The purification factor of this step was ca. 1.3-fold (i.e., 46.7/35.5, Table 1). Prior to estimation of Mrs, the pI 4.2–4.8 CbEs were further purified either by preparative native PAGE followed by electroelution or additional chromatofocusing with narrow pH range (i.e., pH 4–5). The overall purification factors for the purification scheme were 51- and 77-fold as judged by aNC and [14C]trans-permethrin hydrolysis, respectively (Table 1). Native Mr(s) of Permethrin CbE When a purified permethrin CbE sample (i.e., rechromatofocused pI 4.2–4.8 CbEs) was separated by gel filtration chromatography in the presence and absence of PHMB and PMSF, three CbEs with different Mrs (i.e., 59-, 49-, and 32kDa CbEs) were identified (Fig. 2A). Without the specific inhibitors, the three CbEs were not

TABLE 1 Purification of Permethrin CbE from Hemolymph of the Fourth Instar of CPB aNC hydrolysis

Step Hemolymph Ammonium sulfate HIC Ion exchange Gel filtration Chromatofocusing Preparative PAGE

Total protein (mg)

Total activity (nmol/min)

Specific activity (nmol/min/ mg protein)

Yield (%)

185.3 69.2 13.9 1.6 0.5 0.25 0.10

8770 6060 1930 1220 836 540 237

47.3 87.5 138.4 768.2 1680 2160 2390

100 69 22 14 10 6 3

Permethrin hydrolysis

Purification

Total activity (pmol/hr)

Specific activity (pmol/hr/ mg protein)

Yield (%)

Purification

1 1.8 2.9 16.2 35.5 46.7 50.5

99.8 — — — — 5.6 —

0.29 — — — — 22.2 —

100 — — — — 10 —

1 — — — — 76.6 —

36

LEE AND CLARK

FIG. 1. Chromatographic purification of permethrin CbEs (pI 4.2–4.8 CbEs), from hemolymph of the permethrinresistant (PE-R) strain of CPB. The fractions marked with “111” were collected and processed in the next purification step.

PERMETHRIN CARBOXYLESTERASES IN COLORADO POTATO BEETLE

37

FIG. 2. Estimation of native molecular mass (Mr) of the permethrin CbE by Sephadex G-100 gel filtration chromatography (A) and by size-exclusion (SE)-HPLC (B). The aNC hydrolysis activity for each fraction was determined in the presence or absence of PHMB and PMSF.

well resolved (see “aNC only” line in Fig. 2A). In a similar manner, only a single 60-kDa CbE peak was found using gel filtration chromatography (see Fig. 1C). To confirm the native Mrs of the permethrin CbE, fractions containing the 59-, 49-, and 32kDa CbEs were separated by SE–HPLC in the presence and absence of PHMB and PMSF. Similar results were obtained in which a 57- and a 50-kDa CbEs were identified (Fig. 2B). In the absence of the specific inhibitors, the CbE peaks

was detected as a single peak. The smaller 32kDa CbE was not identified by SE–HPLC. In summary, the permethrin CbE was determined to be comprised of at least three separate molecular forms (i.e., 57- to 59-, 49- to 50-, and 32-kDa CbE) under native conditions. The 57to 59- and 49- to 50-kDa forms of CbE were most predominant as judged by aNC hydrolytic activity and protein quantity in the gel filtration chromatography and SE–HPLC separations (note: protein profile as determined by UV280 is

38

LEE AND CLARK

not shown). The 57- to 59- kDa CbE was more sensitive to inhibition by PMSF whereas the 49to 50- and 32-kDa CbEs were more sensitive to inhibition by PHMB. When the selective inhibition by PMSF and PHMB was examined for the fractions from a rechromatofocusing step, the pI 4.5 and pI 4.8 CbEs were determined to be selectively inhibited by PMSF and PHMB, respectively (Fig. 3). Thus, on the basis of the selective inhibition profiles by PHMB and PMSF, the native Mrs of the pI 4.8 and 4.5 CbEs appear to be primarily 49–50 and 57–59 kDa, respectively. Denatured Mr(s) of Permethrin CbE Three major protein bands (i.e., designated as band I, II, and III; ca. 31-, 46- to 48-, and 59kDa proteins, respectively) and trace amounts of minor protein bands (i.e., 39-, 43-, 44-, 67-, 73-, 77-, and 114-kDa proteins) were identified from a purified pI 4.2–4.8 sample following SDS–PAGE (Fig. 4A). The proteins associated with major bands I, II, and III accounted for ca.

46, 33, and 16% of the total amount of proteins as judged by scanning densitogram, respectively (Fig. 4B). No differences in separation patterns were evident in the absence or presence of reducing agent, b-mercaptoethanol (see lanes “2b” and “1b,” respectively), indicating that none of the bands I, II, and III proteins are covalently associated polymeric proteins. Because the Mrs of the three major proteins identified by SDS–PAGE matched well with the Mrs of three native molecular forms of CbE identified previously by gel filtration, the three major protein bends (i.e., band I, II, and III proteins) appear to be the denatured forms of permethrin CbE. To verify this finding, a [3H]DFP-labeling experiment was conducted. High levels of radioactivity (cpm) were detected in five zones of the gel including the 30-kDa zone (gel slice 20–27, 8280 cpm), the 57-kDa zone (gel slice 10–13, 5300 cpm), and the 46-kDa zone (gel slice 15– 17, 1860 cpm) (gel not shown). This finding confirms that all the band I, II, and III proteins (i.e., 30-, 46-, and 57-kDa proteins, in this case)

FIG. 3. Identification of the PHMB-sensitive and PMSF-sensitive CbEs following rechromatofocusing of the purified permethrin CbE. The purified pI 4.2–4.8 CbE were rechromatofocused using a narrow pH gradient (pH 4–5) and the aNC hydrolysis activity for each fraction was determined in the presence or absence of PHMB and PMSF.

PERMETHRIN CARBOXYLESTERASES IN COLORADO POTATO BEETLE

39

FIG. 4. SDS–PAGE of the purified permethrin CbE (A) and densitographic quantification (B) are presented. In the densitogram, the horizontal axis is optical density (arbitrary units) and vertical axis is gel distance (70 mm, running top to bottom). The “2b” and “1b” represent the absence and presence of b-mercaptoethanol, respectively. The major band proteins were designated as bands I, II, and III on the right side of the densitograms.

are true esterases and most likely the denatured forms of the permethrin CbEs. Multiple-Charged Forms of Permethrin CbE Each major protein band identified by SDS– PAGE was further determined to be composed of several proteins with very similar Mrs but different pI values (i.e., multiple-charged forms) through denatured 2-DGE (Fig. 5). By using the median pI value of each major pI protein group, the molecular compositions of the denatured pI 4.8, 4.5, and 4.2 CbEs can be described. The pI 4.8 CbE appears to be composed of at least six to seven charged proteins with similar Mrs of ca. 46–48 kDa). The pI 4.5 CbE appears to be composed of at least four charged proteins with similar Mrs of ca. 31 kDa. The pI 4.2 CbE appears to be composed of at least two charged proteins with similar Mrs of ca. 30–33 kDa.

Glycoprotein Nature of Permethrin CbE Purified permethrin CbE (i.e., pI 4.2–4.8 CbEs) was separated by ConA–sepharose affinity chromatography. No CbE activity was found in the eluted fractions even after extensive washing with binding buffer (fractions 1–12, Fig. 6). Bound CbE was eluted when binding buffer was amended with high concentration of mannopyranoside (i.e., 0.5 M, fractions 13–32). The CbE, however, was not eluted as a single sharp peak but eluted as several (3–4) contiguous peaks. This finding is suggestive of the possibility that the pI 4.2–4.8 CbEs are composed of several forms that differ in their overall levels of glycosylation (11–12). To determine whether the glycosylation levels are directly related with the overall level of negative charge, the permethrin CbE was N-deglycosylated using F/N-glycosidase F. As judged by the separation patterns of CbE bands on the IEF

40

LEE AND CLARK

FIG. 5. Two-dimensional gel electrophoresis (2-DGE) of the purified permethrin CbE by denatured IEF vs SDS–PAGE. The pI values determined by pI marker proteins are shown on the top of the gel. The molecular mass of marker proteins are shown on the right side of the gel.

gel, neither a shift in the position of the CbE bands nor a change in the number of CbE bands was observed (data not shown). These results suggest that the charge heterogeneity of the permethrin CbE is not due to the degree of Nglycosylation. Effects of pH and Temperature on Permethrin CbE Optimum pH for aNC hydrolysis activity of the pI 4.8 and 4.5 CbE was approximately 7 for both CbEs (Fig. 7A). The hydrolytic reaction of the pI 4.5 CbE showed a typical bell-shaped pH profile implicating the involvement of at least two ionizable groups in catalysis (13). The pI 4.8 CbE had an elevated level of hydrolytic activity at higher pH (i.e., pH 7–10) when compared with the pI 4.5 CbE. At lower pH (i.e., pH 3–7), however, their activity profiles closely resembled each other. Both pI 4.8 and 4.5 CbEs showed the highest activity at ca. 508C (Fig.

7B). However, the pI 4.5 CbE lost ca. 50% of its activity at 608C, whereas the pI 4.8 CbE activity was relatively unaffected at this temperature. Naphthyl Substrate Specificities and [14C] Permethrin Hydrolysis Activities of the pI 4.8, 4.5, and 4.2 CbEs A consistent tendency was established with all three substrates that results in decreased Km values as the negative charge associated with the three pI forms of the CbE decreased (Fig. 8A). When the Km values for each substrate were compared using an individual pI CbE form, a consistent tendency of decreasing Km values was observed as the hydrophobicity of substrate increased, regardless of the charge state of the CbE. This result indicates that the catalytic sites for all three pI CbEs are hydrophobic so that they exhibit higher affinities (i.e., lower Km values) toward the more hydrophobic substrates.

PERMETHRIN CARBOXYLESTERASES IN COLORADO POTATO BEETLE

41

FIG. 6. Concanavalin A (ConA)–Sepharose affinity chromatography of the permethrin CbE (i.e., pI 4.2–4.8 CbEs).

A consistent tendency of increasing Vmax values was observed as the negative charge of the pI CbEs decreased, regardless of the naphthyl substrate used (Fig. 8B). This consistent tendency suggests that the pI 4.8 CbE is more efficient in the hydrolysis of these naphthyl substrates. When the Vmax values for each substrate were compared using an individual pI CbE form, the Vmax values decreased as the carbon number of the substrate increased, regardless of the level of negative charge of the CbE. These results indicate that the hydrolytic rate of the CbEs is reduced as relatively more hydrophobic substrate, such as aNC, are used in spite of their relatively higher affinity to the more hydrophobic substrate. This trend was also evident in the hydrolysis of a-naphthyl caprylate (i.e., C8 acyl naphthyl ester) where the Vmax and Km values of the combined pI 4.2–4.8 CbEs were ca. 0.13 mmol/min/mg protein and 13 mM, respectively (data not shown). As seen in the hydrolysis of naphthyl substrates, the pI 4.8 CbE also possessed the highest [14C]trans-permethrin hydrolytic activity (Table

2). However, the [14C]permethrin hydrolysis rates by the permethrin CbEs were ca. 0.5–1 3 106 times lower when compared to the Vmax values obtained with aNC. Characterization of the Hydrophobic Nature of the pI 4.8 and 4.5 CbEs Three CbE peaks (i.e., HIC peak I, II, and III CbEs) were commonly identified from the methyl-HIC of both the PI 4.8 CbE (i.e., 4.8 HIC, Fig. 9A) and the pI 4.5 CbE (i.e., 4.5 HIC, Fig. 9B). The presence of multiple CbEs with different hydrophobicities was indicated by the unique separation pattern. The HIC peak II CbE was the most predominant form of the pI 4.8 CbE from the 4.8 HIC while the HIC peak III CbE was the most predominant form of the pI 4.5 CbE from the 4.5 HIC. The HIC peak II CbE from the 4.8 HIC fractionation possessed the highest specific hydrolytic activity toward aNC (Fig. 9A), a property associated with the pI 4.8 CbE as judged by the ratio of Vmax/Km constants (determined from Fig. 8). The HIC

42

LEE AND CLARK

FIG. 7. Effects of pH (A) and temperature (B) on aNC hydrolysis by the pI 4.5 and 4.8 CbEs.

peak III CbE from both the 4.8 HIC and the 4.5 HIC was more specific in its hydrolytic activity toward aNB (Figs. 9A and 9B), a property associated with the pI 4.5 CbE as judged by the ratio of Vmax/Km constants (determined from Fig. 8). As shown in Fig. 9C, the HIC peak II CbE was more selectively inhibited by PHMB than PMSF whereas the HIC peak III CbE was more selectively inhibited by PMSF than by PHMB. The HIC peak I CbE was less hydrophobic than the HIC peak II CbE (Figs. 9A and 9B)

but was similar to HIC peak II CbE in that it was more specifically inhibited by PHMB. The HIC peak IV CbE was detected only in the 4.8 HIC (Fig. 9A). It possessed more specific hydrolysis activity toward aNC than aNB but was only slightly inhibited by either PHMB or PMSF. The results on substrate and inhibition specificities indicate that HIC peak II CbE and HIC peak III CbE are the major representative forms of CbE present in the pI 4.8 and 4.5 CbE samples, respectively. Using HIC, the pI 4.8 CbE,

PERMETHRIN CARBOXYLESTERASES IN COLORADO POTATO BEETLE

43

FIG. 8. Kinetic analysis of the pI 4.8, 4.5, and 4.2 CbEs using three naphthyl substrates, aNA, aNB, and aNC. (A) Comparisons of Km values; (B) comparisons ofVmax values. Each bar represents the mean 6 SD obtained from four separate determinations.

which has the most hydrophobic catalytic center amongst the CbEs compared, was actually found to be less hydrophobic in terms of surface hydrophobicity than the pI 4.5 CbE. These results indicate that the pI 4.8 form of permethrin CbE should be more freely soluble in the hemolymph but sequester and hydrolyze permethrin to a greater extent than the pI 4.5 form. DISCUSSION

It is apparent from our results that the pI 4.8 and 4.5 forms of permethrin CbE, which are

overexpressed in the hemolymph of permethrinresistant CPB (9), contribute significantly to the sequestration and hydrolysis of hydrophobic esters, such as permethrin. In this manner, the pI 4.8 and 4.5 CbEs are likely responsible for the DEF-synergistic aspect of permethrin resistance in CPB (3). The relatively low purification factor (51- and 77-fold) appears to be largely the result of isolating only the pI 4.2–4.8 CbEs from the entire CbE pool through multiple purification steps, particularly the initial methyl-HIC step and the

44

LEE AND CLARK

FIG. 9. Determination of surface hydrophobicities of the pI 4.8 (A) and 4.5 (B) CbEs by methyl-hydrophobic interaction chromatography (methyl-HIC). The pI 4.8 and 4.5 CbE samples from rechromatofocusing were separated individually by methyl-HIC and hydrolysis activities using aNB and aNC as substrates were determined. The major CbE activity peaks were marked as I, II, III, and IV in an increasing order of hydrophobicity. (C) Inhibition of the HIC peak I, II, III, and IV CbEs by PHMB and PMSF with aNC as substrate.

PERMETHRIN CARBOXYLESTERASES IN COLORADO POTATO BEETLE TABLE 2 [14C]trans-Permethrin Hydrolysis by the pI 4.2, 4.5, and 4.8 CbEs [14C]trans-Permethrin Hydrolysis (pmol/h/mg protein) pI 4.8 CbE

pI 4.5 CbE

pI 4.2 CbE

70.4 6 20.8

65.8 6 14.5

17.7 6 2.6

Note. Values represent the mean 6 SD (n 5 3).

hydrolysis of aNC by other nonspecific esterases (14–16). Also, the relatively low purification level is indicative of the relatively high abundance of the permethrin CbEs. Since 0.1 mg of permethrin CbE was purified from 185 mg of crude hemolymph proteins with an approximate 3% yield, the actual content of the permethrin CbE in the PE-R hemolymph would be approximately 1.8% if the final preparation of the pI 4.2–4.8 CbE is 100% pure (i.e., 0.1/ 185/0.03 3 100 5 1.8%). An esterase whose primary function is sequestration typically constitutes large portion of total proteins in an insect body. The sequestration esterases from the aphid Myzus persicae and the mosquito Culex quinquefasciatus account for as much as 3 and 12% of total protein, respectively, in highly resistant strains (17, 18). Under native conditions, the 49- to 50-kDa and 57- to 59-kDa CbEs were quantitatively the most predominant forms with the smaller form (i.e., 32-kDa CbE) less prominent (Fig. 2). The most predominant form of denatured protein, however, was the 25- to 31-kDa protein (i.e., band I protein), followed by the 46- to 48-kDa protein (i.e., band II protein), and the 57- to 60kDa protein (i.e., band III protein) (Fig. 4). Thus, the amount of the 25- to 31-kDa CbE is increased while the amount of 57- to 59-kDa CbE is decreased under denaturing conditions whereas the reverse is found under native conditions. These findings suggest that majority of the 57to 59-kDa CbE may be converted into the 25to 31-kDa protein under denaturing conditions whereas the amount of the 49- to 50-kDa CbE was present at similar levels under both native and denaturing conditions. Thus, the pI 4.8 CbE

45

appears to be the 49- to 50-kDa monomeric protein, while the pI 4.5 CbE appears to be the 57- to 59-kDa noncovalent dimeric protein. The native Mrs of permethrin CbEs are similar to carboxylesterases from other insects, which range from 59 to 110 kDa (14, 18–22). However, dimeric association of small subunits (i.e., 25–31 kDa), as proposed for the pI 4.5 CbE, has not been previously reported in animal species. One similar case has been reported, nevertheless, for a CbE from an aerobic bacterium Pseudomonas fluorescens, which was composed of two identical subunits, each with a Mr of 23 kDa (23). Each charged form of permethrin CbE (i.e., pI 4.8, 4.5, and 4.2 CbE) is composed of two to seven proteins with slightly different pI values as determined from 2-DGE. The functional significance of the charged forms of each pI CbE in permethrin resistance, however, is not known at this point. The permethrin CbE is also glycosylated as demonstrated by ConA–Sepharose chromatography. Glycosylated CbEs have been reported in other insects including M. persicae and C. quinquefasciatus (24, 25). The N-glycan moieties, however, are not associated with the charge heterogeneity of the permethrin CbE as demonstrated by N-deglycosylation experiments. Possible means resulting in the charge heterogeneity include the polypeptide moieties themselves, O-glycans, and posttranslational modifications, such as phosphorylation/dephosphorylation, amidation/deamidation, and sulfation/desulfation (26, 27). From the kinetic analyses of various naphthyl substrates, Km values for the charged forms of permethrin CbEs (i.e., pI 4.8, 4.5, and 4.2 CbEs) were similar to those for other purified CbEs (14, 18–22, 28). However, the Vmax values were approximately 120–440, 53, and 50 times lower than those from C. quinquefasciatus (21), Nilaparvata lugens (14), and Lucilia cuprina (28), respectively, but 16 times higher than that from Laodelphax striatellus (22). From these comparisons, permethrin CbEs of CPB appear to be relatively poor hydrolases for these artificial substrates when compared with other insect CbEs. The consistent tendency of increased affinities

46

LEE AND CLARK

for all three pI CbEs as the hydrophobicity of naphthyl substrate increase implies that the catalytic sites of all the pI CbEs have hydrophobic properties. Hydrophobicity has been shown to be a common characteristic of the catalytic sites of the rat plasma esterase, ES-1A, (29), the E4 esterase from M. persicae (30), and an esterase from Triatoma infestans (31). The hydrophobicity of the catalytic site of the pI 4.8 and 4.5 CbEs, however, was determined not to be associated with overall surface hydrophobicity of the permethrin CbE, as judged from the behavior of the pI 4.8 and 4.5 CbE in the HIC. The separate nature of the hydrophobicity of the catalytic site from the overall surface hydrophobicity allows the possibility that the catalytic site of the permethrin CbE is concealed from the surface of enzyme, as proposed for several lipases (32–34). The hydrolytic activity of the permethrin CbEs decreases with increasing hydrophobicity of the substrate in spite of an increase in affinity. The decreasing hydrolytic activity toward the substrates with increasing acyl carbon chain indicates that the permethrin CbEs prefer to hydrolyze substrates with shorter acyl chain. This relationship is also in keeping with the low hydrolysis rate of permethrin, a (2, 2-dichlorovinyl)-2,2-dimethyl cyclopropanecarboxylate. Among the three pI forms of permethrin CbE, the pI 4.8 CbEs showed the highest affinity and catalytic efficiency (Vmax/Km) toward all three naphthyl substrates tested and hydrolyzed permethrin at the greatest rate, indicating a major role in the esterase-mediated aspect of permethrin resistance in CPB. The availability of highly purified and biochemically and molecularly characterized forms of permethrin CbE will allow the efficient development of immunochemical (and following microsequencing techniques DNA-based) diagnostics for the monitoring of permethrin resistance in field populations of beetles. Also, these purified proteins will also allow us to determine the relative importance of sequestration versus hydrolytic detoxification in this aspect of permethrin resistance.

REFERENCES 1. A. J. Forgash, Insecticide resistance of the Colorado potato beetle, Leptinotarsa decemlineata (Say), in “Advances in Potato Pest Management” (J. H. Lashomb and H. Casagrande, Eds.), pp. 34–46, Hutchinson Ross, Stroudsburg, PA, 1981. 2. J. Lashomb, Potato pest management: a decade of progress, in “Advances in Potato Pest Biology and Management” (G. W. Zehnder, M. L. Powelson, R. K. Jansson, and K. V. Raman, Eds.), pp. 3–5, APS Press, St. Paul, MN, 1994. 3. J. A. Argentina, J. M. Clark, and D. N. Ferrero, Genetics of azinphosmethyl and permethrin resistance in the Colorado potato beetle (Coleoptera: Chrysomelidae), J. Econ. Entomol. 82, 698 (1989). 4. J. A. Argentine, J. M. Clark, and D. N. Ferro, Relative fitness of insecticide resistance Colorado potato beetle strains (Coleoptera: Chrysomelidae), J. Environ. Entomol. 18, 705 (1989). 5. J. A. Argentine, “Biochemistry and Genetics of Insecticide Resistance in the Colorado Potato Beetle,” Ph.D. dissertation, Dept. of Entomol., Univ. of Massachusetts, 1991. 6. J. A. Argentine, J. M. Clark and D. N. Ferro, Computer simulation for control of Colorado potato beetle (Coleoptera: Chrysomelidae), J. Agric. Entomol. 11, 137 (1994). 7. J. M. Clark and J. A. Argentine, Biochemical mechanisms of insecticide resistance in the Colorado potato beetle, in “Advances in Potato Pest Biology and Management” (G.W. Zehnder, M. L. Powelson, R. K. Jansson, and K. V. Raman, Eds.), pp. 294–308, APS Press, St. Paul, MN, 1994. 8. J. A. Argentine, S. H. Lee, M. A. Sos, S. R. Barry, and J. M. Clark, Permethrin resistance in a near isogenic strain of Colorado potato beetle, Pestic. Biochem. Physiol. 53, 97 (1995). 9. S. Lee and J. M. Clark, Tissue distribution and biochemical characterization of Carboxylesterases associated with permethrin resistance in a near isogenic strain of Colorado potato beetle, Pestic. Biochem. Physiol. 56, 208 (1996). 10. R. C. Switzer, C. R. Merril, and S. Shifrin, A highly sensitive silver stain for detecting proteins and peptides in polyacrylamide gels, Anal. Biochem. 98, 231 (1979). 11. Pharmacia LKB Biotechnology, “Affinity Chromatography: Principles and Methods,” Uppsala, Sweden, 1988. 12. T. W. Rademacher, R. B. Parekh, and R. A. Dwek, Glycobiology, Annu. Rev. Biochem. 57, 785 (1988). 13. K. F. Tipton and H. B. F. Dixon, Effects of pH on enzymes, in “Methods in Enzymology: Enzyme Kinetics and Mechanism” (D. L. Purich, Ed.), Part A, Vol. 63, pp. 183–234, Academic Press, New York, NY, 1979. 14. W.-L. Chen and C.-M. Sun, Purification and characterization of carboxylesterases of a rice brown planthopper,

PERMETHRIN CARBOXYLESTERASES IN COLORADO POTATO BEETLE

15.

16.

17.

18.

19.

20.

21.

22.

23.

Nilaparvata lugens Stal, Insect Biochem. Mol. Biol. 24, 347 (1994). T. Tsujita, T. Miyada, and H. Okuda, Purification of rat kidney carboxylesterase and its comparison with other tissue esterases, J. Biochem. 103, 327 (1988). J. Otto, A. Ronai, and O. Von Deimling, Purification and characterization of esterase 1F, the albumin esterase of the house mouse (Mus musculus), Eur. J. Biochem. 116, 285 (1981). A. L. Devonshire and G. D. Moores, A carboxylesterase with broad substrate specificity causes organophosphorous, carbamate and pyrethroid resistance in peachpotato aphids (Myzus persicae), Pestic. Biochem. Physiol. 18, 235 (1982). A. J. Ketterman, K. G. I. Jayawardena, and J. Hemingway, Purification and charaterization of a carboxylesterase involved in insecticide resistance from the mosquito Culex qinquefasciatus, Biochem. J. 287, 355 (1992). D. Fournier, J.-M. Bride, C. Mouches, M. Raymond, M. Magnin, J.-B. Berge, N. Pasteur, and G. P. Georghiou, Biochemical characterization of the esterases A1 and B1 associated with organophosphate resistance in the Culex pipiens L. complex, Pestic. Biochem. Physiol. 27, 211 (1987). A. Callaghan, C. A. Malcolm, and J. Hemingway, Biochemical studies of A and B carboxylesterases from organophosphate resistant strains of an Italian Culex pipiens (Diptera: Culicidae), Pestic. Biochem. Physiol. 41, 198 (1991). S. H. P. P. Karunaratne, K. G. I. Jayawardena, J. Hemingway, and A. J. Ketterman, Charaterization of a B-type esterase involved in insecticide resistance from mosquito Culex quinquefasciatus, Biochem. J. 294, 575 (1993). K. Sakata and T. Miyata, Biochemical charaterization of carboxylesterase in the small brown planthopper Laodelphax striatellus (Fallen), Pestic. Biochem. Physiol. 50, 247 (1994). K. H. Hong, W. H. Jang, K. D. Choi, and O. J. Yoo, Characterization of Pseudomonas fluorescens carboxylesterase: cloning and expression of the esterase gene in Escherichia coli, Agric. Biol. Chem. 55, 2839 (1991).

47

24. A. L. Devonshire, L. M. Searle, and G. D. Moores, Quantitative and qualitative variation in the mRNA for carboxylesterases in insecticide-susceptible and resistant Myzus persicae (Sulz), Insect Biochem. 16, 659 (1986). 25. M. Raymond, A. Callaghan, P. Fort, and N. Pasteur, Worldwide migration of amplified insecticide resistant genes in mosquitoes, Nature (London) 350, 151 (1991). 26. F. Wold, In vivo chemical modification of protein (posttranslational modification), Annu. Rev. Biochem. 50, 787 (1981). 27. W. B. Huttner, Protein tyrosine sulfation, TIBS 12, 361 (1987). 28. S. Whyard and V. K. Walker, Characterization of malathion carboxylesterase in the sheep blowfly Lucilia cuprina, Pestic. Biochem. Physiol. 50, 198 (1994). 29. H. A. van Lith, M. Haller, I, J. M. van Hoof, M. J. A. van Der Wouw, B. F. M. van Zutphen, and A. C. Beynen, Characterization of rat plasma esterase ES-1A concerning its molecular and catalytic properties, Arch. Biochem. Biophys. 301, 265 (1993). 30. A. L. Devonshire and G. D. Moores, Detoxification of insecticides by esterases from Myzus persicae—Is hydrolysis important? in “Enzymes Hydrolysing Organophosphorus Compounds” (E. Reiner, W. N. Aldridge, and F. C. G. Hoskin, Eds.), pp. 180–192, Wiley, Chichester, England, 1989. 31. N. C. de Malkenson, E. J. Wood, and E. N. Zebra, Isolation and chracterization of an esrerase of Triatoma infestans with a critical role in the degradation of organophosphorus esters, Insect Biochem. 14, 481 (1984). 32. L. Brady, A. M. Brzozowski, Z. S. Derewenda, E. Dodson, G. Dodson, S. Tolly, J. P. Turkenburg, L. Christiansen, B. Huge-jensen, L. Norskov, L. Thim, and U. Menge, A serine protease triad forms the catalytic centre of a triacylglycerol lipase, Nature (London) 343, 767 (1990). 33. F. K. Winkler, A. D’Arcy, and W. Hunziker, Structure of human pancreatic lipase, Nature (London) 343, 771 (1990). 34. D. Blow, Lipases reach the surface, Nature (London) 351, 444 (1991).