Characterization of Affinity-purified Juvenile Hormone Esterase from the Plasma of the Tobacco Hornworm, Manduca sex&*

Juvenile hormone (JH) esterase found primarily in the hemolymph and tissues of insects is a low abundance protein involved in the ester hydrolysis of insect juvenile hormones, JHs. The enzyme was purified from the larval plasma of wild-type Manduca sexta using an affinity column prepared by binding the ligand, 3-[(4'-mercapto)butylthio]-1,1,1-trifluoropropan-2-one (MBTFP), to epoxy-activated Sepharose. The purification was greater than 700-fold with a 72% recovery, and the purified enzyme appeared as a single protein on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoelectrophoresis, reverse phase high performance liquid chromatography, and amino acid sequence analysis. The molecular weight was 66,000. The plasma JH esterase in wild-type, black, and white strains of M. sexta was similar when analyzed by immunotitration, wide range (pH 3.5-9.0) isoelectric focusing, and inhibition with MBTFP and 3-octylthio-1,1,1-trifluoropropan-2-one (OTFP). Inhibition studies revealed a sensitive and insensitive form (I50 = 10(-9) and 10(-6) M, respectively) in these three biotypes. Narrow range isoelectric focusing (pH 4.0-7.0) indicated the presence of two major isoelectric forms with pI values of 6.0 and 5.5, but their inhibition kinetics with OTFP and O,O-diisopropyl phosphorofluoridate were identical.

Juvenile hormone (JH) esterase found primarily in the hemolymph and tissues of insects is a low abundance protein involved in the ester hydrolysis of insect juvenile hormones, JHs. The enzyme was purified from the larval plasma of wild-type Munducu sexta using an affinity column prepared by binding the ligand, 3-[(4'-mercapto)butylthio]-l,l,l-trifluoropropan-2-one (MBTFP), to epoxy-activated Sepharose.
The purification was greater than 700-fold with a 72% recovery, and the purified enzyme appeared as a single protein on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoelectrophoresis, reverse phase high performance liquid chromatography, and amino acid sequence analysis. The molecular weight was 66,000. The plasma JH esterase in wild-type, black, and white strains of M. sexta was similar when analyzed by immunotitration, wide range (pH 3.5-9.0) isoelectric focusing, and inhibition with MBTFP and S-octylthiol,l,l-trifluoropropan-2-one (OTFP). Inhibition studies revealed a sensitive and insensitive form (IBo = lOWe and lo-' M, respectively) in these three biotypes. Narrow range isoelectric focusing (pH 4.0-7.0) indicated the presence of two major isoelectric forms with pI values of 6.0 and 5.5, but their inhibition kinetics with OTFP and O,O-diisopropyl phosphorofluoridate were identical.
Juvenile hormones (JHs)' are sesquiterpenes and are known to regulate development and reproduction in insects (reviewed by de Kort and Granger, 1981). A number of studies have shown that catabolism of JH by specific esterases, in addition to a decrease in JH biosynthesis, is important in regulating JH titer at critical times of insect development, thus allowing for normal metamorphosis from a larva to the adult moth or butterfly (reviewed by Roe and Venkatesh, 1990). In the hemolymph of most insects studied, ester hydrolysis of JH to the metabolite JH acid, by JH esterase(s) is the primary pathway of metabolism  3-1(4'mercapto)but$lthio]-l,l,l-trifluoropropan-2-one; OTFP, 3-o&yl$iol,l,l-trifluoropropan-2-one; PAGE, polyacrylamide gel electrophoresis; SDS, sodium dodecyl sulfate; TBS, Tris-buffered saline; TFA, trifluoroacetic acid.
of JH esterase activity. The first initiates wandering behavior, where the larva moves to the soil to undergo metamorphosis. The second peak regulates the transformation from the larva to the adult moth. The purification and characterization of this important regulatory enzyme is an essential step to furthering our knowledge of the regulation of insect metamorphosis as well as developing molecular approaches to regulate the enzyme activity as a means of selective insect control.
A major focus in JH esterase research has been the cabbage looper, Trichoplusia ni, and the tobacco hornworm, A4. sexta. JH esterase was purified from the hemolymph of T. ni (Yuhas et al., 1983;Rudnicka and Jones, 1987; and M. senta (Coudron et al., 1981) using classical approaches. From structure-activity and kinetic studies with polarized trifluoromethyl ketones, however, the affinity column, Sepharose-S(CH&SCH,COCF,, was discovered that made possible the rapid purification of JH esterase from hemolymph and other tissues (Abdel-Aal and Hammock, , 1986Abdel-Aal et al., 1988;Hanzlik and Hammock, 1987;Roe and Venkatesh, 1990). From these and other studies on M. sexta JH esterase, conflicting reports were made on the number of hemolymph JH esterases (Sanburg et al., 1975;Coudron et al., 1981;Sparks et al., 1983;Wing et al., 1984;Hammock, 1985, 1986;Sparks et al., 1989) and multiple catalytic sites (Abdel-Aal and . There are also no published data available on the immunogenicity, amino acid composition, and sequence information on the JH esterase for this important insect model.

EXPERIMENTAL PROCEDURES
Animals-Wild-type M. sexta larvae from the original Yamamoto colony (Yamamoto, 1968) were reared according to Bell and Joachim (1976) at 27 + 1 "C under a 16-h:8-h 1ieht:dark cvcle. Fourth instar larvae exhibiting head capsule slippage were separated from the main colony and those undergoing ecdysis between 2 and 6 h after lightson were designated as day 0, last instar larvae. Gates were selected as described previously  Preparation and Enzyme Assays-Hemolympb from fifth instar day three (gate I) larvae was collected via a clipped horn and anal prolegs into a culture tube (10 X 75 mm) held at 4 "C, containing a few crystals of phenylthiourea. The hemolymph was centrifuged for 5 min at 1000 x g, and the resulting plasma was filtered through glass wool and stored at -85 "C until used. Freeze-thawing had no detectable effect on the esterase activity. JH esterase activity was assayed by the partition method described by Hammock and Roe (1985). Isoelectric focusing was performed on a LKB 2117 Multiphor II electrophoresis system using LKB Ampholine PagPlates. The broad range precast, 11.0.cm gel (pH 3.5-9.5) was 5% polyacrylamide with 2.4% Ampholine.
The narrow range precast, 11.0~cm gel (pH 4.0-7.0) was 5% polyacrylamide and 2.2% Ampholine. Gels were prefocused for 30 min at 10 "C and 25 watts without samples. Ten microliters of each diluted sample was applied as droplets at six locations in separate lanes excluding 1 cm from the cathode and anode ends. Electrofocusing was performed for 2.5 h at a constant power of 25 watts. The gel was then sliced into twenty 0.5~cm pieces and eluted for 24 h at 4 "C in 0.5 ml of standard sodium phosphate buffer for measurement of enzyme activity or in 0.5 ml of distilled water for measurement of pH. Immunology-Polyclonal antibodies were elicited in BALB mice. A primary dose consisting of 50 pg of purified JH esterase emulsified in Freund's complete adjuvant was injected intraperitoneally. Two weeks later, a booster dose of 50 rg of purified protein was administered in Freund's incomplete adjuvant.
The mice were killed 2 weeks later, and the blood was collected from a heart puncture. Serum was prepared by centrifuging the clotted blood at 10,000 x g for 15 min at 4 "C. The immune serum thus collected was stored at -85 "C and used for enzyme characterization studies. Immunoblotting was performed by transferring proteins from an SDS-PAGE gel onto nitrocellulose according to Towbin et al. (1979 Hewick et al. (1981).

AND DISCUSSION
Purification of JH E&erase-The purification of JH esterase from the plasma of last instar day three (gate I) larvae of M. sexta was accomplished using the affinity matrix MBTFP-Sepharose. A number of other affinity ligands and linkages to Sepharose were also examined, but MBTFP-Sepharose produced the optimum column. When the plasma was passed through this affinity column, there was a 90% decrease in the JH esterase activity without any detectable loss of protein in the column effluent. This indicates the highly selective nature Juvenile Hormone Esterase of M. sexta Lane A, molecular weight (MOL. WT.) markers (Bio-Rad): phosphorylase b (92,500), bovine serum albumin (66,000), ovalbumin (45,000) carbonic anhydrase (31,000), soybean trypsin inhibitor (21,500), and lysozyme (14,400); lone B, plasma (40 pg of protein); lane C, purified JH esterase (2.5 rg); lone D, purified JH esterase (1.25 pg); lanes E and F, Western blots of purified JH esterase using 1 and 0.5 pg of protein, respectively. Plasma and purified JH esterase was from the hemolymph of last instar day 3 (gate I) wildtype M. sexta larvae.
of the ligand used for preparing the affinity matrix. From 100 ml of plasma containing 2100 mg of protein (specific activity of 0.71 nmol of JH III min-' mg-' of protein), 1.95 mg of purified JH esterase was recovered having a specific activity of 554 nmol of JH III min-' mg-' of protein (Table I). This represents a purification factor of 780 with an overall recovery of 72% of the original enzyme activity applied to the column. Treatment of JH esterase bound to this affinity matrix with salt, pH extremes, detergents, and urea was unsuccessful in causing enzyme release, but exposure to OTFP concentrations greater than 10m5 M eluted the esterase in high purity. However, elution of the enzyme bound to the affinity matrix with OTFP took an extended period of time (2 days). Trifluoromethyl ketones are slow, tight-binding inhibitors and sometimes behave as pseudoirreversible inhibitors (Abdel-Aal and . This delay was also apparent in the long dialysis time (6 days) required to regenerate the enzyme activity. The JH esterase is a low abundance protein and cannot be visualized as a discrete band when the plasma is subjected to SDS-PAGE (Fig. 1, lane B). As shown in lanes C and D, SDS-PAGE analysis of affinity-purified JH esterase appears as a single band with an apparent molecular weight of 66,000. This result corresponds well with the earlier estimated molecular weight for M. sextu JH esterase of 62,000 by Coudron et al. (1981) and 65,000 by Abdel-Aal and Hammock ( , 1986. Abdel-Aal and  also reported that when the purified M. sexta JH esterase was exposed to [3H]paraoxon (an irreversible inhibitor of serine esterases) and examined by fluorography, only a single band was observed, corresponding to that detected by Coomassie on SDS-PAGE gels. Immunoblotting of purified JH esterase using mouse polyclonal anti-JH esterase serum showed one clear band, and no other cross-reacting proteins were evident, in- ase. Affinity-purified JH esterase from last instar day 3 (gate I) wildtype M. sexta larvae was injected into an Aquapore octyl300-A reverse phase column in 0.1% TFA graded to 0.1% TFA in 80% acetonitrile, 15% isopropyl alcohol in water. The absorbance of the protein at 214 and 280 nm was determined. The peak at zero time is the solvent peak.
dicating the homogeneity of the purified enzyme (Fig. 1, lanes E and F).
Reverse phase HPLC analysis of purified JH esterase is shown in Fig. 2. Absorbance of the protein determined at both 214 and 280 nm yielded only a single peak, indicating the purity of the protein, and illustrates the advantage of the affinity procedure for selectively purifying this enzyme. The amino acid composition of the HPLC-purified JH esterase is shown in Table II. Based on the residues per molecule, the molecular weight was estimated to be 66,000, which is in agreement with that obtained by SDS-PAGE. There is a predominance of alanine. Glycine was predominant in the JH esterase of the cabbage looper, T. ni, while methionine and histidine were in low amounts . The HPLC-purified JH esterase of M. sextu was subjected to Edman degradation. This analysis resulted in a sequence of 15 residues from the NH2 terminus, indicating that the purified enzyme is likely a single protein (Fig. 3). This is supportive evidence for the appearance of the enzyme as a single band on SDS-PAGE whether visualized by Coomassie Blue or by antibody interaction on Western blots (Fig. 1) and a single peak on HPLC (Fig. 2). Although JH esterase has been purified in several lepidopterans, protein sequence information has only been recently available for Heliothis uirescens (Hanzlik et al., 1989). Fig. 3 compares the protein sequence of the first 15 amino acids of the NH2 terminus reported in this study for M. sexta to that of H. virescens. Edman degradation of affinity-purified plasma JH esterase of H. uirescens yielded a major and minor form in a 3:l ratio, but JH esterase was characterized as a single band on SDS-PAGE and isoelectric focusing. The minor form had a 2-residue extension of Ser-Ala followed by a sequence of 5 residues identical with the 5 NH*-terminal residues of the major form (Hanzlik et Hanzlik et al. (1989Hanzlik et al. ( ). al., 1989. The protein sequence of plasma JH esterase of H. virescens deduced from cDNA clones was found to be homologous to the NH2 terminus of the major form found in purified JH esterase with one exception at position 10. Interestingly, only 1 residue, Val-9 of M. se&a, matches with the NH2terminal amino acid sequence of the major form of H. uirestens determined by Edman degradation and two residues Val-9 and Val-10 of M. sextu matches with the cDNA deduced protein sequence (Fig. 3). The lack of homology is surprising since JH esterase from both species has a K, for JH in the range of 10e7 M. This indicates that either the NH2 terminus is not important in in vitro catalytic activity or a great deal of latitude may exist in this region relative to its function. Questions about the functional significance of the primary structure cannot be addressed fully until the active site has been identified and the three-dimensional structure of JH esterase is described. The differences noted between these two species may also be important in interactions of JH esterase with in viuo elements. We have noted that the protein function for other M. sentu proteins varies significantly between species even in closely related insect species. In vivo interactions will be evaluated in future studies by interspecies transformations of JH esterase.

Immunotitration
of JH Ester-use-The plasma JH esterase activity of wild, black, and white strains of last instar day three (gate I) larvae of M. sexta was compared with that of purified JH esterase from wild-type larvae of the same developmental age by immunotitration (Fig. 4). Plasma JH esterase and the purified, reactivated JH esterase containing equal amounts of enzyme activity were used against different concentrations of mouse polyclonal, anti-JH esterase serum. The results presented in Fig. 4 show that the immunotitration pattern for the three different strains was similar, indicating that the JH esterase activity is due to the same enzyme(s) in all these different strains. It appears that the quantity of immune serum required to immunoprecipitate 50% of the purified enzyme activity was less than that required for plasma (Fig. 4). This was further demonstrated by using an ELBA to test the cross-immunoreactivity of our immune serum using an equal amount of plasma protein and purified JH esterase (Fig. 5). The concentration of JH esterase in last instar day three plasma is roughly 0.1% of the total protein. Therefore, 0.5 pg of plasma protein (the highest concentration tested in Fig. 5) contains 0.5 ng of JH esterase. This amount of JH esterase is below the detection limit of the immune serum used in the present study (see Fig. 5). At a concentration of up to 0.5 pg of plasma protein versus an equal amount of purified JH esterase, the anti-JH esterase serum showed no cross-reactivity with plasma protein. As the protein concentration of the plasma increased beyond 0.5 Kg, the ELISA cross-reactivity became evident (data not shown). An examination of Western blots of plasma revealed the presence of a FIG. 4. Immunoprecipitation of JH esterase. Immunoprecipitation of purified JH esterase (JHE) from wild-type larvae and the plasma JH esterase from wild, black, and white strains of M. senta.
The JH esterase was from last instar day 3 (gate I) larvae. Different concentrations of immune serum in sodium phosphate buffer (I = 0.2 M, pH 7.2) were incubated at 4 "C overnight with JH esterase having equal enzyme activity. Following centrifugation at 10,000 X g for 2 h, the supernatants were assayed for JH esterase activity.
Values are the mean of three independent determinations with a standard deviation less than 5% of the mean. Preimmune sera served as the control and was used against both purified enzyme and plasma from the wild strain with the same results. The plasma control data are plotted. Values are the mean of three independent determinations with a standard deviation less than 5% of the mean.

Juvenile
Hormone E&erase of M. sexta 21731 number of cross-reacting proteins (data not shown). In a similar study, Hanzlik and Hammock (1987) reported that the antibodies developed against affinity-purified JH esterase from the plasma of 2'. ni also cross-reacted with other proteins in the plasma. These authors attributed this cross-reaction to similarities in glycosylation of JH esterase with glycosylation of other serum proteins. Interestingly,  reported that the polyclonal antibodies developed to JH esterase purified from the plasma of T. ni by classical techniques was highly specific to JH esterase and showed no crossreactivity to other plasma proteins. The reasons for the above discrepancy are not clear. It is possible that the affinity purified enzyme used in the present study and that used by Hanzlik and Hammock (1987) was contaminated with other plasma proteins in low abundance and high antigenicity. However, this appears unlikely in the present study, because Western blotting of purified JH esterase detected only a single band, and no other cross-reacting proteins were evident.  also found that in T. ni, the quantity of anti-JH esterase serum needed for immunoprecipitating 50% of the JH esterase activity in the fresh hemolymph was less than that required in the case of purified JH esterase; this was attributed to a possible loss of enzymatic activity during purification.
In our studies, an examination of the plasma before and after immunoprecipitation showed no differences in the (Ynaphthyl acetate esterase activity (data not shown). This indicates that the anti-JH esterase serum was specific to JH esterase and not to other esterases in the plasma. The anti-JH esterase serum of M. sextu showed no cross-reactivity with the JH esterase of T. ni and the Colorado potato beetle, Leptinotarsa decemlineata when examined by immunotitration (data not shown).  also reported that the anti-JH esterase serum of T. ni showed no cross-reactivity toward M. sextu and very little cross-reactivity toward JH esterase from other lepidopterans. This is not surprising considering the differences at the amino terminus ( Fig. 3) discussed earlier. Therefore, the function of the protein is conserved with respect to its action on JH, but the protein structure appears to be quite variable between species.
Multiple Forms of Plasma JH E&erase-The plasma JH esterase activity from the fifth instar day 3 larvae of black, white, and wild strains were compared for their sensitivity to inhibition by the affinity ligand, MBTFP, and the eluting agent, OTFP. The inhibitor concentrations (1O-3 to 10-l' M) were selected to give 0 to 95% inhibition after a lo-min incubation time. The residual enzyme activity was monitored using JH III at a final substrate concentration of 5 pM, and the percent inhibition was plotted against log inhibitor concentration. The inhibition curve for MBTFP for all the three strains was a simple sigmoid curve, and the Ibo values were determined to be 6.43 X 10m5 M, 2.25 X lo-" M, and 8.13 x 10m5 M for wild, black, and white strains, respectively (Fig. 6). The inhibition curves for OTFP in these same strains are shown in Fig. 7 (Fig. 6). The OTFP inhibition pattern, however, indicates the possibility that there are multiple forms of JH esterase in the plasma of M. sextu (Fig. 7). The percent inhibition for each strain was recalculated by assuming equivalent activity for both forms, such that the percent inhibition under 50% is expected to be from the less sensitive form and that above 50% is from the sensitive form. The corrected percentage inhibition thus calculated when plotted against log inhibitor concentration yielded straight lines resembling the inhibition of two homogeneous catalytic sites (Fig. 7). The regression analysis for these lines yielded Iho values in the range of 10m9 M and 1Om6 M for the sensitive and less sensitive forms, respectively, for all three strains, and is similar to that reported earlier by Abdel-Aal and  for a different wild-type strain of M. sextu. The similarity of inhibition curves obtained for all three strains suggests that the JH esterase activity is due to the same enzyme(s) and that the catalytic activity is not altered by strain differences.
As shown in this study, OTFP is a stronger inhibitor of plasma JH esterase activity than is MBTFP. Thus, OTFP can be used to displace the enzyme bound to MBTFP-Sepharose.
The inhibition curves obtained for OTFP provide evidence for the presence of multiple catalytic forms of JH esterase in the plasma of M. sextu. There are conflicting reports regarding the number of forms of JH esterase in the plasma of M. sertu. Sanburg et al. (1975), Abdel-Aal and Hammock (1986), and Sparks et al. (1989) report multiple isoelectric forms, but Coudron et al. (1981) and Wing et al. (1984) report only a single form. Recent studies by Jesudason et al. (1990) provide evidence for the existence of multiple isoforms of JH esterase throughout the entire life cycle of M. sextu. In addition, the presence of unique forms of JH esterase activity was identified by differences in the molecular weight of JH esterase activity in the preovipositional eggs of M. sextu (Share et al., 1988) and by isoelectric focusing in the corpora allata (Sparks et al., 1989). Broad range (pH 3.5-9.5) isoelectric focusing of plasma from wild, black, and white strains and purified enzyme resolved a broad peak of activity in the pH range of 4.5-7.0 (Fig. 8A). The major peak of activity in all three strains and the purified enzyme had the same isoelectric point of 6.0. However, when freshly collected plasma from day 3 fifth instar larvae was subjected to narrow range (pH 4.0-7.0) isoelectric focusing, two peaks of activity were resolved which were found roughly in the same proportion (Fig. 8B). The two isoelectric forms were resolved at pH values 6.0 and 5.5. Jesudason et al. (1990) also resolved the same multiple forms in the plasma of wild-type M. sextu throughout larval, pupal, and adult development; in the plasma of last instar day 3 wild, white, and black strains of M. sexta, and for purified JH esterase provided from this study. M. sextu JH esterase is glycosylated' as in T. ni (Hanzlik and Hammock, 1987), and JH esterase does not bind to ampholytes (Jones et al., 1986).
To determine whether the isoelectric forms have different chemical reactivities, their sensitivity to inhibition by OTFP and DFP was determined. DFP at lo-* M gave no inhibition of either of the JH esterase activities (data not shown). This confirms the hypothesis Roe and Venkatesh, 1990) that a specific plasma JH esterase is responsible for JH metabolism and is insensitive to inhibition by the general esterase inhibitor, DFP. The OTFP inhibition curves for both isoforms were identical (data not shown). Jesudason and Roe3 have also shown multiple catalytic function in individual isoelectric forms. This study has shown that although JH esterase is functionally similar between lepidopteran families, it appears to differ significantly immunologically and in the NHz-terminal amino acid sequence. The JH esterase activity in M. sextu is due to a group of highly similar proteins distinguishable only by high resolution isoelectric focusing and contains multiple catalytic sites.