Development of a recombinant baculovirus expressing a modified juvenile hormone esterase with potential for insect control.код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 30:177-194 (1995) Development of a Recombinant Baculovirus Expressing a Modified Juvenile Hormone Esterase With Potential for Insect Control B.C. Bonning, K. Hoover, T.F. Booth, S. Duffey, and B.D. Hammock Departments of Entomology and Environmental Toxicology, University of California, Davis, California (B.C.B., K.H., S.D., B.D.H.); NEXCInstituteof Virology and Environmental Microbiology, Oafo~d,UK (TEB.) Baculovirus insecticides are receiving renewed attention as insect pest control agents following the development of fast-acting recombinant baculoviruses. Here we report on the construction and biological activity of a recombinant baculovirus derived from the nuclear polyhedrosis virus of Autograph californica which expresses a modified form of juvenile hormone esterase (JHE).The serine at the catalytic site of the JHE has been mutated to a glycine residue so that the protein does not degrade JH. The recombinant baculovirus expressing this modified form of JHE, named AcJHE-SG, has enhanced activity against lepidopteran larvae. Lethal times of the recombinant are 20 to 30% lower than for the wild type virus, and a 66% reduction in feeding damage caused by infected larvae is observed. This result is comparable to the best recombinant baculovirus developed to date, AcAalT, which expresses an insect-selective scorpion toxin. The potential of these recombinant viruses for commercialization as insecticides is discussed. Bioassays of AcJHE-SG in conjunction with anti-JH agents indicate that the virus is not killing by an anti-JH mechanism. Larvae apparently die from contraction-paralysis, or disruption of the normal sequence of events at the molt. o 1995 Wiley-Liss, Inc. Key words: juvenile hormone esterase, Autographa californica nuclear polyhedrosis virus, Heliothis virescens INTRODUCTION Baculoviruses are arthropod specific viruses which have long been in use as insect pest control agents for protection of numerous crop plants (Entwistle Acknowledgments: We thank B.F. McCutchen for provision of the virus AcAalT and A. Vaughen for assistance with bioassays. This work was supported by DCB-91-19332 from NSF, 91 -373026186 and 92-37302-7601 from USDA, and 23-696 from the USDA Forest Service. B.C.B. was supported by a NATO collaborative research grant (CRG 900955) and B.D.H. by the Burroughs Wellcome Foundation to work with R.D. Possee and T.F. Booth, NERC Institute of Virology and Environmental Microbiology, Oxford, U.K. Received October 7, 1994; accepted March 8, 1995. Address reprint requests to Dr. B.D. Hammock, Departments of Entomology and Environmental Toxicology, University of California, Davis, CA 9561 6. B.C. Bonning i s now at Department of Entomology, Iowa State University, Ames, IA 5001 1 . 0 1995 Wiley-Liss, Inc. 178 Bonning et al. and Evans, 1985). According to environmental conditions, death can occur anywhere from several days to weeks post infection. A reduction in the lethal time of the virus would enable broader use of the virus for protection of crops less able to sustain foliar damage without economic loss. The main aim of this work is to develop such a fast-acting baculovirus for commercial use as a crop protection agent. Recent advances in recombinant DNA technology have facilitated genetic engineering of baculoviruses to reduce the time taken for the virus to kill its larval host (Bonning and Hammock, 1992).Several recombinant baculoviruses derived from Autographa californica nuclear polyhedrosis virus (AcNPV) with reduced lethal times have been constructed (Bonning and Hammock, 1994; Maeda, 1989; McCutchen et al., 1991; OReilly and Miller, 1989; Stewart et al., 1991; Tomalski and Miller, 1992). The virus AcAaIT", which expresses an insect-selective scorpion toxin (McCutchen et al., 1991; Stewart et al., 1991), is widely considered to be the best. AcAaIT kills consistently between 25 and 30% more quickly than the wild type parent virus, and does not suffer from the stability problems seen for some of the other recombinants. Juvenile hormone (JH) and juvenile hormone esterase (JHE) are key components in the regulation of development in lepidopteran larvae (Hammock, 1985; Riddiford, 1994).As such they have been exploited for use in pest control with development of the juvenoids (Staal, 1982)and anti-JH agents (Staal, 1986) and insertion of JHE into recombinant baculoviruses (Bonning and Hammock, 1994; Bonning et al., 1992; Hammock et al., 1990). A series of mutant forms of JHE were made by site-directed mutagenesis for analysis of key residues involved in catalysis of JH (Ward et al., 1992). JHE is a serine esterase with a Ser201-Hi~446-G1~332 catalytic triad (Hanzlik et al., 1989; Ward et al., 1992). The recombinant viruses expressing the modified forms of JHE were tested for insecticidal efficacyby bioassay in Heliothis vivescens and Trickoplusia ni. The virus AcJHE-SG, which expresses the protein with the catalytic site serine (Serzol)changed to Gly, was found to be highly insecticidal. The efficacy of the virus against lepidopteran larvae is comparable to that of the virus AcAaIT (McCutchen et al., 1991; Stewart et al., 1991). In order to gain insight into the mechanism of action of AcJHE-SG, a series of bioassays were carried out using viruses in conjunction with various chemical agents designed to interfere with regulation of JH. The anti-JH agent ETB acts as an anti-JH agent at low doses at the tissue level. ETB may reduce JH *Abbreviations used: AcNPV C6 = Autograph californica nuclear polyhedrosis virus strain C6; AcAalT = AcNPV expressing the scorpion toxin Androctonus australis Insect-selective Toxin; AcUW2(B).JHE,AcJHE = AcNPV expressing wild type JHE; AcJHE-DN = AcNPV expressing JHE with Asp173 changed to Asn; AcJHE-HK = AcNPV expressing JHE with His446 changed to Lys; AcJHE-RH = AcNPV expressing JHE with Argd7 changed to His; AcJHE-SG = AcNPV expressing JHE with Serzol changed to Gly; ANOVA = analysis of variance; BSA = bovine serum albumin; CL = confidence limits; ECL = enhanced chemi-luminescence; EPPAT = O-ethyl-S phenylphosphoramidothiolate; ETB = ethyl 4-(2-(tert-butyl-carbonyloxy)butoxy)benzoate); FCS = fetal calf serum; FMev = Fluoromevalonolactone, tetrahydro-4-fluoromethyl-4-hydroxy-2H-pyran-2-one; J H = juvenile hormone; JHE = juvenile hormone esterase; LRD = lethal ratio for dose; LRT = lethal ratio for time; OTFP = 3-octylthio-1, I , I -trifluoropropan-2-one; PAGE = polyacrylamide gel electrophoresis; PEG = polyethylene glycol; pibs = polyhedron inclusion bodies; rJHBP = recombinant juvenile hormone binding protein; SDS = sodium dodecyl sulphate; TLC = thin layer chromatography. Baculovirus Expressing Modified JHE 179 titers by acting on the feedback regulation of JH production by the corpora allata (Staal, 1986).At high doses ETB acts as a juvenoid (Kiguchi et al., 1984; Kramer and Staal, 1981; Sparks et al., 1979). FMev is an anti-JH agent which acts directly on the corpora allata. FMev acts as a competitive inhibitor of mevalonic acid in conversion of mevalonate to JH (Quistad et al., 1981). The juvenoids, epofenonane, and fenoxycarb were used in bioassays along with the JHE inhibitors EPPAT (Sparks et al., 1983) and OTFP (Abdel-Aal and Hammock, 1985). These compounds are very potent inhibitors of JHE but also inhibit other esterases to varying degrees. Here we report on the action of AcJHE-SG and demonstrate its potential for insect control. MATERIALS AND METHODS Chemical Reagents The juvenoids fenoxycarb (Dorn et al., 1976) and epofenonane [Ro-10-3108; 1(4'-ethylphenoxy)-6,7-epoxy-3-ethyl-7-methylnonane (Zurfluh et al., 196811, supplied by U. Schweiter, Hoffman-LaRoche(Neuilly, France), were used in bioassays. ETB (ZR 2646) was supplied by G. Quistad and G.B. Staal formerly of Zoecon Corporation (Palo Alto, California). FMev was supplied by D.A. Schooley, formerly of Zoecon Corporation. The JHE inhibitor EPPAT was supplied by T.R. Fukuto (Universityof California, Riverside) and J. Sanborn (California EPA, Sacramento, California). OTFP was synthesized as described (Hammock et al., 1984) and purified by vacuum distillation. Further analyses showed it to be a single spot on normal phase TLC and a single peak on capillary GLC. Viruses AcJHE-SG is a recombinant baculovirus which expresses juvenile hormone esterase with the catalytic serine (Ser201)mutated to Gly. Hence, this modified protein has no catalytic activity against JH (Ward et al., 1992). The viruses AcUW2(B).JHE (AcJHE; Bonning et al., 1992), AcJHE-HK, AcJHE-DN, and AcJHE-RH (Ward et al., 1992) were employed as recombinant control viruses. AcJHE-HK expresses JHE with His446changed to Lys; AcJHE-DN expresses JHE with changed to Asn, and AcJHE-RH expresses JHE with Argd7changed to His. AcJHE-HK, AcJHE-DN, and AcJHE-RH were constructed in addition to AcJHE-SG for analysis of the catalytic mechanism of JHE (Ward et al., 1992).None of the JHEs produced by these viruses are highly active in hydrolyzing JH. All recombinant viruses were derived from the wildtype AcNPV strain C6, and were constructed using the plasmid pAcUW2 (Weyer et al., 1990) such that expression of the modified JHE gene was driven by a duplicated viral p10 promoter. Expression of recombinant proteins from these viruses in vitro is similar (Ward et al., 1992). The recombinant virus AcAaIT, which expresses an insect-selective toxin derived from a scorpion (McCutchen et al., 1991), was used for comparative purposes in bioassays. Binding Assay JHE-SG was produced in insect cell culture in the absence of FCS, and purified using DEAE anion exchange chromatography as described previously (Ichinose et al., 1992a). Binding assays were carried out using equilibrium 180 Bonning et al. dialysis with racemic 3H JH I11 (2.5 x M final concentration) in PEGcoated glass vials as described elsewhere (Park et al., 1993) with the modification that OTFP was not used. Dialysis tubing was prepared as described previously (Klotz, 1990). Purified recombinant juvenile hormone binding protein (rJHBP;Touhara et al., 1993) and P-galactosidase were used as positive and negative controls, respectively. Samples of JHE-SG and P-galactosidase were added to dialysis tubing in excess (10 to 20 pg) compared to rJHBP (400 ng). Dialysis bags were incubated at 25°C for 16 h with rotation. The amount of JH bound to each sample was determined by subtracting the concentration of hormone in the external reservoir from that inside the dialysis bag. Immunohistochemistry and Western Blot Analysis The JHE-SG purified by DEAE anion exchange was injected into fourth stadium larvae of Manduca sexfa (Ichinose et al., 1992b).Control larvae were injected with BSA. Ten micrograms of sample was injected in each case. One hour post injection, the pericardial cells were dissected out (Ichinose et al., 1992b) and placed directly into 0.1% paraformaldehyde in 0.1M phosphate buffer, pH 7.2, for immunohistochemical analysis. Sections were prepared and immunogold labeled as described previously (Booth et al., 1992). For Western blot analysis, pericardial cells were placed directly into cracking buffer following dissection, boiled for 4 min, and run on a 10%SDS polyacrylamide gel. Proteins were transferred to a nitrocellulose membrane, and detected using the ECL detection kit according to the manufacturer’s instructions (Amersham International plc, Buckinghamshire, U.K.). Bioassay Larvae of Trichoplusia ni were maintained as described previously (Bonning et al., 1992). Larvae of Heliothis virescens were maintained at 26°C on a semisynthetic diet (Hunter et al., 1984; BioServ, Frenchtown, NJ). For determination of lethal times of the recombinant baculoviruses, neonate larvae were infected at 2,000 pibs/pl using the droplet feeding technique (Hughes et al., 1986). Mortality was scored every 4,6, or 8 h according to the mortality rate. The response of each larva to gentle stimulation at the head end using a pipette tip was noted. Larvae which did not respond were scored as dead. Pigmentation of the cuticle, the bloated or contracted nature of the thorax and abdomen, and partial paralysis were also noted as symptoms. Thirty larvae were infected per bioassay for each virus, and bioassays were replicated three times. LTs0values were determined using a probit analysis program which assumes normal distribution and independence of observations (Russell et al., 1977). Bioassays were also carried out to determine lethal times of recombinant baculoviruses in larvae of H . virescens in conjunction with topical application of juvenoids, anti-JH agents, or inhibitors of JHE. The doses used for juvenoids, anti-JH agents, and JHE inhibitors were 30 ng, 5 pg, and 3 pg, respectively (Bonning and Hammock, unpublished report). Chemical reagents were suspended in ethanol or acetone and 1 p1 was applied topically to the dorsal surface of each larva between 4 and 8 h Baculovirus Expressing Modified JHE 181 post infection. Control larvae were treated with the chemical agent or solvent alone, without virus infection. For determination of the lethal doses of recombinant viruses, neonate larvae of H. virescens were placed individually into the wells of 24 well plates (Falcon, Franklin Lake, NJ) with diet for 5 days at 26°C. Larvae that had reached the third stadium by this time were infected with virus on a diet plug as described previously (Bonning et al., 1992).Larvae were infected with one of five doses of virus (3,000, 1,000, 333, 111, 37 pibs) applied previously to the diet plug in a volume of 1 p1 of distilled water. Only larvae that had completely consumed the diet plug after 24 h were transferred to cups of diet and maintained at 26°C. Mortality was scored 10 days later. Thirty larvae were infected at each dose per experiment and bioassays were repeated three times. Lethal doses were determined using a probit analysis program (Russell et al., 1977). There is no significant difference in the lethal dose between early and late third stadium larvae of H. virescens (Hoover, unpublished results). Lethal ratios for time (LRT) and dose (LRD) were calculated at the 50% level from LTs0and LD50 data (Bonning and Hammock, 1993). For example, LRTs0= LTs0of test virus LT50 of wild type virus. In Vivo Expression of Modified JHE In order to assess the expression levels of modified JHEs in vivo, hemolymph samples from larvae infected with wild-type AcNPV, AcJHE-SG, AcJHE-HK, AcJHE, AcJHE-RH, and AcJHE-DN (n = 3 or 4 per virus) were run on SDS polyacrylamide gels and analyzed by Western blot as described above. Larvae of H . virescens were reared to the third stadium and infected as described above for lethal dose bioassays. Hemolymph samples were taken from a small incision made in the proleg 3 days post infection for SDS PAGE. Samples taken from larvae infected with wild-type virus, AcJHE, AcJHE-HK, or AcJHESG (8 to 10 larvae per virus treatment) were also assayed for catalytic activity using radiolabeled JH (Hammock and Roe, 1985). Feeding-Damage Bioassay Newly hatched neonate larvae of H . virescens were dosed at 2,000 pibs/pl by droplet feeding (Hughes et al., 1986). Thirty larvae were infected with each test virus. Larvae were transferred individually to inverted 90 mm Petri dishes containing dampened circles of filter paper and iceburg lettuce (variety Salinas). Lettuce was grown in the greenhouse in the absence of pesticides. Petri dishes were sealed using parafilm and maintained at 26°C. Pieces of iceburg lettuce (which were not limiting), were changed every 48 h and the area of each piece scanned before and after feeding using a C1-202 area meter (CID, Inc., Vancouver, WA). Data for feeding-damage were analyzed by one-way ANOVA followed by the Scheffe means separation F-test (Steel and Torrie, 1980) which controls for the experiment-wise type 1 error rate. Feeding damage assays were also carried out using cotton leaves (variety 182 Bonning et al. Acala SJ-2). However, because the area of cotton leaves eaten by uninfected or infected larvae is only about 20% that of lettuce, feeding assays on cotton were done using third stadium larvae for accurate assessment of differences in feeding damage between treatments. Larvae were reared to third stadium (as described above) and starved for 24 h prior to infection with 3,000 pibs per diet plug. After 16 h, larvae that had completely consumed the diet plug were transferred to prescanned leaves and the protocol continued as described above for lettuce. RESULTS Binding Assays Binding assays were carried out using equilibrium dialysis with radiolabeled JH I11 to determine whether the protein JHE-SG will bind JH. No binding of JH to JHE-SG or P-galactosidase was detected despite the presence of these proteins in excess. A significant amount of JH (18 pMol) was bound to the rJHBP under the conditions used. This is within the range expected for racemic JH I11 (Park et al., 1993). Immunohistochemistry and Western Blot Analysis Immunohistochemicalanalysis confirmed that when JHE-SG is injected into larvae of M . sexta, it is removed from the hemolymph by the pericardial cells. Immunogold labeling revealed concentration of JHE-SG in the lysosome-like granules within the pericardial cells as seen for wild-type JHE (Booth et al., 1992).Labeling was above the background levels seen for sections prepared from larvae injected with BSA. Western blot analysis of pericardial cells taken from larvae injected with JHE-SG confirmed this result. No JHE was detected in pericardial cells taken from larvae injected with BSA when analyzed by Western blot. Bioassays Bioassay data for the recombinant baculoviruses expressing JHEs that had been modified for analysis of the catalytic mechanism (Ward et al., 1992) are shown in Table 1. No significant differences were seen in time to death of neonate larvae of H. vivescens except for AcJHE-SG which expresses JHE with the Ser at the catalytic site mutated to Gly ( P < 0.05). A similar reduction in lethal time was seen on bioassay of AcJHE-SG in T. ni relative to the wildtype virus (Table 1). Larvae infected with the wild-type virus were pale and swollen in appearance and lysed rapidly following death of the larva. Larvae infected with the JHE viruses AcJHE, AcJHE-HK, AcJHE-DN, and AcJHE-RH were pale in appearance at death similar to the wild-type virus-infected larvae (Fig. la), but lysis was delayed by 24 to 48 h compared to wild type virus-infected larvae. Of the larvae infected with AcJHE-SG, 27% (n = 159) died at the molt with two major morphological symptoms. In one case, the larvae showed extensive cuticular blackening and died after head capsule slippage because the old cuticle failed to split (Fig. lb). Dissection revealed that the larva was trapped within the old cuticle. In the other case, the old cuticle was not com- Baculovirus Expressing Modified JHE 183 TABLE 1. Lethal Times of Recombinant Baculoviruses Expressing Modified JHEs in Trichoplusia ni and Heliothis virescens Virus AcNPV C6 AcJHE AcJHE-SG' ACJHE-HK~ AcJHE-RH' ACJHE-DN~ T. ni LTs0(95%CL) 107 (105-108) 107 (103-111) 90 (87-93)* - - H.virescens LRT5R LT5, (95%CLIb 1 .o 1.o 0.84 - 114 (111-117) 118 (114-122) 81 (77-85)* 113 (110-116) 103 (97-108) 94 (85-100) LRTm 1.0 1.o 0.71 0.97 1.13 1.03 "LRTj0= LTj, recombinant virus LT5, wild type virus (Bonning and Hammock, 1993). bData taken from different bioassays. 'Baculovirus expressing JHE with Serzolmutated to Gly. dBaculovirusexpressing JHE with mutated to Lys. 'Baculovirus expressing JHE with Arg,, mutated to His. 'Baculovirus expressing JHE with Asp,, mutated to Asn. *Significantlydifferent from AcNPV C6 ( P < 0.05). pletely shed (Fig. lc). The remaining larvae infected with AcJHE-SG (73%) were pale and swollen in appearance similar to larvae infected with AcJHE (Fig. la). Some of these larvae exhibited contractive paralysis with only partial or no response to stimulation. Lethal doses for the recombinant baculoviruses in H . virescens were not significantly different from the wild-type virus ( P > 0.05; Table 2). The high variance is typical for these bioassays (Bonning et al., 1995). In Vivo Expression of Modified JHEs Samples of hemolymph taken from larvae infected with the recombinant baculoviruses expressing the modified JHEs were analyzed by SDS PAGE and Western blotting, and assayed for JHE activity. Western blots confirmed that in vivo expression of the modified forms of JHE was similar to the in vivo expression of the wild-type JHE. Catalytic activity detected using radiolabeled JH I11 to assay hemolymph taken from larvae infected with wild-type AcNPV, AcJHE-SG, or AcJHE-HK was not above background levels (5 to 10 nMol JH hydrolyzed/min/ml hemolymph). Activity detected in hemolymph from larvae infected with AcJHE was between 100 and 150 nMol JH hydrolyzed /min/ml hemolymph. Feeding-Damage Bioassay Feeding-damage caused by uninfected neonate larvae (control) and larvae infected with wild-type AcNPV, AcJHE-SG, or AcAaIT to iceburg lettuce was compared (Table 3A). Damage caused by the wild-type virus was significantly less (by 50%) than that caused by uninfected larvae (F = 5.16; P < 0.05; Table 3B). Damage caused by larvae infected with AcJHE-SG was significantly less (by 66%) than that caused by wild type virus-infected larvae (F = 2.70; P < 0.05), and not significantly different from damage caused by larvae infected with AcAaIT (F = 0.029; P > 0.05; Table 3B). Damage caused by larvae in- 184 Bonning et al. Fig. 1 . Symptoms of larvae of H. virescens infected with recombinant baculoviruses expressing JHEs. a: Larva infected with AcJHE showing characteristic pale coloration. Larvae infected with the corresponding wild-type virus are also characteristically pale in appearance and lyse rapidly. b: Larva infected with AcJHE-SG showing cuticular blackening. This larva died following head capsule slippage when the old cuticle failed to split. c: Larva infected with AcJHE-SG which died after an unsuccessful molt with incomplete shedding of the old cuticle. TABLE 2. Lethal Doses of Recombinant Baculoviruses Expressing Modified JHEs in Heliothis virescens Virus AcNPV C6 AcJHE AcJHE-SG AcJHE-HK LD50 (95% CL) 106 (73-156) 53 (25-82) 80 (16-170) 47 (22-73) Baculovirus Expressing Modified JHE 185 TABLE 3A. Feeding Damage to Iceburg Lettuce, and to Cotton Leaves Caused by Larvae of Heliothis virescens When Infected With Recombinant Baculoviruses Compared to WildType Virus and Uninfected Controls Mean area consumed" (cm') Virus Lettuceb SD Cotton' SD 2.88 1.45 0.49 0.39 1.61 1.50 0.63 0.57 6.92 4.52 2.07 2.92 4.96 2.50 0.90 Control AcNPV C6 AcJHE-SG AcAaITd 0.88 an = 30 larvae per treatment. bFeeding damage caused by neonate larvae. 'Feeding damage caused by third instar larvae. dBaculovirus expressing an insect-selective scorpion toxin (McCutchen et al., 1991). TABLE 3B. Statistical Analysis of Feeding Damage Datat F value Comparison Lettuce Cotton AcNPV C6 vs. control AcAaIT vs. control AcJHE-SG vs. control AcNPV C6 vs. AcAaIT AcNPV C6 VS. AcJHE-SG AcTHE-SG vs. AcAaIT 5.16' 15.78' 14.54' 3.28, 2.70* 0.02 0.04 0.11 0.17 0.02 0.04 0.05 'Groups compared by one-way ANOVA followed by the Scheffe means separation F-test. *Significantlydifferent ( P < 0.05). fected with AcJHE was not significantly different to damage caused by the wild-type virus (data not shown). Similar results were seen for feeding bioassays carried out on cotton with third stadium larvae (Table 3A; Fig. 2). On cotton, damage caused by wild-type AcNPV-infected third stadium larvae was 35% less than that caused by uninfected larvae, and damage caused by larvae infected with AcJHE-SG was 54% less than that caused by AcNPV-infected larvae (Table 3A). However, the greater variability in feeding-damage caused by third stadium larvae (which is not seen for neonate larvae) was such that none of the comparisons between treatments on cotton showed significant differences when compared by one-way ANOVA and the Scheffe means separation F test (Table 38). This was also seen for feeding-damage bioassays with third stadium larvae on iceburg lettuce (data not shown). Tables 4 to 6 show data for bioassays of wild-type and recombinant viruses in combination with various chemical agents designed to disrupt JH regulation. Table 4 shows data for the wild-type AcNPV. This table shows that two structurally distinct juvenoids and two separate chemical classes of JHE inhibitors tend to decrease the LTs0of the wild-type virus on infection and treatment of neonate larvae, while anti-JH agents tend to increase the lethal time. This may be related to the effects of JH and juvenoids on the fat body which is the primary site of virus replication (Bonning and Hammock, unpublished 186 Bonning et a!. Fig. 2. Feeding damage on cotton caused by an uninfected third stadium larva of H. virescens (control) or larvae infected with wild-type AcNPV (WT AcMNPV), or AcJHE-SG (S201G AcMNPV). An undamaged leaf (untreated) i s shown for comparison. Damage caused by w i l d type virusinfected larvae is 35% less than that caused by uninfected larvae. Damage caused by AcJHE%-infected larvae i s 54% less than that caused by wild-type virus-infected larvae (Table 3). report). Table 5 shows bioassay data for AcJHE. The juvenoids and JHE inhibitors increase the LT50where significant differences are seen between treated and untreated larvae infected with AcJHE ( P < 0.05). Greater variation was seen in data for the anti-JH agents, with only one out of six data sets being significantly different. In this case the lethal time of AcJHE was increased by FMev. Table 6 shows data for AcJHE-SG. In contrast to the wild-type virus, the juvenoids and JHE inhibitors increase the lethal time of AcJHE-SG as they do with AcJHE. The anti-JH agents FMev and ETB increase the lethal time of AcJHE-SG fairly consistently, as they do for the wild-type virus. These data are not consistent with the hypothesis that AcJHE-SG acts by reducing the JH titer. No mortality was observed for uninfected larvae treated with the chemical agents or solvents alone. Baculovirus Expressing Modified JHE 187 TABLE 4. Effect of Various Chemical Agents on the Lethal Time of Wild-Type AcNPV on Infection of Neonate Larvae of H. zlirescelzs* LTSo (95% CL) Treatment JH analogs Epofenonane Fenoxycarb JHE inhibitors OTFP EPPAT Anti-JH agents FMev ETB Control Treated 162 (157-168) 124 (120-129) 108 (103-111) 163 (156-1 70) 162 (157-168) 124 (120-129) 108 (103-111) 163 (156-170) 146 (139-155)” 108 (103-113)“ 97 (95-100)” 148 (137-162) 167 (158-178) 117 (112-122) 106 (101-109) 142 (133-151)” 162 (157-168) 131 (126-336) 131 (125-135) 100 (94-105) 131 (126-136) 164 (161-168) 121 (118-125)” 104 (98-109)” 101 (98-105) 115 (105-125)’ 133 (122-143) 146 (139-153) 115 (111-119) 137 (132-141) 145 (139-150) 144 (139-149) 144 (139-149) 131 (125-138) 166 (158-175)” 118 (112-123) 158 (150-166)” 149 (144-155) 161 (155-167)” 154 (150-159)” *No mortality was seen in the uninfected control larvae treated with solvent or the chemical agent alone. ”Significantlydifferent at 95% CL. DISCUSSION We have developed a fast-acting recombinant baculovirus which expresses a modified form of JHE which is inactive with respect to its function of JH catalysis. The most obvious hypothesis for the action of AcJHE-SG is that it reduces the effective JH titer at the tissue level synonymous with the function of the wild-type JHE. It follows from this hypothesis that both the injected recombinant protein (JHE-SG)and the recombinant baculovirus would act as biochemical anti-JH agents. Several lines of evidence refute this hypothesis. First the catalytic activity of JHE-SG is dramatically reduced in vitro (Ward et al., 1992) making it very unlikely that the ex ressed enzyme in infected insects is converting JH to JH acid. Assays with H-JH showed that the catalytic activity of THE-SG was reduced > lo6 fold over that of recombinant wild-type JHE (Ward et al., 1992). A second hypothesis is that expression of high levels of JHE-SG spare the natural JHE from degradation. In this case, the JH titer would be reduced by the natural THE. This hypothesis is not supported by the fact that the stage of insects treated is not anticipated to have high levels of endogenous JHE (Hammock, 1985). In addition, the hemolymph from insects infected with AcJHE-SG did not show significant catalytic activity on JH, in contrast to hemolymph from larvae infected with 9 188 Bonning et al. TABLE 5. Effect of Various Agents on the LTs0of AcJHE on Infection of Neonate Larvae of H . virescens LT50 (95% CL) Treatment JH analogs Epofenonane Fenoxycarb JHE inhibitors OTFP EPPAT Anti-JH agents FMev ETB LRT5t Control Treated 1.02 1.02 1.01 1.02 0.98 1.02 111 (108-114) 104 (98-108) 102 (93-108) 97 (90-106) 145 (137-155) 104 (98-108) 109 (106-113) 118 (112-125)b 126 (119-133)b 111 (108-114)b 147 (140-155) 125 (118-131)b 1.02 1.01 1.02 0.97 1.04 111 (108-114) 102 (93-108) 104 (98-108) 131 (127-135) 107 (104-110) 108 (104-113) 120 (114-125)b 119 (lll-126)b 153 (145-162)b 111 (104-118) 0.95 0.97 1.02 1.04 1.04 1.02 127 (121-133) 131 (127-135) 104 (98-108) 107 (104-110) 107 (104-110) 105 (101-109) 127 (120-133) 122 (117-127) 116 (109-123)b 110 (106-114) 115 (106-124) 107 (103-111) aLRT50= LTs0recombinant virus LTs0wild type virus. bSignificantlydifferent at 95% CL. AcJHE which expresses the catalytically active enzyme. The pharmacological data discussed below also argue against this hypothesis. A third hypothesis is that the THE-SG has lost its ability to catalyze JH but still binds the hormone tightly enough to sequester it. For this hypothesis to be correct, the protein must have a high affinity for JH and be at a high enough concentration in vivo to effectively prevent JH from binding to its receptor. We have a good model for comparison in the hemolymph JH binding protein (JHBP). This protein is at relatively high concentration in the hemolymph of both Manduca sexta (Goodman and Gilbert, 1978; Hidayat and Goodman, 1994) and T. ni (Wing et al., 1981) yet is compatible with normal insect metamorphosis. Thus, for JHE-SG to act by this last mechanism it would have to be at significantly higher concentration and/or bind JH with much higher affinity than the natural JHBP. As a control, the JHBP from M.sexta (Touhara et al., 1993) was expressed in the baculovirus system. The resulting virus failed to show any significant effects on the speed of kill or symptoms of infected larvae although production of this protein by the baculovirus system was quite high (50 mg 1-' in Sf21 cell culture; Bonning and Touhara, unpublished results). Under conditions where the recombinant JHBPbound JH 111, no binding to THE-SG was detected. We interpret these data to suggest first that the K, of the wild-type enzyme (Abdel-Aal et al., 1988) is largely the result of catalytic events leading to the formation of the acyl enzyme or that the Baculovirus Expressing Modified JHE 189 TABLE 6. Effect of Various Agents on the LTs0of AcJHE-SG on Infection of Neonate Larvae of H.virescens LTso (95% CL) Treatment JH analogs Epofenonane Fenoxycarb JHE inhibitors OTFP EPPAT Anti-JH agents FMev ETB LRTma 0.75 0.74 0.84 0.75 0.74 0.84 0.75 0.73 0.73 0.73 0.78 0.83 0.82 0.88 0.86 0.86 Control 93 122 102 93 122 102 Treated (88-98) (116-129) (98-106) (88-98) (116-129) (98-106) 105 147 104 92 140 109 (101-109)b (141-152)b (100-109) (87-97) (134-147)b (102-116) 93 (88-98) 94 (84-102) 94 (82-102) 94 (84-102) 102 105 101 100 (98-106)b (98-110) (94-106) (93-107) 130 143 105 143 100 91 (122-137)b (136-149)b (100-109) (137-148)b (95-106)b (87-95) 105 126 103 128 88 88 (98-111) (121-132) (98-107) (122-134) (84-92) (84-92) aLRT50= LTsorecombinant virus LTs0wild type virus. bSignificantlydifferent at 95% CL. protein’s structure is so altered by Gly,,, that it no longer binds JH. The JHESG protein does not appear to be at a higher concentration in the hemolymph of larvae infected with AcJHE-SG than wild-type JHE in larvae infected with AcJHE. Western blots of the hemolymph of insects infected with AcJHE, AcJHESG, and other mutants show similar levels of recombinant proteins. The bioassay data discussed below also argue against the JH binding hypothesis. The best argument against the catalytic mechanism of AcJHE-SG directly involving JH comes from the biological assays described here. Experiments based on the use of two structurally distinct juvenoids and two structurally distinct JHE inhibitors support the hypothesis that the decrease in lethal time of AcJHE-SG is not the result of an anti-JH mechanism. Anti-JH agents would be expected to decrease the lethal time of a recombinant with an anti-JH action, rather than increase the lethal time as seen for AcJHE-SG. Chemical antiJH agents would thus be expected to show additive and/or synergistic effects with a recombinant virus showing anti-JH effects which would lead to more rapid kill. This has been shown for another modified JHE which has been stabilized against degradation and expressed in a baculovims for insect control (Bonning et al., unpublished report). JHE is exported from the baculovirus-infected cell (Bonning et al., 1992; Hammock et al., 1990) into the hemolymph of the larva. From here it is removed by an active uptake process by the pericardial cells (Booth et al., 1992; Ichinose et al., 1992a,b). Using JHE-SG purified by DEAE anion exchange, 190 Bonning et al. we have shown using Western blotting and electron microscopy, that alteration of the catalytic site serine has not affected uptake of the protein by the pericardial cells. The following three hypotheses for the mechanism of AcJHE-SG are exclusive of JH. First, expression of JHE-SG by the recombinant virus could be saturating the pericardial cells resulting in disruption of homeostasis. This hypothesis is not supported by data for other recombinant baculoviruses which, when injected into larvae for very high levels of infection, express wild-type JHE to extremely high levels in the hemolymph without any toxic effect (for review see Bonning and Hammock, 1994). Secondly, the normal process of degradation of the enzyme within the pericardial cells could have been disrupted by alteration of the serine residue at the catalytic site. This hypothesis is currently under investigation. Finally, proteins which are structurally related to esterases have a number of roles unrelated to catalytic activity. These include both structural and transport functions (Cygler et al., 1993; Ollis et al., 1992). It is conceivable that by replacing the SerZolof JHE, we have inadvertently generated a protein which has such properties. Events at the larval molt appear to be particularly susceptible to the effects of JHE-SG. The exact mechanism whereby the old cuticle splits at the start of the molt is unknown. Failure of the cuticle of AcJHE-SG infected larvae to split may be related to generation of inadequate fluid pressure which facilitates molting. Failure to escape completely from the old cuticle may be related to the contraction paralysis seen in some specimens. Partial paralysis would hinder the muscular movements associated with escape from the exuvium. It cannot be assumed that a reduced lethal time of a recombinant baculovirus is proportional to the resultant feeding damage. Larvae in bioassays are scored as dead only when no response is observed in reaction to stimulation at the head end. Hence, if larvae are trapped at the molt, or partially paralyzed, the lethal time will underestimate the reduction in feeding damage. The variation in LRTs0for AcJHE-SG (from 0.73 to 0.88) can be seen in Table 6 and may be related to the exact age of neonate larvae at the time of infection. This is currently under investigation. Despite this variation, feeding damage on iceburg lettuce was not significantly different from that caused by AcAaIT which kills fairly consistently with an LRT50of 0.7 (Bonning and Hammock, 1993; McCutchen et al., 1991; Stewart et al., 1991).Larvae infected with AcJHESG remain on the plant but become sluggish and stop feeding, whereas larvae infected with AcAaIT fall off the plant as a result of rapid paralysis (Cory et al., 1994). The fact that cadavers of larvae infected with AcJHE-SG remain on the plant will facilitate dispersal of the virus following liquefaction of the cadaver, but may not be so appealing from the perspective of the farmer or consumer. We have demonstrated the potential of the recombinant baculovirus AcJHESG as an effective insecticidal agent. A significant reduction in feeding damage is seen on infection of larvae with AcJHE-SG which is comparable to that for AcAaIT, widely seen as the most effective recombinant baculovirus with reduced lethal time developed to date (McCutchen et al., 1991; Stewart et al, 1991). However, neither recombinant virus reduces feeding damage sufficiently to warrant commercialization. Additives or synergists used in con- Baculovirus Expressing Modified JHE 191 junction with these recombinant baculoviruses (McCutchen et al., 1994; Shapiro and Dougherty, 1993) can be expected to facilitate their efficacy as crop protection agents. An understanding of the mechanism by which JHESSG kills larvae, and the pharmacokinetics of the recombinant protein may facilitate development of recombinant viruses with a greater reduction in the lethal time on infection of pest insects. LITERATURE CITED Abdel-Aal YAI, Hammock BD (1985): 3-Octylthio-l,l,l-trifluoro-2-propanone, a high affinity and slow binding inhibitor of juvenile hormone esterase from Trichoplusia ni (Hiibner). Insect Biochem 15:111-122. Abdel-Aal YAI, Hanzlik TN, Hammock BD, Harshman LG, Prestwich G (1988): Juvenile hormone esterases in two Heliothines: Kinetic, biochemical and immunogenic characterization. Comp Biochem Physiol90B:117-124. Bonning BC, Hammock BD (1992): Development and potential of genetically engineered viral insecticides. Biotechnol Genet Eng Rev 10:453487. Bonning BC, Hammock BD (1993): Lethal ratios: An optimized strategy for presentation of bioassay data generated from genetically engineered baculovirus. J Invert Pathol62:196-197. Bonning BC, Hammock BD (1994): Insect control by use of recombinant baculoviruses expressing juvenile hormone esterase. In Hedin P, Menn JJ, Hollingworth R (eds): Natural and Engineered Pest Management Agents. Washington, DC: ACS Symposium Series, vol. 551, pp 368-383. Bonning BC, Hirst M, Possee RD, Hammock BD (1992): Further development of a recombinant baculovirus insecticide expressing the enzyme juvenile hormone esterase from Heliothis virescens. Insect Biochem Mol Biol22:453458. Bonning BC, Hoover K, Duffey S, Hammock BD (1995): Production of polyhedra of the Autographa californica nuclear polyhedrosis virus using the Sf21 and Tn5B1-4 cell lines and comparison with host-derived polyhedra by bioassay. J Invert Pathol (in press). Booth TF, Bonning BC, Hammock BD (1992): Localization of juvenile hormone esterase during development in normal and in recombinant baculovirus-infected larvae of the moth Trichoplusia ni. Tissue Cell 24:267-282. Cory JS, Hirst ML, Williams T, Hails RS, Goulson D, Green BM, Carty TM, Possee RD, Cayley PJ, Bishop DHL (1994): Field trial of a genetically improved baculovirus insecticide. Nature 370:138-140. Cygler M, Schrag JD, Sussman JL, Hare1 M, Silman I, Gentry MG, Doctor BP (1993): Relationship between sequence conservation and three-dimensional structure in a large family of esterases, lipases, and related proteins. Protein Sci 2:366-382. Dorn S, Oesterhelt G, Suchy M, Trautmann KH, Wipf HK (1976): Environmental degradation of the insect growth regulator 6,7-epoxy-l-(p-ethylphenoxy)-3-ethyl-7-methylnonane (Ro 103108) in polluted water. J Agric Food Chem 24:637-640. Entwistle PL, Evans HF (1985): Viral control. In Gilbert LI, Kerkut GA (eds): Comprehensive Insect Physiology, Biochemistry and Pharmacology. Oxford: Pergamon Press, vol. 12, pp 347-412. 192 Bonning et d. Goodman W, Gilbert LI (1978):The hemolymph titer of juvenile hormone binding protein and binding sites during the fourth larval instar of Manduca sexta. Gen Comp Endocrinol35:27-34. Hammock BD (1985): Regulation of juvenile hormone titer: Degradation. In Kerkut GA, Gilbert LI (eds): Comprehensive Insect Physiology, Biochemistry, and Pharmacology. New York: Pergamon Press, pp 431472. Hammock BD, Roe RM (1985): Analysis of juvenile hormone esterase activity. In Law JH, Rilling HC (eds): Methods in Enzymology. Orlando, FL: Academic Press, pp 487-494. Hammock BD, Abdel-Aal YAI, Mullin CA, Hanzlik TN, Roe RM (1984): Substituted thiotrifluoropropanones as potent selective inhibitors of juvenile hormone esterase. Pestic Biochem Physiol22209-223. Hammock BD, Bonning BC, Possee RD, Hanzlik TN, Maeda S (1990): Expression and effects of the juvenile hormone esterase in a baculovirus vector. Nature 344:458461. Hanzlik TN, Abdel-Aal YAI, Harshman LG, Hammock BD (1989): Isolation and sequencing of cDNA clones coding for juvenile hormone esterase from Heliotkis virescens: Evidence for a charge relay network of the serine esterases different from the serine proteases. J Biol Chem 264112419-12425. Hidayat P, Goodman WG (1994):Juvenile hormone and hemolymph juvenile hormone binding protein titers and their interaction in the hemolymph of fourth stadium Manduca sexta. Insect Biochem Mol Biol24:709-715. Hughes PR, van Beek NAM, Wood HA (1986): A modified droplet feeding method for rapid assay of Bacillus tkuringiensis and baculoviruses of noctuid larvae. J Invert Pathol48:187-192. Hunter FR, Crook NE, Entwistle PF (1984): Viruses as pathogens for the control of insects. In Grainger JM, Lynch JM (eds): Microbial Methods for Environmental Biotechnology. New York Academic Press, pp 323-347. Ichinose R, Kamita SG, Maeda S, Hammock BD (1992a): Pharmacokinetic studies of the recombinant juvenile hormone esterase in Manduca sexta. Pestic Biochem Physiol42:13-23. Ichinose R, Nakamura A, Yamoto T, Booth TF, Maeda S, Hammock BD (199213):Uptake of juvenile hormone esterase by pericardial cells of Manduca sexta. Insect Biochem Mol Biol22:893-904. Kiguchi K, Mori T, Akai H (1984): Effects of anti-juvenile hormone "ETB on the development and metamorphosis of the silkworm, Bombyx mori. J Insect Physiol30:499-506. Klotz IM (1990): Ligand-protein binding affinities. In Creighton TE (ed): Protein Function: A Practical Approach. Oxford: IRL Press, pp 25-54. Kramer SJ, Staal GB (1981): In vitro studies on the mechanism of action of anti-juvenile hormone agents in larvae of Manduca sexta. In Pratt GE, Brooks GT (eds): Juvenile hormone biochemistry. New York Elsevier /North-Holland, pp 425437. Maeda S (1989): Increased insecticidal effect by a recombinant baculovirus carrying a synthetic diuretic hormone gene. Biochem Biophys Res Commun 165:1177-1183. McCutchen BF, Choudary PV, Crenshaw R, Maddox D, Kamita SG, Palekar N, Volrath S, Fowler E, Hammock BD, Maeda S (1991): Development of a recombinant baculovirus expressing an insect-selective neurotoxin: Potential for pest control. Bio/Technology. 93348-852. McCutchen BF, Betana MD, Herrmann R, Hammock BD (1995): Interactions of recombinant Baculovirus Expressing Modified JHE 193 and wild-type baculoviruses with classical insecticides and pyrethroid-resistant tobacco budworm (Lepidoptera: Noctuidae). J Econ Entomol (in press). Ollis DL, Cheah E, Cygler M, Dijkstra B, Frolow F, Franken SM, Hare1 M, Remington SJ, Silman I, Schrag J, Sussman JL, Verschueren KHG, Goldman A (1992): The alp hydrolase fold. Protein Eng 5:197-211. OReilly DR, Miller LK (1989): A baculovirus blocks insect molting by producing ecdysteroid UDP-glucosyl transferase. Science 245:lllO-1112. Park YC, Tesch MJ, Toong YC, Goodman WG (1993): Affinity purification and binding analysis of the hemolymph juvenile hormone binding protein from Manduca sexta. Biochemistry 327909-7915. Quistad GB, Cerf DC, Schooley DA, Staal GB (1981): Fluoromevalonate acts as an inhibitor of insect juvenile hormone biosynthesis. Nature 289:176-177. Riddiford LM (1994): Cellular and molecular actions of JH I. General considerations and premetamorphic actions. Adv Insect Physiol24:213-274. Russell RM, Robertson JL, Savin NE (1977): POLO: A new computer program for probit analysis. Bull Entomol SOCAm 23:209-213. Shapiro M, Dougherty EM (1993): The use of fluorescent brighteners as activity enhancers for insect pathogenic viruses. In Lumsden RD, Vaughn JL (eds): Pest Management; Biologically Based Technologies. Washington DC: ACS Series, pp 40-46. Sparks TC, Wing KD, Hammock BD (1979): Effects of the anti-juvenile hormone mimic ETB on the induction of insect juvenile hormone esterase in Trichoplusia ni. Life Sci 25:445-450. Sparks TC, Hammock BD, Riddiford LM (1983): The haemolymph juvenile hormone esterase of Manduca sexta (L.): Inhibition and regulation. Insect Biochem 13:529-541. Staal GB (1982): Insect control with growth regulators interfering with the endocrine system. Entomol Exp Appl31:15-23. Staal GB (1986): Anti juvenile hormone agents. Ann Rev Entomol31:391-429. Steel RGD, Torrie JH (1980):Principles and Procedures of Statistics: A Biometrical Approach. New York: McGraw-Hill. Stewart LMD, Hirst M, Ferber ML, Merryweather AT, Cayley PJ, Possee RD (1991):Construction of an improved baculovirus insecticide containing an insect-specific toxin gene. Nature 35285-88. Tomalski MD, Miller LK (1992): Expression of a paralytic neurotoxin gene to improve insect baculoviruses as biopesticides. Bio/Technology 10:545-549. Touhara K, Lerro KA, Bonning BC, Hammock BD, Prestwich GD (1993): Ligand binding by a recombinant insect juvenile hormone binding protein. Biochemistry 322068-2075. Ward VK, Bonning BC, Huang TL, Shiotsuki T, Griffeth VN, Hammock BD (1992): Analysis of the catalytic mechanism of juvenile hormone esterase by site-directed mutagenesis. Int J Biochem 24:1933-1941. Weyer U, Knight S, Possee RD (1990): Analysis of very late gene expression by Autograph 194 Bonning et al. culifornicu nuclear polyhedrosis virus and the further development of multiple expression vectors. J Gen Virol71:1525-1534. Wing KD, Sparks TC, Love11 VM, Levinson SO, Hammock BD (1981): The distribution of juvenile hormone esterase and its interrelationship with other proteins influencing juvenile hormone metabolism in the cabbage looper, Tvichoplusia ni. Insect Biochem 11:473485. Zurfluh R, Wall EN, Siddall JB, Edwards JA (1968): Synthetic studies on insect hormones. VII. An approach to stereospecific synthesis of juvenile hormones. J Am Chem SOC90:6224-6225.